2018-ADA-Standards-of-Care

S60 Glycemic Targets Diabetes Care Volume 41, Supplement 1, January 2018

Table 6.2—Summary of glycemic recommendations for many nonpregnant adults have been associated with increased car-
with diabetes diovascular risk independent of fasting
A1C <7.0% (53 mmol/mol)* plasma glucose in some epidemiological
studies, but intervention trials have not
Preprandial capillary plasma glucose 80–130 mg/dL* (4.4–7.2 mmol/L) shown postprandial glucose to be a cardio-
vascular risk factor independent of A1C. In
Peak postprandial capillary plasma glucose† <180 mg/dL* (10.0 mmol/L) subjects with diabetes, surrogate measures
of vascular pathology, such as endothelial
*More or less stringent glycemic goals may be appropriate for individual patients. Goals should be dysfunction, are negatively affected by post-
prandial hyperglycemia. It is clear that post-
individualized based on duration of diabetes, age/life expectancy, comorbid conditions, known CVD prandial hyperglycemia, like preprandial
hyperglycemia, contributes to elevated
or advanced microvascular complications, hypoglycemia unawareness, and individual patient A1C levels, with its relative contribution be-
considerations. †Postprandial glucose may be targeted if A1C goals are not met despite reaching ing greater at A1C levels that are closer to
preprandial glucose goals. Postprandial glucose measurements should be made 1–2 h after the 7% (53 mmol/mol). However, outcome
beginning of the meal, generally peak levels in patients with diabetes. studies have clearly shown A1C to be the
primary predictor of complications, and
of the ACCORD data did not identify a proposes optimal targets, but each target landmark trials of glycemic control such as
clear explanation for the excess mortality must be individualized to the needs of each the DCCT and UKPDS relied overwhelmingly
in the intensive treatment arm (61). patient and his or her disease factors. on preprandial SMBG. Additionally, a ran-
domized controlled trial in patients with
Longer-term follow-up has shown no ev- When possible, such decisions should be known CVD found no CVD benefit of insulin
idence of cardiovascular benefit or harm in made with the patient, reflecting his or her regimens targeting postprandial glucose
the ADVANCE trial (66). The end-stage re- preferences, needs, and values. Fig. 6.1 compared with those targeting preprandial
nal disease rate was lower in the intensive is not designed to be applied rigidly but glucose (74). Therefore, it is reasonable for
treatment group over follow-up. However, to be used as a broad construct to guide postprandial testing to be recommended
10-year follow-up of the VADT cohort clinical decision-making (72), in both type 1 for individuals who have premeal glucose
(67) showed a reduction in the risk of car- and type 2 diabetes. values within target but have A1C values
diovascular events (52.7 [control group] above target. Measuring postprandial
vs. 44.1 [intervention group] events per Recommended glycemic targets for plasma glucose 1–2 h after the start of a
1,000 person-years) with no benefit in many nonpregnant adults are shown in meal and using treatments aimed at
cardiovascular or overall mortality. Hetero- Table 6.2. The recommendations include
geneity of mortality effects across studies blood glucose levels that appear to corre-
was noted, which may reflect differences late with achievement of an A1C of ,7%
in glycemic targets, therapeutic approaches, (53 mmol/mol). The issue of preprandial
and population characteristics (68). versus postprandial SMBG targets is com-
plex (73). Elevated postchallenge (2-h oral
Mortality findings in ACCORD (61) and glucose tolerance test) glucose values
subgroup analyses of VADT (69) suggest
that the potential risks of intensive glyce-
mic control may outweigh its benefits in
higher-risk patients. In all three trials, se-
vere hypoglycemia was significantly more
likely in participants who were randomly
assigned to the intensive glycemic control
arm. Those patients with long duration of
diabetes, a known history of hypoglyce-
mia, advanced atherosclerosis, or ad-
vanced age/frailty may benefit from less
aggressive targets (70,71).

Providers should be vigilant in preventing
hypoglycemia and should not aggressively
attempt to achieve near-normal A1C levels
in patients in whom such targets cannot be
safely and reasonably achieved. Severe or
frequent hypoglycemia is an absolute indi-
cation for the modification of treatment
regimens, including setting higher glycemic
goals.

Many factors, including patient prefer-
ences, should be taken into account when
developing a patient’s individualized
goals (Table 6.2).

A1C and Glycemic Targets Figure 6.1—Depicted are patient and disease factors used to determine optimal A1C targets.
Numerous aspects must be considered Characteristics and predicaments toward the left justify more stringent efforts to lower A1C; those
when setting glycemic targets. The ADA toward the right suggest less stringent efforts. Adapted with permission from Inzucchi et al. (72).

care.diabetesjournals.org Glycemic Targets S61

Table 6.3—Classification of hypoglycemia*

Level Glycemic criteria Description

Hypoglycemia alert value (level 1) #70 mg/dL (3.9 mmol/L) Sufficiently low for treatment with fast-acting carbohydrate and dose

adjustment of glucose-lowering therapy

Clinically significant hypoglycemia (level 2) ,54 mg/dL (3.0 mmol/L) Sufficiently low to indicate serious, clinically important hypoglycemia

Severe hypoglycemia (level 3) No specific glucose threshold Hypoglycemia associated with severe cognitive impairment requiring

external assistance for recovery

*Adapted from ref. 75.

reducing postprandial plasma glucose val- c Hypoglycemia unawareness or one Hypoglycemia may be inconvenient or
ues to ,180 mg/dL (10.0 mmol/L) may or more episodes of severe hypo- frightening to patients with diabetes. Se-
help to lower A1C. glycemia should trigger reevalua- vere hypoglycemia may be recognized or
tion of the treatment regimen. E unrecognized and can progress to loss of
An analysis of data from 470 participants consciousness, seizure, coma, or death. It is
in the ADAG study (237 with type 1 diabe- c Insulin-treated patients with hy- reversed by administration of rapid-acting
tes and 147 with type 2 diabetes) found poglycemia unawareness or an glucose or glucagon. Clinically significant
that actual average glucose levels associ- episode of clinically significant hy- hypoglycemia can cause acute harm to
ated with conventional A1C targets were poglycemia should be advised to the person with diabetes or others, espe-
higher than older DCCT and ADA targets raise their glycemic targets to cially if it causes falls, motor vehicle acci-
(Table 6.1) (37,39). These findings support strictly avoid hypoglycemia for at dents, or other injury. A large cohort study
that premeal glucose targets may be relaxed least several weeks in order to par- suggested that among older adults with
without undermining overall glycemic con- tially reverse hypoglycemia un- type 2 diabetes, a history of severe hypo-
trol as measured by A1C. These data promp- awareness and reduce risk of future glycemia was associated with greater risk
ted the revision in the ADA-recommended episodes. A of dementia (77). Conversely, in a sub-
premeal glucose target to 80–130 mg/dL study of the ACCORD trial, cognitive
(4.4–7.2 mmol/L) but did not affect the c Ongoing assessment of cognitive impairment at baseline or decline in cog-
definition of hypoglycemia. function is suggested with increased nitive function during the trial was sig-
vigilance for hypoglycemia by the nificantly associated with subsequent
HYPOGLYCEMIA clinician, patient, and caregivers if episodes of severe hypoglycemia (78). Ev-
low cognition or declining cognition idence from DCCT/EDIC, which involved
Recommendations is found. B adolescents and younger adults with
type 1 diabetes, found no association be-
c Individuals at risk for hypoglycemia Hypoglycemia is the major limiting factor tween frequency of severe hypoglycemia
should be asked about symptom- in the glycemic management of type 1 and cognitive decline (79), as discussed in
atic and asymptomatic hypoglyce- and type 2 diabetes. Recommendations Section 12 “Children and Adolescents.”
mia at each encounter. C from the International Hypoglycemia Study
Group regarding the classification of hypo- Severe hypoglycemia was associated
c Glucose (15–20 g) is the preferred glycemia in clinical trials are outlined in Ta- with mortality in participants in both the
treatment for the conscious individ- ble 6.3 (75). Of note, this classification standard and the intensive glycemia arms
ual with blood glucose #70 mg/dL scheme considers a blood glucose ,54 of the ACCORD trial, but the relationships
[3.9 mmol/L]), although any form of mg/dL (3.0 mmol/L) detected by SMBG, between hypoglycemia, achieved A1C,
carbohydrate that contains glucose CGM (for at least 20 min), or laboratory and treatment intensity were not straight-
may be used. Fifteen minutes after measurement of plasma glucose as suffi- forward. An association of severe hypo-
treatment, if SMBG shows contin- ciently low to indicate clinically significant glycemia with mortality was also found
ued hypoglycemia, the treatment hypoglycemia that should be included in in the ADVANCE trial (80). An association
should be repeated. Once SMBG reports of clinical trials of glucose-lowering between self-reported severe hypoglyce-
returns to normal, the individual drugs for the treatment of diabetes (75). mia and 5-year mortality has also been
should consume a meal or snack to However, a hypoglycemia alert value reported in clinical practice (81).
prevent recurrence of hypoglycemia. E of #70 mg/dL (3.9 mmol/L) can be impor-
tant for therapeutic dose adjustment of Young children with type 1 diabetes and
c Glucagon should be prescribed for glucose-lowering drugs in clinical care and the elderly, including those with type 1 and
all individuals at increased risk of is often related to symptomatic hypogly- type 2 diabetes (77,82), are noted as par-
clinically significant hypoglycemia, cemia. Severe hypoglycemia is defined as ticularly vulnerable to clinically significant
defined as blood glucose ,54 mg/dL severe cognitive impairment requiring as- hypoglycemia because of their reduced
(3.0 mmol/L), so it is available should sistance from another person for recov- ability to recognize hypoglycemic symp-
it be needed. Caregivers, school ery (76). toms and effectively communicate their
personnel, or family members of needs. Individualized glucose targets, pa-
these individuals should know Symptoms of hypoglycemia include, tient education, dietary intervention (e.g.,
where it is and when and how to but are not limited to, shakiness, irritabil- bedtime snack to prevent overnight hypo-
administer it. Glucagon administra- ity, confusion, tachycardia, and hunger. glycemia when specifically needed to treat
tion is not limited to health care
professionals. E

S62 Glycemic Targets Diabetes Care Volume 41, Supplement 1, January 2018

low blood glucose), exercise management, with hypoglycemia-prone diabetes (fam- prone patients also require urine or
medication adjustment, glucose monitor- ily members, roommates, school person- blood ketone monitoring. If accompa-
ing, and routine clinical surveillance may nel, child care providers, correctional nied by ketosis, vomiting, or alteration in
improve patient outcomes (76). CGM with institution staff, or coworkers) should be the level of consciousness, marked hyper-
automated low glucose suspend has been instructed on the use of glucagon kits in- glycemia requires temporary adjustment of
shown to be effective in reducing hypogly- cluding where the kit is and when and the treatment regimen and immediate in-
cemia in type 1 diabetes (34). For patients how to administer glucagon. An individual teraction with the diabetes care team. The
with type 1 diabetes with severe hypogly- does not need to be a health care pro- patient treated with noninsulin therapies or
cemia and hypoglycemia unawareness fessional to safely administer glucagon. medical nutrition therapy alone may tem-
that persists despite medical treatment, Care should be taken to ensure that glu- porarily require insulin. Adequate fluid and
human islet transplantation may be an op- cagon kits are not expired. caloric intake must be ensured. Infection or
tion, but the approach remains experimen- dehydration is more likely to necessitate
tal (83,84). Hypoglycemia Prevention hospitalization of the person with diabetes
Hypoglycemia prevention is a critical com- than the person without diabetes.
In 2015, the ADA changed its prepran- ponent of diabetes management. SMBG
dial glycemic target from 70–130 mg/dL and, for some patients, CGM are essential A physician with expertise in diabetes
(3.9–7.2 mmol/L) to 80–130 mg/dL (4.4– tools to assess therapy and detect incipient management should treat the hospital-
7.2 mmol/L). This change reflects the results hypoglycemia. Patients should understand ized patient. For further information on
of the ADAG study, which demonstrated situations that increase their risk of hypo- the management of diabetic ketoacidosis
that higher glycemic targets corresponded glycemia, such as fasting for tests or pro- and the hyperglycemic nonketotic hyper-
to A1C goals (37). An additional goal of cedures, delayed meals, during or after osmolar state, please refer to the ADA con-
raising the lower range of the glycemic intense exercise, and during sleep. Hypo- sensus report “Hyperglycemic Crises in
target was to limit overtreatment and glycemia may increase the risk of harm to Adult Patients With Diabetes” (87).
provide a safety margin in patients titrat- self or others, such as with driving. Teach-
ing glucose-lowering drugs such as insulin ing people with diabetes to balance insulin References
to glycemic targets. use and carbohydrate intake and exercise 1. Aleppo G, Ruedy KJ, Riddlesworth TD, et al.;
are necessary, but these strategies are not REPLACE-BG Study Group. REPLACE-BG: a ran-
Hypoglycemia Treatment always sufficient for prevention. domized trial comparing continuous glucose mon-
Providers should continue to counsel itoring with and without routine blood glucose
patients to treat hypoglycemia with fast- In type 1 diabetes and severely insulin- monitoring in adults with well-controlled type 1
acting carbohydrates at the hypoglycemia deficient type 2 diabetes, hypoglycemia diabetes. Diabetes Care 2017;40:538–545
alert value of 70 mg/dL (3.9 mmol/L) or unawareness (or hypoglycemia-associated 2. The Diabetes Control and Complications Trial
less. Hypoglycemia treatment requires autonomic failure) can severely com- Research Group. The effect of intensive treatment
ingestion of glucose- or carbohydrate- promise stringent diabetes control and quality of diabetes on the development and progression of
containing foods. The acute glycemic re- of life. This syndrome is characterized by de- long-term complications in insulin-dependent dia-
sponse correlates better with the glucose ficient counterregulatory hormone release, betes mellitus. N Engl J Med 1993;329:977–986
content of food than with the carbohy- especially in older adults, and a diminished 3. Miller KM, Beck RW, Bergenstal RM, et al.; T1D
drate content of food. Pure glucose is autonomic response, which both are risk fac- Exchange Clinic Network. Evidence of a strong
the preferred treatment, but any form tors for, and caused by, hypoglycemia. A cor- association between frequency of self-monitoring
of carbohydrate that contains glucose ollary to this “vicious cycle” is that several of blood glucose and hemoglobin A1c levels in T1D
will raise blood glucose. Added fat may weeks of avoidance of hypoglycemia has Exchange clinic registry participants. Diabetes
retard and then prolong the acute glyce- been demonstrated to improve counterregu- Care 2013;36:2009–2014
mic response. In type 2 diabetes, ingested lation and hypoglycemia awareness in many 4. Polonsky WH, Fisher L, Schikman CH, et al.
protein may increase insulin response patients (86). Hence, patients with one or Structured self-monitoring of blood glucose signif-
without increasing plasma glucose con- more episodes of clinically significant hy- icantly reduces A1C levels in poorly controlled,
centrations (85). Therefore, carbohydrate poglycemia may benefit from at least noninsulin-treated type 2 diabetes: results from
sources high in protein should not be short-term relaxation of glycemic targets. the Structured Testing Program study. Diabetes
used to treat or prevent hypoglycemia. Care 2011;34:262–267
Ongoing insulin activity or insulin secreta- INTERCURRENT ILLNESS 5. Gellad WF, Zhao X, Thorpe CT, Mor MK, Good
gogues may lead to recurrent hypoglyce- CB, Fine MJ. Dual use of Department of Veterans
mia unless further food is ingested after For further information on management Affairs and Medicare benefits and use of test
recovery. Once the glucose returns to of patients with hyperglycemia in the hos- strips in veterans with type 2 diabetes mellitus.
normal, the individual should be coun- pital, please refer to Section 14 “Diabetes JAMA Intern Med 2015;175:26–34
seled to eat a meal or snack to prevent Care in the Hospital.” 6. Grant RW, Huang ES, Wexler DJ, et al. Patients
recurrent hypoglycemia. who self-monitor blood glucose and their unused
Stressful events (e.g., illness, trauma, sur- testing results. Am J Manag Care 2015;21:e119–
Glucagon gery, etc.) may worsen glycemic control and e129
precipitate diabetic ketoacidosis or nonke- 7. Endocrine Society. Choosing wisely [Internet],
The use of glucagon is indicated for the totic hyperosmolar state, life-threatening 2013. Available from http://www.choosingwisely
treatment of hypoglycemia in people un- conditions that require immediate medical .org/societies/endocrine-society/. Accessed 18 Au-
able or unwilling to consume carbohy- care to prevent complications and death. gust 2015
drates by mouth. Those in close contact Any condition leading to deterioration 8. Ziegler R, Heidtmann B, Hilgard D, Hofer S,
with, or having custodial care of, people in glycemic control necessitates more fre- Rosenbauer J, Holl R; DPV-Wiss-Initiative. Fre-
quent monitoring of blood glucose; ketosis- quency of SMBG correlates with HbA1c and acute
complications in children and adolescents with
type 1 diabetes. Pediatr Diabetes 2011;12:11–17
9. Rosenstock J, Davies M, Home PD, Larsen J,
Koenen C, Schernthaner G. A randomised, 52-week,

care.diabetesjournals.org Glycemic Targets S63

treat-to-target trial comparing insulin detemir with GOLD randomized clinical trial. JAMA 2017;317: 37. Wei N, Zheng H, Nathan DM. Empirically es-
insulin glargine when administered as add-on to 379–387 tablishing blood glucose targets to achieve HbA1c
glucose-lowering drugs in insulin-naive people 23. Beck RW, Riddlesworth T, Ruedy K, et al. Ef- goals. Diabetes Care 2014;37:1048–1051
with type 2 diabetes. Diabetologia 2008;51:408– fect of continuous glucose monitoring on glyce- 38. Herman WH, Ilag LL, Johnson SL, et al. A clin-
416 mic control in adults with type 1 diabetes using ical trial of continuous subcutaneous insulin infusion
10. Garber AJ. Treat-to-target trials: uses, inter- insulin injections: the DIAMOND randomized clin- versus multiple daily injections in older adults with
pretation and review of concepts. Diabetes Obes ical trial. JAMA 2017;317:371–378 type 2 diabetes. Diabetes Care 2005;28:1568–1573
Metab 2014;16:193–205 24. Deiss D, Bolinder J, Riveline J-P, et al. Im- 39. Albers JW, Herman WH, Pop-Busui R, et al.; Dia-
11. Elgart JF, Gonza´lez L, Prestes M, Rucci E, proved glycemic control in poorly controlled pa- betes Control and Complications Trial/Epidemiology
Gagliardino JJ. Frequency of self-monitoring blood tients with type 1 diabetes using real-time of Diabetes Interventions and Complications
glucose and attainment of HbA1c target values. continuous glucose monitoring. Diabetes Care Research Group. Effect of prior intensive insulin
Acta Diabetol 2016;53:57–62 2006;29:2730–2732 treatment during the Diabetes Control and Com-
12. Farmer A, Wade A, Goyder E, et al. Impact of 25. O’Connell MA, Donath S, O’Neal DN, et al. plications Trial (DCCT) on peripheral neuropathy
self monitoring of blood glucose in the manage- Glycaemic impact of patient-led use of sensor- in type 1 diabetes during the Epidemiology of Di-
ment of patients with non-insulin treated diabe- guided pump therapy in type 1 diabetes: a rando- abetes Interventions and Complications (EDIC)
tes: open parallel group randomised trial. BMJ mised controlled trial. Diabetologia 2009;52: Study. Diabetes Care 2010;33:1090–1096
2007;335:132 1250–1257 40. Stratton IM, Adler AI, Neil HAW, et al. Asso-
13. O’Kane MJ, Bunting B, Copeland M, Coates 26. Wong JC, Foster NC, Maahs DM, et al.; T1D ciation of glycaemia with macrovascular and mi-
VE; ESMON study group. Efficacy of self moni- Exchange Clinic Network. Real-time continuous crovascular complications of type 2 diabetes
toring of blood glucose in patients with newly glucose monitoring among participants in the (UKPDS 35): prospective observational study.
diagnosed type 2 diabetes (ESMON study): T1D Exchange clinic registry. Diabetes Care BMJ 2000;321:405–412
randomised controlled trial. BMJ 2008;336: 2014;37:2702–2709 41. Jovanovicˇ L, Savas H, Mehta M, Trujillo A,
1174–1177 27. Hommel E, Olsen B, Battelino T, et al.; Pettitt DJ. Frequent monitoring of A1C during
14. Simon J, Gray A, Clarke P, Wade A, Neil A, SWITCH Study Group. Impact of continuous glu- pregnancy as a treatment tool to guide therapy.
Farmer A; Diabetes Glycaemic Education and cose monitoring on quality of life, treatment sat- Diabetes Care 2011;34:53–54
Monitoring Trial Group. Cost effectiveness of isfaction, and use of medical care resources: 42. Beck RW, Connor CG, Mullen DM, Wesley
self monitoring of blood glucose in patients with analyses from the SWITCH study. Acta Diabetol DM, Bergenstal RM. The fallacy of average: how
non-insulin treated type 2 diabetes: economic 2014;51:845–851 using HbA1c alone to assess glycemic control can
evaluation of data from the DiGEM trial. BMJ 28. Battelino T, Phillip M, Bratina N, Nimri R, be misleading. Diabetes Care 2017;40:994–999
2008;336:1177–1180 Oskarsson P, Bolinder J. Effect of continuous glu- 43. Nathan DM, Kuenen J, Borg R, Zheng H,
15. Young LA, Buse JB, Weaver MA, et al.; Mon- cose monitoring on hypoglycemia in type 1 diabe- Schoenfeld D, Heine RJ; A1c-Derived Average Glu-
itor Trial Group. Glucose self-monitoring in non- tes. Diabetes Care 2011;34:795–800 cose Study Group. Translating the A1C assay into
insulin-treated patients with type 2 diabetes in 29. Beck RW, Hirsch IB, Laffel L, et al.; Juvenile estimated average glucose values. Diabetes Care
primary care settings: a randomized trial. JAMA Diabetes Research Foundation Continuous Glu- 2008;31:1473–1478
Intern Med 2017;177:920–929 cose Monitoring Study Group. The effect of con- 44. Selvin E. Are there clinical implications of ra-
16. Malanda UL, Welschen LMC, Riphagen II, tinuous glucose monitoring in well-controlled cial differences in HbA1c? A difference, to be a
Dekker JM, Nijpels G, Bot SDM. Self-monitoring type 1 diabetes. Diabetes Care 2009;32:1378– difference, must make a difference. Diabetes
of blood glucose in patients with type 2 diabetes 1383 Care 2016;39:1462–1467
mellitus who are not using insulin. Cochrane Da- 30. Bode B, Beck RW, Xing D, et al.; Juvenile Di- 45. Bergenstal RM, Gal RL, Connor CG, et al.; T1D
tabase Syst Rev 2012;1:CD005060 abetes Research Foundation Continuous Glucose Exchange Racial Differences Study Group. Racial
17. Willett LR. ACP Journal Club. Meta-analysis: Monitoring Study Group. Sustained benefit of differences in the relationship of glucose concen-
self-monitoring in non-insulin-treated type 2 di- continuous glucose monitoring on A1C, glucose trations and hemoglobin A1c levels. Ann Intern
abetes improved HbA1c by 0.25%. Ann Intern profiles, and hypoglycemia in adults with type 1 Med 2017;167:95–102
Med 2012;156:JC6–JC12 diabetes. Diabetes Care 2009;32:2047–2049 46. Lacy ME, Wellenius GA, Sumner AE, et al. As-
18. U.S. Food and Drug Administration. FDA ap- 31. Yeh H-C, Brown TT, Maruthur N, et al. Com- sociation of sickle cell trait with hemoglobin A1c in
proves first continuous glucose monitoring system parative effectiveness and safety of methods African Americans. JAMA 2017;317:507–515
for adults not requiring blood sample calibration of insulin delivery and glucose monitoring 47. Wheeler E, Leong A, Liu C-T, et al.; EPIC-CVD
[Internet]. Available from https://www.fda.gov/ for diabetes mellitus: a systematic review and Consortium; EPIC-InterAct Consortium; Lifelines
NewsEvents/Newsroom/PressAnnouncements/ meta-analysis. Ann Intern Med 2012;157:336– Cohort Study. Impact of common genetic deter-
ucm577890.htm. Accessed 2 October 2017 347 minants of hemoglobin A1c on type 2 diabetes
19. Bolinder J, Antuna R, Geelhoed-Duijvestijn P, 32. Choudhary P, Ramasamy S, Green L, et al. risk and diagnosis in ancestrally diverse popula-
Kro¨ger J, Weitgasser R. Novel glucose-sensing Real-time continuous glucose monitoring signifi- tions: a transethnic genome-wide meta-analysis.
technology and hypoglycaemia in type 1 diabetes: cantly reduces severe hypoglycemia in hypoglycemia- PLoS Med 2017;14:e1002383
a multicentre, non-masked, randomised con- unaware patients with type 1 diabetes. Diabetes 48. Wilson DM, Kollman; Diabetes Research in
trolled trial. Lancet 2016;388:2254–2263 Care 2013;36:4160–4162 Children Network (DirecNet) Study Group. Re-
20. U.S. Food and Drug Administration. FDA ex- 33. Choudhary P, Rickels MR, Senior PA, et al. lationship of A1C to glucose concentrations in
pands indication for continuous glucose monitor- Evidence-informed clinical practice recommenda- children with type 1 diabetes: assessments by
ing system, first to replace fingerstick testing for tions for treatment of type 1 diabetes compli- high-frequency glucose determinations by sen-
diabetes treatment decisions [Internet]. Avail- cated by problematic hypoglycemia. Diabetes sors. Diabetes Care 2008;31:381–385
able from https://www.fda.gov/newsevents/ Care 2015;38:1016–1029 49. Buse JB, Kaufman FR, Linder B, Hirst K, El
newsroom/pressannouncements/ucm534056 34. Bergenstal RM, Klonoff DC, Garg SK, et al.; Ghormli L, Willi S; HEALTHY Study Group. Diabetes
.htm. Accessed 14 September 2017 ASPIRE In-Home Study Group. Threshold-based screening with hemoglobin A1c versus fasting
21. Tamborlane WV, Beck RW, Bode BW, et al.; insulin-pump interruption for reduction of hypo- plasma glucose in a multiethnic middle-school co-
Juvenile Diabetes Research Foundation Continu- glycemia. N Engl J Med 2013;369:224–232 hort. Diabetes Care 2013;36:429–435
ous Glucose Monitoring Study Group. Continuous 35. Bergenstal RM, Garg S, Weinzimer SA, et al. 50. Kamps JL, Hempe JM, Chalew SA. Racial dis-
glucose monitoring and intensive treatment of Safety of a hybrid closed-loop insulin delivery sys- parity in A1C independent of mean blood glucose
type 1 diabetes. N Engl J Med 2008;359:1464– tem in patients with type 1 diabetes. JAMA 2016; in children with type 1 diabetes. Diabetes Care
1476 316:1407–1408 2010;33:1025–1027
22. Lind M, Polonsky W, Hirsch IB, et al. Contin- 36. Stewart ZA, Wilinska ME, Hartnell S, et al. 51. Diabetes Control and Complications Trial
uous glucose monitoring vs conventional therapy Closed-loop insulin delivery during pregnancy in (DCCT)/Epidemiology of Diabetes Interventions
for glycemic control in adults with type 1 diabetes women with type 1 diabetes. N Engl J Med 2016; and Complications (EDIC) Research Group. Effect
treated with multiple daily insulin injections: the 375:644–654 of intensive diabetes therapy on the progression

S64 Glycemic Targets Diabetes Care Volume 41, Supplement 1, January 2018

of diabetic retinopathy in patients with type 1 di- 63. Nathan DM, Zinman B, Cleary PA, et al.; 74. Raz I, Wilson PWF, Strojek K, et al. Effects of
abetes: 18 years of follow-up in the DCCT/EDIC. Diabetes Control and Complications Trial/ prandial versus fasting glycemia on cardiovascular
Diabetes 2015;64:631–642 Epidemiology of Diabetes Interventions and Com- outcomes in type 2 diabetes: the HEART2D trial.
52. Lachin JM, Genuth S, Cleary P, Davis MD, plications (DCCT/EDIC) Research Group. Modern- Diabetes Care 2009;32:381–386
Nathan DM; Diabetes Control and Complications day clinical course of type 1 diabetes mellitus after 75. International Hypoglycaemia Study Group.
Trial/Epidemiology of Diabetes Interventions and 30 years’ duration: the Diabetes Control and Com- Glucose concentrations of less than 3.0 mmol/L
Complications Research Group. Retinopathy and plications Trial/Epidemiology of Diabetes Interven- (54 mg/dL) should be reported in clinical trials:
nephropathy in patients with type 1 diabetes four tions and Complications and Pittsburgh Epidemiology a joint position statement of the American Diabe-
years after a trial of intensive therapy. N Engl J of Diabetes Complications experience (1983-2005). tes Association and the European Association for
Med 2000;342:381–389 Arch Intern Med 2009;169:1307–1316 the Study of Diabetes. Diabetes Care 2017;40:
53. Ohkubo Y, Kishikawa H, Araki E, et al. Inten- 64. Orchard TJ, Nathan DM, Zinman B, et al.; 155–157
sive insulin therapy prevents the progression of Writing Group for the DCCT/EDIC Research Group. 76. Seaquist ER, Anderson J, Childs B, et al. Hypo-
diabetic microvascular complications in Japanese Association between 7 years of intensive treat- glycemia and diabetes: a report of a workgroup of
patients with non-insulin-dependent diabetes ment of type 1 diabetes and long-term mortality. the American Diabetes Association and The Endo-
mellitus: a randomized prospective 6-year study. JAMA 2015;313:45–53 crine Society. Diabetes Care 2013;36:1384–1395
Diabetes Res Clin Pract 1995;28:103–117 65. Skyler JS, Bergenstal R, Bonow RO, et al.; 77. Whitmer RA, Karter AJ, Yaffe K, Quesenberry
54. UK Prospective Diabetes Study (UKPDS) American Diabetes Association; American College CP Jr, Selby JV. Hypoglycemic episodes and risk of
Group. Effect of intensive blood-glucose control of Cardiology Foundation; American Heart Asso- dementia in older patients with type 2 diabetes
with metformin on complications in overweight ciation. Intensive glycemic control and the pre- mellitus. JAMA 2009;301:1565–1572
patients with type 2 diabetes (UKPDS 34). Lancet vention of cardiovascular events: implications of 78. Punthakee Z, Miller ME, Launer LJ, et al.;
1998;352:854–865 the ACCORD, ADVANCE, and VA diabetes trials: ACCORD Group of Investigators; ACCORD-MIND
55. UK Prospective Diabetes Study (UKPDS) a position statement of the American Diabetes Investigators. Poor cognitive function and risk of
Group. Intensive blood-glucose control with sul- Association and a scientific statement of the severe hypoglycemia in type 2 diabetes: post hoc
phonylureas or insulin compared with conven- American College of Cardiology Foundation and epidemiologic analysis of the ACCORD trial. Dia-
tional treatment and risk of complications in the American Heart Association. Diabetes Care betes Care 2012;35:787–793
patients with type 2 diabetes (UKPDS 33). Lancet 2009;32:187–192 79. Jacobson AM, Musen G, Ryan CM, et al.;
1998;352:837–853 66. Zoungas S, Chalmers J, Neal B, et al.; Diabetes Control and Complications Trial/
56. Holman RR, Paul SK, Bethel MA, Matthews ADVANCE-ON Collaborative Group. Follow-up of Epidemiology of Diabetes Interventions and Com-
DR, Neil HAW. 10-year follow-up of intensive glu- blood-pressure lowering and glucose control in plications Study Research Group. Long-term ef-
cose control in type 2 diabetes. N Engl J Med 2008; type 2 diabetes. N Engl J Med 2014;371:1392–1406 fect of diabetes and its treatment on cognitive
359:1577–1589 67. Hayward RA, Reaven PD, Wiitala WL, et al.; function. N Engl J Med 2007;356:1842–1852
57. Adler AI, Stratton IM, Neil HAW, et al. Asso- VADT Investigators. Follow-up of glycemic control 80. Zoungas S, Patel A, Chalmers J, et al.; ADVANCE
ciation of systolic blood pressure with macrovas- and cardiovascular outcomes in type 2 diabetes. N Collaborative Group. Severe hypoglycemia and
cular and microvascular complications of type 2 Engl J Med 2015;372:2197–2206 risks of vascular events and death. N Engl J Med
diabetes (UKPDS 36): prospective observational 68. Turnbull FM, Abraira C, Anderson RJ, et al.; 2010;363:1410–1418
study. BMJ 2000;321:412–419 Control Group. Intensive glucose control and mac- 81. McCoy RG, Van Houten HK, Ziegenfuss JY,
58. Duckworth W, Abraira C, Moritz T, et al.; rovascular outcomes in type 2 diabetes [published Shah ND, Wermers RA, Smith SA. Increased mor-
VADT Investigators. Glucose control and vascular correction appears in Diabetologia 2009;52: tality of patients with diabetes reporting severe
complications in veterans with type 2 diabetes. N 2470]. Diabetologia 2009;52:2288–2298 hypoglycemia. Diabetes Care 2012;35:1897–
Engl J Med 2009;360:129–139 69. Duckworth WC, Abraira C, Moritz TE, et al.; 1901
59. Patel A, MacMahon S, Chalmers J, et al.; Investigators of the VADT. The duration of diabe- 82. DuBose SN, Weinstock RS, Beck RW, et al.
ADVANCE Collaborative Group. Intensive blood glu- tes affects the response to intensive glucose con- Hypoglycemia in older adults with type 1 diabe-
cose control and vascular outcomes in patients with trol in type 2 subjects: the VA Diabetes Trial. J tes. Diabetes Technol Ther 2016;18:765–771
type 2 diabetes. N Engl J Med 2008;358:2560–2572 Diabetes Complications 2011;25:355–361 83. Hering BJ, Clarke WR, Bridges ND, et al.; Clin-
60. Ismail-Beigi F, Craven T, Banerji MA, et al.; 70. Lipska KJ, Ross JS, Miao Y, Shah ND, Lee SJ, ical Islet Transplantation Consortium. Phase 3 trial
ACCORD trial group. Effect of intensive treatment Steinman MA. Potential overtreatment of diabe- of transplantation of human islets in type 1 dia-
of hyperglycaemia on microvascular outcomes in tes mellitus in older adults with tight glycemic betes complicated by severe hypoglycemia. Dia-
type 2 diabetes: an analysis of the ACCORD rand- control. JAMA Intern Med 2015;175:356–362 betes Care 2016;39:1230–1240
omised trial. Lancet 2010;376:419–430 71. Vijan S, Sussman JB, Yudkin JS, Hayward RA. 84. Harlan DM. Islet transplantation for hypogly-
61. Gerstein HC, Miller ME, Byington RP, et al.; Effect of patients’ risks and preferences on health cemia unawareness/severe hypoglycemia: caveat
Action to Control Cardiovascular Risk in Diabetes gains with plasma glucose level lowering in type 2 emptor. Diabetes Care 2016;39:1072–1074
Study Group. Effects of intensive glucose lowering diabetes mellitus. JAMA Intern Med 2014;174: 85. Layman DK, Clifton P, Gannon MC, Krauss
in type 2 diabetes. N Engl J Med 2008;358:2545– 1227–1234 RM, Nuttall FQ. Protein in optimal health: heart
2559 72. Inzucchi SE, Bergenstal RM, Buse JB, et al. disease and type 2 diabetes. Am J Clin Nutr 2008;
62. Nathan DM, Cleary PA, Backlund J-YC, et al.; Management of hyperglycemia in type 2 diabetes, 87:1571S–1575S
Diabetes Control and Complications Trial/ 2015: a patient-centered approach: update to a 86. Cryer PE. Diverse causes of hypoglycemia-
Epidemiology of Diabetes Interventions and Com- position statement of the American Diabetes Asso- associated autonomic failure in diabetes. N Engl
plications (DCCT/EDIC) Study Research Group. In- ciation and the European Association for the Study J Med 2004;350:2272–2279
tensive diabetes treatment and cardiovascular of Diabetes. Diabetes Care 2015;38:140–149 87. Kitabchi AE, Umpierrez GE, Miles JM, Fisher
disease in patients with type 1 diabetes. N Engl J 73. American Diabetes Association. Postprandial JN. Hyperglycemic crises in adult patients with
Med 2005;353:2643–2653 blood glucose. Diabetes Care 2001;24:775–778 diabetes. Diabetes Care 2009;32:1335–1343

Diabetes Care Volume 41, Supplement 1, January 2018 S65

7. Obesity Management for the American Diabetes Association
Treatment of Type 2 Diabetes:
Standards of Medical Care in
Diabetesd2018

Diabetes Care 2018;41(Suppl. 1):S65–S72 | https://doi.org/10.2337/dc18-S007

The American Diabetes Association (ADA) “Standards of Medical Care in Diabetes” 7. OBESITY MANAGEMENT FOR THE TREATMENT OF TYPE 2 DIABETES
includes ADA’s current clinical practice recommendations and is intended to provide
the components of diabetes care, general treatment goals and guidelines, and tools
to evaluate quality of care. Members of the ADA Professional Practice Committee, a
multidisciplinary expert committee, are responsible for updating the Standards of
Care annually, or more frequently as warranted. For a detailed description of ADA
standards, statements, and reports, as well as the evidence-grading system for ADA’s
clinical practice recommendations, please refer to the Standards of Care Introduction.
Readers who wish to comment on the Standards of Care are invited to do so at
professional.diabetes.org/SOC.

There is strong and consistent evidence that obesity management can delay the progression
from prediabetes to type 2 diabetes (1,2) and may be beneficial in the treatment of type 2
diabetes (3–8). In overweight and obese patients with type 2 diabetes, modest and
sustained weight loss has been shown to improve glycemic control and to reduce the
need for glucose-lowering medications (3–5). Small studies have demonstrated that in
obese patients with type 2 diabetes more extreme dietary energy restriction with very-
low-calorie diets can reduce A1C to ,6.5% (48 mmol/mol) and fasting glucose
to ,126 mg/dL (7.0 mmol/L) in the absence of pharmacologic therapy or ongoing
procedures (7,9,10). Weight loss–induced improvements in glycemia are most likely to
occur early in the natural history of type 2 diabetes when obesity-associated insulin
resistance has caused reversible b-cell dysfunction but insulin secretory capacity re-
mains relatively preserved (5,8,10,11).The goal of this section is to provide evidence-

based recommendations for dietary, pharmacologic, and surgical interventions for

obesity management as treatments for hyperglycemia in type 2 diabetes.

ASSESSMENT

Recommendation Suggested citation: American Diabetes Associa-
tion. 7. Obesity management for the treatment
c At each patient encounter, BMI should be calculated and documented in the of type 2 diabetes: Standards of Medical Care in
medical record. B Diabetesd2018. Diabetes Care 2018;41(Suppl. 1):
S65–S72
At each routine patient encounter, BMI should be calculated as weight divided by
height squared (kg/m2) (12). BMI should be classified to determine the presence of © 2017 by the American Diabetes Association.
overweight or obesity, discussed with the patient, and documented in the patient Readers may use this article as long as the work
record. In Asian Americans, the BMI cutoff points to define overweight and obesity is properly cited, the use is educational and not
are lower than in other populations (Table 7.1) (13,14). Providers should advise over- for profit, and the work is not altered. More infor-
weight and obese patients that, in general, higher BMIs increase the risk of mation is available at http://www.diabetesjournals
.org/content/license.

S66 Obesity Management for the Treatment of Type 2 Diabetes Diabetes Care Volume 41, Supplement 1, January 2018

Table 7.1—Treatment options for overweight and obesity in type 2 diabetes
BMI category (kg/m2)

25.0–26.9 27.0–29.9 30.0–34.9 35.0–39.9 $40
(or 23.0–26.9*) (or 27.5–32.4*) (or 32.5–37.4*) (or $37.5*)
Treatment
†
Diet, physical activity, and behavioral therapy † † † † †
† †
Pharmacotherapy †† †

Metabolic surgery †

*Cutoff points for Asian American individuals. †Treatment may be indicated for selected motivated patients.

cardiovascular disease and all-cause mor- carefully selected patients by functioning, and health-related quality
tality. Providers should assess each pa- trained practitioners in medical of life (17). A post hoc analysis of the
tient’s readiness to achieve weight loss care settings with close medical Look AHEAD study suggests that hetero-
and jointly determine weight loss goals monitoring. To maintain weight geneous treatment effects may have
and intervention strategies. Strategies in- loss, such programs must incorpo- been present. Participants who had mod-
clude diet, physical activity, behavioral rate long-term comprehensive erately or poorly controlled diabetes (A1C
therapy, pharmacologic therapy, and met- weight maintenance counseling. B 6.8% or higher) as well as both those with
abolic surgery (Table 7.1). The latter two well-controlled diabetes (A1C less than
strategies may be prescribed for carefully Among overweight or obese patients with 6.8%) and good self-reported health
selected patients as adjuncts to diet, type 2 diabetes and inadequate glycemic, were found to have significantly reduced
physical activity, and behavioral therapy. blood pressure, and lipid control and/or cardiovascular events with intensive life-
other obesity-related medical conditions, style intervention during follow-up (18).
DIET, PHYSICAL ACTIVITY, AND lifestyle changes that result in modest
BEHAVIORAL THERAPY and sustained weight loss produce clini- Lifestyle Interventions
cally meaningful reductions in blood glu- Weight loss can be attained with lifestyle
Recommendations cose, A1C, and triglycerides (3–5). Greater programs that achieve a 500–750 kcal/day
weight loss produces even greater bene- energy deficit or provide approximately
c Diet, physical activity, and behavior- fits, including reductions in blood pres- 1,200–1,500 kcal/day for women and
al therapy designed to achieve .5% sure, improvements in LDL and HDL 1,500–1,800 kcal/day for men, adjusted
weight loss should be prescribed for cholesterol, and reductions in the need for the individual’s baseline body weight.
overweight and obese patients with for medications to control blood glucose, Although benefits may be seen with as
type 2 diabetes ready to achieve blood pressure, and lipids (3–5). little as 5% weight loss (19), sustained
weight loss. A weight loss of $7% is optimal.
Look AHEAD Trial
c Such interventions should be high Although the Action for Health in Diabe- These diets may differ in the types of
intensity ($16 sessions in 6 months) tes (Look AHEAD) trial did not show that foods they restrict (such as high-fat or
and focus on diet, physical activity, an intensive lifestyle intervention reduced high-carbohydrate foods) but are effec-
and behavioral strategies to achieve cardiovascular events in overweight or tive if they create the necessary energy
a 500–750 kcal/day energy deficit. A obese adults with type 2 diabetes (15), it deficit (12,20–22). Use of meal replace-
did show the feasibility of achieving and ment plans prescribed by trained practi-
c Diets should be individualized, as maintaining long-term weight loss in pa- tioners, with close patient monitoring,
those that provide the same caloric tients with type 2 diabetes. In the Look can be beneficial. Within the intensive
restriction but differ in protein, carbo- AHEAD intensive lifestyle intervention lifestyle intervention group of the Look
hydrate, and fat content are equally group, mean weight loss was 4.7% at AHEAD trial, for example, use of a partial
effective in achieving weight loss. A 8 years (16). Approximately 50% of inten- meal replacement plan was associated
sive lifestyle intervention participants with improvements in diet quality (23).
c For patients who achieve short- lost $5%, and 27% lost $10% of their The diet choice should be based on the
term weight-loss goals, long-term initial body weight at 8 years (16). Partic- patient’s health status and preferences.
($1 year) comprehensive weight ipants randomly assigned to the intensive
maintenance programs should be lifestyle group achieved equivalent risk Intensive behavioral lifestyle interven-
prescribed. Such programs should factor control but required fewer glucose-, tions should include $16 sessions in
provide at least monthly contact blood pressure–, and lipid-lowering med- 6 months and focus on diet, physical ac-
and encourage ongoing monitoring ications than those randomly assigned to tivity, and behavioral strategies to achieve
of body weight (weekly or more fre- standard care. Secondary analyses of the an ;500–750 kcal/day energy deficit. In-
quently), continued consumption Look AHEAD trial and other large cardio- terventions should be provided by trained
of a reduced-calorie diet, and par- vascular outcome studies document interventionists in either individual or
ticipation in high levels of physical other benefits of weight loss in patients group sessions (19).
activity (200–300 min/week). A with type 2 diabetes, including improve-
ments in mobility, physical and sexual Overweight and obese patients with
c To achieve weight loss of .5%, type 2 diabetes who have lost weight
short-term (3-month) interventions during the 6-month intensive behavioral
that use very-low-calorie diets lifestyle intervention should be enrolled
(#800 kcal/day) and total meal re- in long-term ($1 year) comprehensive
placements may be prescribed for

care.diabetesjournals.org Obesity Management for the Treatment of Type 2 Diabetes S67

weight loss maintenance programs that promote weight loss or to be weight neu- adhere to low-calorie diets and to rein-
provide at least monthly contact with a tral. Agents associated with weight loss force lifestyle changes including physical
trained interventionist and focus on on- include metformin, a-glucosidase inhibi- activity. Providers should be knowledge-
going monitoring of body weight (weekly tors, sodium–glucose cotransporter 2 in- able about the product label and should
or more frequently), continued consump- hibitors, glucagon-like peptide 1 agonists, balance the potential benefits of success-
tion of a reduced-calorie diet, and partic- and amylin mimetics. Dipeptidyl peptidase ful weight loss against the potential risks
ipation in high levels of physical activity 4 inhibitors appear to be weight neutral. of the medication for each patient. These
(200–300 min/week [24]). Some com- Unlike these agents, insulin secretagogues, medications are contraindicated in women
mercial and proprietary weight loss pro- thiazolidinediones, and insulin have often who are or may become pregnant. Women
grams have shown promising weight loss been associated with weight gain (see in their reproductive years must be cautioned
results (25). Section 8. Pharmacologic Approaches to use a reliable method of contraception.
to Glycemic Treatment”).
When provided by trained practitioners Assessing Efficacy and Safety
in medical care settings with close medical A recent meta-analysis of 227 random- Efficacy and safety should be assessed at least
monitoring, short-term (3-month) inter- ized controlled trials of antihyperglycemic monthly for the first 3 months of treatment.
ventions that use very-low-calorie diets treatments in type 2 diabetes found that If a patient’s response is deemed insuffi-
(defined as #800 kcal/day) and total A1C changes were not associated with cient (weight loss ,5%) after 3 months or
meal replacements may achieve greater baseline BMI, indicating that obese pa- if there are any safety or tolerability is-
short-term weight loss (10–15%) than in- tients can benefit from the same types sues at any time, the medication should
tensive behavioral lifestyle interventions of treatments for diabetes as normal- be discontinued and alternative medica-
that typically achieve 5% weight loss. weight patients (28). tions or treatment approaches should be
However, weight regain following the ces- considered.
sation of very-low-calorie diets is greater Concomitant Medications
than following intensive behavioral life- Providers should carefully review the pa- In general, pharmacologic treatment of
style interventions unless a long-term tient’s concomitant medications and, obesity has been limited by low adherence,
comprehensive weight loss maintenance whenever possible, minimize or provide modest efficacy, adverse effects, and weight
program is provided (26,27). alternatives for medications that pro- regain after medication cessation (30).
mote weight gain. Medications associ-
PHARMACOTHERAPY ated with weight gain include atypical METABOLIC SURGERY
antipsychotics (e.g., clozapine, olanza-
Recommendations pine, risperidone, etc.) and antidepres- Recommendations
sants (e.g., tricyclic antidepressants,
c When choosing glucose-lowering selective serotonin reuptake inhibitors, c Metabolic surgery should be recom-
medications for overweight or obese and monoamine oxidase inhibitors), glu- mended as an option to treat type 2
patients with type 2 diabetes, con- cocorticoids, oral contraceptives that diabetes in appropriate surgical
sider their effect on weight. E contain progestins, anticonvulsants in- candidates with BMI $40 kg/m2
cluding gabapentin, and a number of an- (BMI $37.5 kg/m2 in Asian Ameri-
c Whenever possible, minimize the tihistamines and anticholinergics. cans), regardless of the level of gly-
medications for comorbid conditions cemic control or complexity of
that are associated with weight gain. E Approved Weight Loss Medications glucose-lowering regimens, and in
The U.S. Food and Drug Administration adults with BMI 35.0–39.9 kg/m2
c Weight loss medications may be ef- (FDA) has approved medications for (32.5–37.4 kg/m2 in Asian Ameri-
fective as adjuncts to diet, physical ac- both short-term and long-term weight cans) when hyperglycemia is inade-
tivity, and behavioral counseling for management. Phentermine is indicated quately controlled despite lifestyle
selected patients with type 2 diabetes as short-term (a few weeks) adjunct in and optimal medical therapy. A
and BMI $27 kg/m2. Potential ben- conjunction with lifestyle and behavioral
efits must be weighed against the weight loss interventions (29). Five c Metabolic surgery should be con-
potential risks of the medications. A weight loss medications (or combination sidered as an option for adults with
medications) are FDA-approved for long- type 2 diabetes and BMI 30.0–
c If a patient’s response to weight loss term use (more than a few weeks) by 34.9 kg/m2 (27.5–32.4 kg/m2 in
medications is ,5% weight loss af- patients with BMI $27 kg/m2 with one Asian Americans) if hyperglycemia
ter 3 months or if there are any or more obesity-associated comorbid is inadequately controlled despite
safety or tolerability issues at any conditions (e.g., type 2 diabetes, hyperten- optimal medical control by either
time, the medication should be dis- sion, and dyslipidemia) and by patients oral or injectable medications (in-
continued and alternative medica- with BMI $30 kg/m2 who are motivated cluding insulin). B
tions or treatment approaches to lose weight (30–34). Medications ap-
should be considered. A proved by the FDA for the treatment of c Metabolic surgery should be per-
obesity and their advantages and disad- formed in high-volume centers
Antihyperglycemic Therapy vantages are summarized in Table 7.2. with multidisciplinary teams that
When evaluating pharmacologic treat- The rationale for weight loss medications understand and are experienced in
ments for overweight or obese patients is to help patients to more consistently the management of diabetes and
with type 2 diabetes, providers should gastrointestinal surgery. C
first consider their choice of glucose-
lowering medications. Whenever possi- c Long-term lifestyle support and rou-
ble, medications should be chosen to tine monitoring of micronutrient

Table 7.2—Medications approved by the FDA for the treatment of obesity S68 Obesity Management for the Treatment of Type 2 Diabetes

Generic drug name National Average Drug 1-Year weight change status1–4 Adverse effects1,5–12
(proprietary name[s]), dosage,
strength, and form Usual adult dosing Average wholesale Acquisition Cost (per Average weight loss % Patients with $5%
frequency relative to placebo loss of baseline weight
price (per month)13 month)14 Common6 Serious6

Short-term treatment (a few weeks)

Phentermine (Lomaira) 37.5 mg q.d. or 8 mg t.i.d. $5-$76 (37.5 mg); $3-$60 (37.5 mg); N/A* N/A* Headache, elevated blood Dyspnea, angina pectoris,
$52 (8 mg) Unavailable (8 mg) 35–73%
pressure, elevated syncope, severe
38–48%
38–48% heart rate, insomnia, hypertension
45–70%
dry mouth,

constipation, anxiety,

palpitations

Long-term treatment (more than a few weeks)

Lipase inhibitor

Orlistat (Alli) 60 mg caps 60 mg or 120 mg t.i.d. $41–82 (60 mg); $42 (60 mg); 2.5 kg (60 mg); Abdominal pain/ Liver failure and oxalate
$703 (120 mg) $556 (120 mg) 3.4 kg (120 mg) discomfort, oily spotting/ nephropathy
or orlistat (Xenical) (during or up to 1 h stool, fecal urgency,
flatulence,
120 mg caps after a low-fat meal) malabsorption of fat
soluble vitamins (A, D, E,
K) and medications (e.g.,
cyclosporine, thyroid
hormone replacement,
or anticonvulsants),
potentiation of the
effects of warfarin

Selective serotonin (5-HT) 5-HT2C receptor agonist $289 $230 3.2 kg Hypoglycemia, headache, Serotonin syndrome or
Lorcaserin (Belviq) 10 mg 10 mg b.i.d. fatigue NMS-like reactions,
tabs suicidal ideation, heart
Hypoglycemia, headache, valve disorder (,2.4%),
Lorcaserin (Belviq XR) 20 mg q.d. $289 $232 3.2 kg fatigue bradycardia Diabetes Care Volume 41, Supplement 1, January 2018
20 mg extended-release Serotonin syndrome or
tabs NMS-like reactions,
suicidal ideation, heart
valve disorder (,2.4%),
bradycardia

Sympathomimetic amine anorectic/antiepileptic combination

Phentermine/topiramate Recommended dose: $239 (maximum dose $192 (maximum dose 6.7 kg (7.5 mg/46 mg); Paresthesia, xerostomia, Topiramate is teratogenic
using the highest 8.9 kg (15 mg/92 mg) constipation, headache and has been associated
ER (Qsymia) 3.75 mg/ 3.75 mg/23 mg q.d. using the highest strength)
with cleft lip/palate
23 mg caps, 7.5 mg/ for 14 days, then strength)

46 mg caps, 11.25 mg/ increase to 7.5 mg/

69 mg caps, 15 mg/ 46 mg q.d.

92 mg caps Maximum dose:

15 mg/92 mg q.d.

Continued on p. S69

Table 7.2—Continued Usual adult dosing Average wholesale National Average Drug 1-Year weight change status1–4 Adverse effects1,5–12 care.diabetesjournals.org
frequency price (per month)13
Generic drug name Acquisition Cost (per Average weight loss % Patients with $5% Common6 Serious6
(proprietary name[s]), dosage, month)14 relative to placebo loss of baseline weight
strength, and form

Opioid antagonist/aminoketone antidepressant combination

Naltrexone/bupropion Maximum dose: two $290 (maximum dose) $231 (maximum dose) 2.0–4.1 kg 36–57% Nausea, constipation, Depression, precipitation of
(32 mg/360 mg) 51–73% headache, vomiting mania, contraindicated in
(Contrave) 8 mg/90 mg tablets of Contrave patients with a seizure
5.8–5.9 kg disorder
tabs b.i.d. for a total daily

dosage of naltrexone

32 mg/bupropion

360 mg

Glucagon-like peptide 1 receptor agonist

Liraglutide (Saxenda) Maintenance dose: $1,385 $1,105 Hypoglycemia, nausea, Pancreatitis, thyroid C-cell
vomiting, diarrhea, tumors in rodents,
6 mg/mL prefilled pen 3 mg s.c. q.d. constipation, headache contraindicated in
patients with personal/
family history of MTC or
MEN2, acute renal
failure

All medications are contraindicated in women who are or may become pregnant. Women in their reproductive years must be cautioned to use a reliable method of contraception. Caps, capsules; ER, extended release; Obesity Management for the Treatment of Type 2 Diabetes S69
MEN2, multiple endocrine neoplasia type 2; MTC, medullary thyroid carcinoma; N/A, not applicable; NMS, neuroleptic malignant syndrome; s.c., subcutaneous; tabs, tablets. *Phentermine is FDA-approved as a short-
term adjunct (a few weeks) in a regimen of weight reduction based on exercise, behavioral modification, and caloric restriction.
1Physicians’ Desk Reference. PDR Network, LLC (electronic version). Truven Health Analytics, Greenwood Village, CO.
2Yanovski SZ, Yanovski JA. Long-term drug treatment for obesity: a systematic and clinical review. JAMA 2014;311:74–86 (30).
3Astrup A, Carraro R, Finer N, et al.; NN8022–1807 Investigators. Safety, tolerability and sustained weight loss over 2 years with the once-daily human GLP-1 analog, liraglutide. Int J Obes (Lond) 2012;36:843–854.
4Wadden TA, Hollander P, Klein S, et al.; NN8022–1923 Investigators. Weight maintenance and additional weight loss with liraglutide after low-calorie-diet-induced weight loss: the SCALE Maintenance randomized study.
Int J Obes (Lond) 2013;37:1443–1451.
5DrugPoints System (electronic version). Truven Health Analytics, Greenwood Village, CO.

6Selective common (defined as an incidence of .5%) and serious adverse effects are noted. Refer to the medication package inserts for full information about adverse effects, cautions, and contraindications.
7Data of common adverse effects for Xenical were derived from seven double-blind, placebo-controlled clinical trials in mixed-type study populations (i.e., patients with or without type 2 diabetes), but the percentage of
patients with type 2 diabetes was not reported. In clinical trials in obese patients with diabetes, hypoglycemia and abdominal distension were also observed.

8Data of common adverse effects for Belviq were derived from placebo-controlled clinical trials in patients with type 2 diabetes.

9Data of common adverse effects for Qsymia were derived from four clinical trials in mixed-type study populations (i.e., patients with or without type 2 diabetes); 13% had type 2 diabetes.

10Data of common adverse effects for Contrave were derived from five double-blind, placebo-controlled clinical trials in mixed-type study populations (i.e., patients with or without type 2 diabetes); 13% had type 2
diabetes.

11Data of common adverse effects for Saxenda were derived from clinical trials in mixed-type study populations (i.e., patients with or without type 2 diabetes). Percentage of patients with type 2 diabetes was not
reported.

12Phentermine. FDA prescribing information, side effects and uses [Internet], 2017. Available from https://www.drugs.com/pro/phentermine.html. Accessed 22 September 2017 (29).

13RED BOOK Online. Micromedex 2.0 (electronic version). Truven Health Analytics, Greenwood Village, CO. Accessed 18 July 2017.

14National Average Drug Acquisition Cost data available at: https://data.medicaid.gov/. Accessed 19 July 2017.

S70 Obesity Management for the Treatment of Type 2 Diabetes Diabetes Care Volume 41, Supplement 1, January 2018

and nutritional status must be pro- Please refer to “Metabolic Surgery in the mortality, complications, reoperations,
vided to patients after surgery, accord- Treatment Algorithm for Type 2 Diabe- and readmissions (71).
ing to guidelines for postoperative tes: A Joint Statement by International
management of metabolic surgery Diabetes Organizations” for a thorough Although metabolic surgery has been
by national and international profes- review (35). shown to improve the metabolic profiles
sional societies. C of morbidly obese patients with type 1
c People presenting for metabolic Randomized controlled trials with diabetes, establishing the role of meta-
surgery should receive a compre- postoperative follow up ranging from bolic surgery in such patients will require
hensive mental health assessment. 1 to 5 years have documented sustained larger and longer studies (72).
B Surgery should be postponed in diabetes remission in 30–63% of patients
patients with histories of alcohol or (35). Available data suggest an erosion of Retrospective analyses and modeling
substance abuse, significant depres- diabetes remission over time (51): 35– studies suggest that metabolic surgery
sion, suicidal ideation, or other mental 50% or more of patients who initially may be cost-effective or even cost-saving
health conditions until these condi- achieve remission of diabetes eventually for patients with type 2 diabetes, but the
tions have been fully addressed. E experience recurrence. However, the me- results are largely dependent on assump-
c People who undergo metabolic sur- dian disease-free period among such in- tions about the long-term effectiveness
gery should be evaluated to assess dividuals following Roux-en-Y gastric and safety of the procedures (73,74).
the need for ongoing mental health bypass (RYGB) is 8.3 years (52,53). With
services to help them adjust to or without diabetes relapse, the majority Adverse Effects
medical and psychosocial changes of patients who undergo surgery main- Metabolic surgery is costly and has associ-
after surgery. C tain substantial improvement of glycemic ated risks. Longer-term concerns include
control from baseline for at least 5 (54,55) dumping syndrome (nausea, colic, diarrhea),
Several gastrointestinal (GI) operations to 15 (38,39,53,56–58) years. vitamin and mineral deficiencies, anemia,
including partial gastrectomies and bari- osteoporosis, and, rarely (75), severe hypo-
atric procedures (35) promote dramatic Younger age, shorter duration of diabe- glycemia from insulin hypersecretion.
and durable improvement of type 2 diabe- tes (e.g., ,8 years) (59), nonuse of insulin, Long-term nutritional and micronutrient
tes. Given the magnitude and rapidity of and better glycemic control are consis- deficiencies and related complications oc-
the effect of GI surgery on hyperglycemia, tently associated with higher rates of di- cur with variable frequency depending on
and experimental evidence that rearrange- abetes remission and/or lower risk of the type of procedure and require lifelong
ments of GI anatomy similar to those in recidivism (38,57,59). Greater baseline vitamin/nutritional supplementation
some metabolic procedures directly affect visceral fat area may also help to predict (76,77). Postprandial hypoglycemia is
glucose homeostasis (36), GI interventions better postoperative outcomes, espe- most likely to occur with RYGB (77,78).
have been suggested as treatments for cially among Asian American patients The exact prevalence of symptomatic hy-
type 2 diabetes, and in that context are with type 2 diabetes, who typically poglycemia is unknown. In one study, it
termed “metabolic surgery.” have more visceral fat compared with affected 11% of 450 patients who had un-
Caucasians with diabetes of the same dergone RYGB or vertical sleeve gastrectomy
A substantial body of evidence has now BMI (60). (75). Patients who undergo metabolic sur-
accumulated, including data from numer- gery may be at increased risk for sub-
ous randomized controlled clinical trials, Beyond improving glycemia, metabolic stance use, including drug and alcohol
demonstrating that metabolic surgery surgery has been shown to confer addi- use and cigarette smoking (79).
achieves superior glycemic control and re- tional health benefits in randomized con-
duction of cardiovascular risk factors in trolled trials, including greater reductions People with diabetes presenting for
obese patients with type 2 diabetes com- in cardiovascular disease risk factors (35) metabolic surgery also have increased rates
pared with various lifestyle/medical inter- and enhancements in quality of life of depression and other major psychiatric
ventions (35). Improvements in micro- and (54,59,61). disorders (80). Candidates for metabolic
macrovascular complications of diabetes, surgery with histories of alcohol or sub-
cardiovascular disease, and cancer have The safety of metabolic surgery has im- stance abuse, significant depression, sui-
been observed only in nonrandomized proved significantly over the past two de- cidal ideation, or other mental health
observational studies (37–46). Cohort cades, with continued refinement of conditions should therefore first be as-
studies attempting to match surgical minimally invasive approaches (laparo- sessed by a mental health professional
and nonsurgical subjects suggest that scopic surgery), enhanced training and with expertise in obesity management prior
the procedure may reduce longer-term credentialing, and involvement of multi- to consideration for surgery (81). Individu-
mortality (38). disciplinary teams. Mortality rates with als with preoperative psychopathology
metabolic operations are typically 0.1– should be assessed regularly following
On the basis of this mounting evidence, 0.5%, similar to cholecystectomy or hys- metabolic surgery to optimize mental
several organizations and government terectomy (62–66). Morbidity has also health management and to ensure psy-
agencies have recommended expanding dramatically declined with laparoscopic chiatric symptoms do not interfere with
the indications for metabolic surgery to approaches. Major complications rates weight loss and lifestyle changes.
include patients with inadequately controlled are 2–6%, with minor complications in
type 2 diabetes and BMI as low as 30 kg/m2 up to 15% (62–70), comparing favorably References
(27.5 kg/m2 for Asian Americans) (47–50). with other commonly performed elective 1. Tuomilehto J. The emerging global epidemic
operations (66). Empirical data suggest of type 1 diabetes. Curr Diab Rep 2013;13:795–
that proficiency of the operating surgeon 804
is an important factor for determining

care.diabetesjournals.org Obesity Management for the Treatment of Type 2 Diabetes S71

2. Knowler WC, Barrett-Connor E, Fowler SE, et al.; 18. Baum A, Scarpa J, Bruzelius E, Tamler R, Basu of liraglutide in weight management. N Engl J Med
Diabetes Prevention Program Research Group. S, Faghmous J. Targeting weight loss interventions 2015;373:11–22
Reduction in the incidence of type 2 diabetes to reduce cardiovascular complications of type 2 33. Khera R, Murad MH, Chandar AK, et al. Asso-
with lifestyle intervention or metformin. N Engl J diabetes: a machine learning-based post-hoc ciation of pharmacological treatments for obesity
Med 2002;346:393–403 analysis of heterogeneous treatment effects in with weight loss and adverse events: a systematic
3. UK Prospective Diabetes Study 7. UK Prospec- the Look AHEAD trial. Lancet Diabetes Endocrinol review and meta-analysis. JAMA 2016;315:2424–
tive Diabetes Study 7: response of fasting plasma 2017;5:808–815 2434
glucose to diet therapy in newly presenting type II 19. Franz MJ, Boucher JL, Rutten-Ramos S, 34. O’Neil PM, Smith SR, Weissman NJ, et al. Ran-
diabetic patients, UKPDS Group. Metabolism VanWormer JJ. Lifestyle weight-loss intervention domized placebo-controlled clinical trial of lorca-
1990;39:905–912 outcomes in overweight and obese adults with serin for weight loss in type 2 diabetes mellitus:
4. Goldstein DJ. Beneficial health effects of mod- type 2 diabetes: a systematic review and meta- the BLOOM-DM study. Obesity (Silver Spring)
est weight loss. Int J Obes Relat Metab Disord analysis of randomized clinical trials. J Acad Nutr 2012;20:1426–1436
1992;16:397–415 Diet 2015;115:1447–1463 35. Rubino F, Nathan DM, Eckel RH, et al.; Dele-
5. Pastors JG, Warshaw H, Daly A, Franz M, 20. Sacks FM, Bray GA, Carey VJ, et al. Compari- gates of the 2nd Diabetes Surgery Summit. Met-
Kulkarni K. The evidence for the effectiveness of son of weight-loss diets with different composi- abolic surgery in the treatment algorithm for
medical nutrition therapy in diabetes manage- tions of fat, protein, and carbohydrates. N Engl J type 2 diabetes: a joint statement by international
ment. Diabetes Care 2002;25:608–613 Med 2009;360:859–873 diabetes organizations. Diabetes Care 2016;39:
6. Lim EL, Hollingsworth KG, Aribisala BS, Chen 21. de Souza RJ, Bray GA, Carey VJ, et al. Effects 861–877
MJ, Mathers JC, Taylor R. Reversal of type 2 di- of 4 weight-loss diets differing in fat, protein, and 36. Rubino F, Marescaux J. Effect of duodenal-
abetes: normalisation of beta cell function in as- carbohydrate on fat mass, lean mass, visceral ad- jejunal exclusion in a non-obese animal model
sociation with decreased pancreas and liver ipose tissue, and hepatic fat: results from the of type 2 diabetes: a new perspective for an old
triacylglycerol. Diabetologia 2011;54:2506–2514 POUNDS LOST trial. Am J Clin Nutr 2012;95:614– disease. Ann Surg 2004;239:1–11
7. Jackness C, Karmally W, Febres G, et al. Very 625 37. Sjo¨stro¨m L, Lindroos A-K, Peltonen M, et al.;
low-calorie diet mimics the early beneficial effect 22. Johnston BC, Kanters S, Bandayrel K, et al. Swedish Obese Subjects Study Scientific Group.
of Roux-en-Y gastric bypass on insulin sensitivity Comparison of weight loss among named diet Lifestyle, diabetes, and cardiovascular risk factors
and b-cell function in type 2 diabetic patients. programs in overweight and obese adults: a 10 years after bariatric surgery. N Engl J Med
Diabetes 2013;62:3027–3032 meta-analysis. JAMA 2014;312:923–933 2004;351:2683–2693
8. Rothberg AE, McEwen LN, Kraftson AT, Fowler 23. Raynor HA, Anderson AM, Miller GD, et al.; 38. Sjo¨stro¨m L, Peltonen M, Jacobson P, et al.
CE, Herman WH. Very-low-energy diet for type 2 Look AHEAD Research Group. Partial meal re- Association of bariatric surgery with long-term re-
diabetes: an underutilized therapy? J Diabetes placement plan and quality of the diet at 1 year: mission of type 2 diabetes and with microvascular
Complications 2014;28:506–510 Action for Health in Diabetes (Look AHEAD) trial. and macrovascular complications. JAMA 2014;
9. Day JW, Ottaway N, Patterson JT, et al. A new J Acad Nutr Diet 2015;115:731–742 311:2297–2304
glucagon and GLP-1 co-agonist eliminates obesity 24. Donnelly JE, Blair SN, Jakicic JM, Manore MM, 39. Adams TD, Davidson LE, Litwin SE, et al.
in rodents. Nat Chem Biol 2009;5:749–757 Rankin JW, Smith BK; American College of Sports Health benefits of gastric bypass surgery after
10. Steven S, Hollingsworth KG, Al-Mrabeh A, Medicine. American College of Sports Medicine 6 years. JAMA 2012;308:1122–1131
et al. Very low-calorie diet and 6 months of weight Position Stand. Appropriate physical activity inter- 40. Sjo¨stro¨m L, Narbro K, Sjo¨stro¨m CD, et al.;
stability in type 2 diabetes: pathophysiological vention strategies for weight loss and prevention Swedish Obese Subjects Study. Effects of bariatric
changes in responders and nonresponders. Diabe- of weight regain for adults. Med Sci Sports Exerc surgery on mortality in Swedish obese subjects.
tes Care 2016;39:808–815 2009;41:459–471 N Engl J Med 2007;357:741–752
11. Schauer PR, Mingrone G, Ikramuddin S, Wolfe 25. Gudzune KA, Doshi RS, Mehta AK, et al. Effi- 41. Sjo¨stro¨m L, Gummesson A, Sjo¨stro¨m CD,
B. Clinical outcomes of metabolic surgery: efficacy cacy of commercial weight-loss programs: an up- et al.; Swedish Obese Subjects Study. Effects of
of glycemic control, weight loss, and remission of dated systematic review. Ann Intern Med 2015; bariatric surgery on cancer incidence in obese pa-
diabetes. Diabetes Care 2016;39:902–911 162:501–512 tients in Sweden (Swedish Obese Subjects Study):
12. Jensen MD, Ryan DH, Apovian CM, et al.; 26. Tsai AG, Wadden TA. The evolution of very- a prospective, controlled intervention trial. Lancet
American College of Cardiology/American Heart low-calorie diets: an update and meta-analysis. Oncol 2009;10:653–662
Association Task Force on Practice Guidelines; Obesity (Silver Spring) 2006;14:1283–1293 42. Sjo¨stro¨m L, Peltonen M, Jacobson P, et al.
Obesity Society. 2013 AHA/ACC/TOS guideline 27. Johansson K, Neovius M, Hemmingsson E. Ef- Bariatric surgery and long-term cardiovascular
for the management of overweight and obesity fects of anti-obesity drugs, diet, and exercise on events. JAMA 2012;307:56–65
in adults: a report of the American College of Car- weight-loss maintenance after a very-low-calorie 43. Adams TD, Gress RE, Smith SC, et al. Long-
diology/American Heart Association Task Force on diet or low-calorie diet: a systematic review and term mortality after gastric bypass surgery. N Engl
Practice Guidelines and The Obesity Society. J Am meta-analysis of randomized controlled trials. Am J Med 2007;357:753–761
Coll Cardiol 2014;63(25 Pt B):2985–3023 J Clin Nutr 2014;99:14–23 44. Arterburn DE, Olsen MK, Smith VA, et al. As-
13. Expert Consultation WHO; WHO Expert Con- 28. Cai X, Yang W, Gao X, Zhou L, Han X, Ji L. sociation between bariatric surgery and long-
sultation. Appropriate body-mass index for Asian Baseline body mass index and the efficacy of hy- term survival. JAMA 2015;313:62–70
populations and its implications for policy and in- poglycemic treatment in type 2 diabetes: a meta- 45. Adams TD, Arterburn DE, Nathan DM, Eckel
tervention strategies. Lancet 2004;363:157–163 analysis. PLoS One 2016;11:e0166625 RH. Clinical outcomes of metabolic surgery: mi-
14. Araneta MR, Grandinetti A, Chang HK. Opti- 29. Phentermine. FDA prescribing information, crovascular and macrovascular complications. Di-
mum BMI cut points to screen Asian Americans side effects and uses [Internet], 2017. Available abetes Care 2016;39:912–923
for type 2 diabetes: the UCSD Filipino Health from https://www.drugs.com/pro/phentermine 46. Sheng B, Truong K, Spitler H, Zhang L, Tong X,
Study and the North Kohala Study (Abstract). Di- .html. Accessed 22 September 2017 Chen L. The long-term effects of bariatric surgery
abetes 2014;63(Suppl. 1):A20 30. Yanovski SZ, Yanovski JA. Long-term drug on type 2 diabetes remission, microvascular and
15. Wing RR, Bolin P, Brancati FL, et al.; Look treatment for obesity: a systematic and clinical macrovascular complications, and mortality: a
AHEAD Research Group. Cardiovascular effects review. JAMA 2014;311:74–86 systematic review and meta-analysis. Obes Surg.
of intensive lifestyle intervention in type 2 diabe- 31. Greenway FL, Fujioka K, Plodkowski RA, et al.; 11 August 2017 [Epub ahead of print]. DOI:
tes. N Engl J Med 2013;369:145–154 COR-I Study Group. Effect of naltrexone plus bu- 10.1007/s11695-017-2866-4
16. Look AHEAD Research Group. Eight-year propion on weight loss in overweight and obese 47. Rubino F, Kaplan LM, Schauer PR, Cummings
weight losses with an intensive lifestyle interven- adults (COR-I): a multicentre, randomised, double- DE; Diabetes Surgery Summit Delegates. The Di-
tion: the Look AHEAD study. Obesity (Silver blind, placebo-controlled, phase 3 trial. Lancet abetes Surgery Summit consensus conference:
Spring) 2014;22:5–13 2010;376:595–605 recommendations for the evaluation and use of
17. Wilding JPH. The importance of weight man- 32. Pi-Sunyer X, Astrup A, Fujioka K, et al.; SCALE gastrointestinal surgery to treat type 2 diabetes
agement in type 2 diabetes mellitus. Int J Clin Obesity and Prediabetes NN8022-1839 Study mellitus. Ann Surg 2010;251:399–405
Pract 2014;68:682–691 Group. A randomized, controlled trial of 3.0 mg

S72 Obesity Management for the Treatment of Type 2 Diabetes Diabetes Care Volume 41, Supplement 1, January 2018

48. Cummings DE, Cohen RV. Beyond BMI: the 60. Yu H, Di J, Bao Y, et al. Visceral fat area as a and complication rates after bariatric surgery.
need for new guidelines governing the use of new predictor of short-term diabetes remission N Engl J Med 2013;369:1434–1442
bariatric and metabolic surgery. Lancet Diabetes after Roux-en-Y gastric bypass surgery in Chinese 72. Kirwan JP, Aminian A, Kashyap SR, Burguera
Endocrinol 2014;2:175–181 patients with a body mass index less than 35 kg/m2. B, Brethauer SA, Schauer PR. Bariatric surgery in
49. Zimmet P, Alberti KGMM, Rubino F, Dixon JB. Surg Obes Relat Dis 2015;11:6–11 obese patients with type 1 diabetes. Diabetes
IDF’s view of bariatric surgery in type 2 diabetes. 61. Halperin F, Ding S-A, Simonson DC, et al. Care 2016;39:941–948
Lancet 2011;378:108–110 Roux-en-Y gastric bypass surgery or lifestyle with 73. Rubin JK, Hinrichs-Krapels S, Hesketh R,
50. Kasama K, Mui W, Lee WJ, et al. IFSO-APC intensive medical management in patients with Martin A, Herman WH, Rubino F. Identifying bar-
consensus statements 2011. Obes Surg 2012;22: type 2 diabetes: feasibility and 1-year results riers to appropriate use of metabolic/bariatric sur-
677–684 of a randomized clinical trial. JAMA Surg 2014; gery for type 2 diabetes treatment: policy lab
51. Ikramuddin S, Korner J, Lee W-J, et al. Dura- 149:716–726 results. Diabetes Care 2016;39:954–963
bility of addition of Roux-en-Y gastric bypass to 62. Flum DR, Belle SH, King WC, et al.; Longitudi- 74. Fouse T, Schauer P. The socioeconomic im-
lifestyle intervention and medical management in nal Assessment of Bariatric Surgery (LABS) Con- pact of morbid obesity and factors affecting
achieving primary treatment goals for uncon- sortium. Perioperative safety in the longitudinal access to obesity surgery. Surg Clin North Am
trolled type 2 diabetes in mild to moderate obe- assessment of bariatric surgery. N Engl J Med 2016;96:669–679
sity: a randomized control trial. Diabetes Care 2009;361:445–454 75. Service GJ, Thompson GB, Service FJ,
2016;39:1510–1518 63. Courcoulas AP, Christian NJ, Belle SH, et al.; Andrews JC, Collazo-Clavell ML, Lloyd RV. Hyper-
52. Sjo¨holm K, Pajunen P, Jacobson P, et al. In- Longitudinal Assessment of Bariatric Surgery insulinemic hypoglycemia with nesidioblastosis
cidence and remission of type 2 diabetes in re- (LABS) Consortium. Weight change and health after gastric-bypass surgery. N Engl J Med 2005;
lation to degree of obesity at baseline and 2 year outcomes at 3 years after bariatric surgery among 353:249–254
weight change: the Swedish Obese Subjects (SOS) individuals with severe obesity. JAMA 2013;310: 76. Mechanick JI, Kushner RF, Sugerman HJ,
study. Diabetologia 2015;58:1448–1453 2416–2425 et al.; American Association of Clinical Endocrinol-
53. Arterburn DE, Bogart A, Sherwood NE, et al. A 64. Arterburn DE, Courcoulas AP. Bariatric sur- ogists; Obesity Society; American Society for Met-
multisite study of long-term remission and re- gery for obesity and metabolic conditions in abolic & Bariatric Surgery. American Association
lapse of type 2 diabetes mellitus following gastric adults. BMJ 2014;349:g3961 of Clinical Endocrinologists, The Obesity Society,
bypass. Obes Surg 2013;23:93–102 65. Young MT, Gebhart A, Phelan MJ, Nguyen NT. and American Society for Metabolic & Bariatric
54. Mingrone G, Panunzi S, De Gaetano A, et al. Use and outcomes of laparoscopic sleeve gastrec- Surgery medical guidelines for clinical practice
Bariatric-metabolic surgery versus conventional tomy vs laparoscopic gastric bypass: analysis of for the perioperative nutritional, metabolic, and
medical treatment in obese patients with type 2 the American College of Surgeons NSQIP. J Am nonsurgical support of the bariatric surgery pa-
diabetes: 5 year follow-up of an open-label, single- Coll Surg 2015;220:880–885 tient. Obesity (Silver Spring) 2009;17(Suppl. 1):
centre, randomised controlled trial. Lancet 2015; 66. Aminian A, Brethauer SA, Kirwan JP, Kashyap S1–S70, v
386:964–973 SR, Burguera B, Schauer PR. How safe is metabolic/ 77. Mechanick JI, Youdim A, Jones DB, et al.;
55. Schauer PR, Bhatt DL, Kirwan JP, et al.; diabetes surgery? Diabetes Obes Metab 2015;17: American Association of Clinical Endocrinologists;
STAMPEDE Investigators. Bariatric surgery versus 198–201 Obesity Society; American Society for Metabolic &
intensive medical therapy for diabetes: 5-year 67. Birkmeyer NJO, Dimick JB, Share D, et al.; Bariatric Surgery. Clinical practice guidelines for
outcomes. N Engl J Med 2017;376:641–651 Michigan Bariatric Surgery Collaborative. Hospital the perioperative nutritional, metabolic, and non-
56. Cohen RV, Pinheiro JC, Schiavon CA, Salles JE, complication rates with bariatric surgery in Mich- surgical support of the bariatric surgery patientd
Wajchenberg BL, Cummings DE. Effects of gastric igan. JAMA 2010;304:435–442 2013 update: cosponsored by American Association
bypass surgery in patients with type 2 diabetes and 68. Altieri MS, Yang J, Telem DA, et al. Lap band of Clinical Endocrinologists, The Obesity Society,
only mild obesity. Diabetes Care 2012;35:1420– outcomes from 19,221 patients across centers and American Society for Metabolic & Bariatric
1428 and over a decade within the state of New York. Surgery. Obesity (Silver Spring) 2013;21(Suppl. 1):
57. Brethauer SA, Aminian A, Romero-Talama´s H, Surg Endosc 2016;30:1725–1732 S1–S27
et al. Can diabetes be surgically cured? Long-term 69. Hutter MM, Schirmer BD, Jones DB, et al. First 78. Lee CJ, Clark JM, Schweitzer M, et al. Preva-
metabolic effects of bariatric surgery in obese pa- report from the American College of Surgeons lence of and risk factors for hypoglycemic symp-
tients with type 2 diabetes mellitus [discussion Bariatric Surgery Center Network: laparoscopic toms after gastric bypass and sleeve gastrectomy.
appears in Ann Surg 2013;258:636–637]. Ann sleeve gastrectomy has morbidity and effective- Obesity (Silver Spring) 2015;23:1079–1084
Surg 2013;258:628–636 ness positioned between the band and the bypass 79. Conason A, Teixeira J, Hsu C-H, Puma L, Knafo
58. Hsu C-C, Almulaifi A, Chen J-C, et al. Effect of [discussion appears in Ann Surg 2011;254:420– D, Geliebter A. Substance use following bariatric
bariatric surgery vs medical treatment on type 2 422]. Ann Surg 2011;254:410–420 weight loss surgery. JAMA Surg 2013;148:145–
diabetes in patients with body mass index lower 70. Nguyen NT, Slone JA, Nguyen X-MT, Hartman 150
than 35: five-year outcomes. JAMA Surg 2015; JS, Hoyt DB. A prospective randomized trial of 80. Young-Hyman D, Peyrot M. Psychosocial Care
150:1117–1124 laparoscopic gastric bypass versus laparoscopic for People with Diabetes. 1st ed. Virginia, Ameri-
59. Schauer PR, Bhatt DL, Kirwan JP, et al.; adjustable gastric banding for the treatment of can Diabetes Association, 2012, p. 240
STAMPEDE Investigators. Bariatric surgery versus morbid obesity: outcomes, quality of life, and 81. Greenberg I, Sogg S, M Perna F. Behavioral
intensive medical therapy for diabetes: 3-year costs. Ann Surg 2009;250:631–641 and psychological care in weight loss surgery: best
outcomes. N Engl J Med 2014;370:2002–2013 71. Birkmeyer JD, Finks JF, O’Reilly A, et al.; Mich- practice update. Obesity (Silver Spring) 2009;17:
igan Bariatric Surgery Collaborative. Surgical skill 880–884

Diabetes Care Volume 41, Supplement 1, January 2018 S73

8. Pharmacologic Approaches to American Diabetes Association
Glycemic Treatment: Standards of
Medical Care in Diabetesd2018

Diabetes Care 2018;41(Suppl. 1):S73–S85 | https://doi.org/10.2337/dc18-S008

The American Diabetes Association (ADA) “Standards of Medical Care in Diabetes” 8. PHARMACOLOGIC APPROACHES TO GLYCEMIC TREATMENT
includes ADA’s current clinical practice recommendations and is intended to provide
the components of diabetes care, general treatment goals and guidelines, and tools
to evaluate quality of care. Members of the ADA Professional Practice Committee, a
multidisciplinary expert committee, are responsible for updating the Standards
of Care annually, or more frequently as warranted. For a detailed description of
ADA standards, statements, and reports, as well as the evidence-grading system
for ADA’s clinical practice recommendations, please refer to the Standards of Care
Introduction. Readers who wish to comment on the Standards of Care are invited to
do so at professional.diabetes.org/SOC.

PHARMACOLOGIC THERAPY FOR TYPE 1 DIABETES

Recommendations

c Most people with type 1 diabetes should be treated with multiple daily in-
jections of prandial insulin and basal insulin or continuous subcutaneous
insulin infusion. A

c Most individuals with type 1 diabetes should use rapid-acting insulin analogs to
reduce hypoglycemia risk. A

c Consider educating individuals with type 1 diabetes on matching prandial insulin
doses to carbohydrate intake, premeal blood glucose levels, and anticipated
physical activity. E

c Individuals with type 1 diabetes who have been successfully using continuous
subcutaneous insulin infusion should have continued access to this therapy after
they turn 65 years of age. E

Insulin Therapy Suggested citation: American Diabetes Associ-
Insulin is the mainstay of therapy for individuals with type 1 diabetes. Generally, ation. 8. Pharmacologic approaches to glyce-
mic treatment: Standards of Medical Care in
the starting insulin dose is based on weight, with doses ranging from 0.4 to Diabetesd2018. Diabetes Care 2018;41(Suppl. 1):
S73–S85
1.0 units/kg/day of total insulin with higher amounts required during puberty.
The American Diabetes Association/JDRF Type 1 Diabetes Sourcebook notes © 2017 by the American Diabetes Association.
0.5 units/kg/day as a typical starting dose in patients with type 1 diabetes who Readers may use this article as long as the work
is properly cited, the use is educational and not
are metabolically stable, with higher weight-based dosing required immediately for profit, and the work is not altered. More infor-
mation is available at http://www.diabetesjournals
following presentation with ketoacidosis (1), and provides detailed information .org/content/license.
on intensification of therapy to meet individualized needs. The American Diabetes
Association (ADA) position statement “Type 1 Diabetes Management Through
the Life Span” additionally provides a thorough overview of type 1 diabetes
treatment (2).

S74 Pharmacologic Approaches to Glycemic Treatment Diabetes Care Volume 41, Supplement 1, January 2018

Education regarding matching prandial compared with U-100 glargine in patients placebo (23). The Reducing With Metformin
insulin dosing to carbohydrate intake, with type 1 diabetes (19,20). Vascular Adverse Lesions in Type 1 Diabetes
premeal glucose levels, and anticipated (REMOVAL) trial investigated the addition
activity should be considered, and se- Rapid-acting inhaled insulin used be- of metformin therapy to titrated insulin
lected individuals who have mastered fore meals in patients with type 1 diabe- therapy in adults with type 1 diabetes at
carbohydrate counting should be edu- tes was shown to be noninferior when increased risk for cardiovascular disease
cated on fat and protein gram estimation compared with aspart insulin for A1C low- and found that metformin did not signifi-
(3–5). Although most studies of multiple ering, with less hypoglycemia observed cantly improve glycemic control beyond
daily injections versus continuous subcu- with inhaled insulin therapy (21). How- the first 3 months of treatment and that
taneous insulin infusion (CSII) have been ever, the mean reduction in A1C was progression of atherosclerosis (measured
small and of short duration, a systematic greater with aspart (–0.21% vs. –0.40%, by carotid artery intima-media thickness)
review and meta-analysis concluded that satisfying the noninferiority margin of was not significantly reduced, although
there are minimal differences between 0.4%), and more patients in the insulin other cardiovascular risk factors such as
the two forms of intensive insulin therapy aspart group achieved A1C goals of body weight and LDL cholesterol im-
in A1C (combined mean between-group #7.0% (53 mmol/mol) and #6.5% (48 proved (24). Metformin is not FDA-
difference favoring insulin pump therapy mmol/mol). Because inhaled insulin car- approved for use in patients with type 1
–0.30% [95% CI –0.58 to –0.02]) and se- tridges are only available in 4-, 8-, and diabetes.
vere hypoglycemia rates in children and 12-unit doses, limited dosing increments
adults (6). A 3-month randomized trial in to fine-tune prandial insulin doses in type 1 Incretin-Based Therapies
patients with type 1 diabetes with noctur- diabetes are a potential limitation.
nal hypoglycemia reported that sensor- Due to their potential protection of b-cell
augmented insulin pump therapy with Postprandial glucose excursions may mass and suppression of glucagon release,
the threshold suspend feature reduced be better controlled by adjusting the tim- glucagon-like peptide 1 (GLP-1) receptor
nocturnal hypoglycemia without increas- ing of prandial (bolus) insulin dose admin- agonists (25) and dipeptidyl peptidase
ing glycated hemoglobin levels (7). The istration. The optimal time to administer 4 (DPP-4) inhibitors (26) are being studied
U.S. Food and Drug Administration (FDA) prandial insulin varies, based on the type in patients with type 1 diabetes but are
has also approved the first hybrid closed- of insulin used (regular, rapid-acting ana- not currently FDA-approved for use in pa-
loop system pump. The safety and effi- log, inhaled, etc.), measured blood glucose tients with type 1 diabetes.
cacy of hybrid closed-loop systems has level, timing of meals, and carbohydrate
been supported in the literature in ado- consumption. Recommendations for pran- Sodium–Glucose Cotransporter 2 Inhibitors
lescents and adults with type 1 diabetes dial insulin dose administration should
(8,9). therefore be individualized. Sodium–glucose cotransporter 2 (SGLT2)
inhibitors provide insulin-independent
Intensive management using CSII and Pramlintide glucose lowering by blocking glucose re-
continuous glucose monitoring should be Pramlintide, an amylin analog, is an agent absorption in the proximal renal tubule by
encouraged in selected patients when that delays gastric emptying, blunts pan- inhibiting SGLT2. These agents provide
there is active patient/family participa- creatic secretion of glucagon, and en- modest weight loss and blood pressure
tion (10–12). hances satiety. It is FDA-approved for use reduction in type 2 diabetes. There are
in adults with type 1 diabetes. It has been three FDA-approved agents for patients
The Diabetes Control and Complica- shown to induce weight loss and lower in- with type 2 diabetes, but none are FDA-
tions Trial (DCCT) clearly showed that in- sulin doses. Concurrent reduction of pran- approved for the treatment of patients
tensive therapy with multiple daily dial insulin dosing is required to reduce the with type 1 diabetes (2). SGLT2 inhibitors
injections or CSII delivered by multidisci- risk of severe hypoglycemia. may have glycemic benefits in patients
plinary teams of physicians, nurses, dieti- with type 1 or type 2 diabetes on insulin
tians, and behavioral scientists improved Investigational Agents therapy (27). The FDA issued a warning
glycemia and resulted in better long-term about the risk of ketoacidosis occurring
outcomes (13–15). The study was carried Metformin in the absence of significant hyperglyce-
out with short-acting and intermediate- mia (euglycemic diabetic ketoacidosis)
acting human insulins. Despite better mi- Adding metformin to insulin therapy may in patients with type 1 or type 2 diabe-
crovascular, macrovascular, and all-cause reduce insulin requirements and improve tes treated with SGLT2 inhibitors.
mortality outcomes, intensive therapy metabolic control in patients with type 1 Symptoms of ketoacidosis include dysp-
was associated with a high rate of severe diabetes. In one study, metformin was nea, nausea, vomiting, and abdominal
hypoglycemia (61 episodes per 100 patient- found to reduce insulin requirements pain. Patients should be instructed to
years of therapy). Since the DCCT, a number (6.6 units/day, P , 0.001), and led to stop taking SGLT2 inhibitors and seek
of rapid-acting and long-acting insulin an- small reductions in weight and total and medical attention immediately if they
alogs have been developed. These analogs LDL cholesterol but not to improved gly- have symptoms or signs of ketoacidosis
are associated with less hypoglycemia, cemic control (absolute A1C reduction (28).
less weight gain, and lower A1C than human 0.11%, P 5 0.42) (22). A randomized clin-
insulins in people with type 1 diabetes ical trial similarly found that, among over- SURGICAL TREATMENT FOR
(16–18). Longer-acting basal analogs weight adolescents with type 1 diabetes, TYPE 1 DIABETES
(U-300 glargine or degludec) may addi- the addition of metformin to insulin did
tionally convey a lower hypoglycemia risk not improve glycemic control and in- Pancreas and Islet Transplantation
creased risk for gastrointestinal adverse Pancreas and islet transplantation have
events after 6 months compared with been shown to normalize glucose levels

care.diabetesjournals.org Pharmacologic Approaches to Glycemic Treatment S75

but require life-long immunosuppression therapy should begin with lifestyle and may reduce risk of cardiovascular
to prevent graft rejection and recurrence management and metformin and events and death (32). Compared with
of autoimmune islet destruction. Given subsequently incorporate an agent sulfonylureas, metformin as first-line
the potential adverse effects of immuno- proven to reduce major adverse car- therapy has beneficial effects on A1C,
suppressive therapy, pancreas transplan- diovascular events and cardiovascu- weight, and cardiovascular mortality
tation should be reserved for patients lar mortality (currently empagliflozin (33). Metformin may be safely used in
with type 1 diabetes undergoing simulta- and liraglutide), after considering patients with estimated glomerular filtra-
neous renal transplantation, following re- drug-specific and patient factors tion rate (eGFR) as low as 30 mL/min/
nal transplantation, or for those with (Table 8.1). A* 1.73 m2, and the FDA recently revised
recurrent ketoacidosis or severe hypogly- c In patients with type 2 diabetes and the label for metformin to reflect its
cemia despite intensive glycemic man- established atherosclerotic cardiovascu- safety in patients with eGFR $30 mL/
agement (29). lar disease, after lifestyle management min/1.73 m2 (34). Patients should be ad-
and metformin, the antihyperglycemic vised to stop the medication in cases of
PHARMACOLOGIC THERAPY FOR agent canagliflozin may be considered nausea, vomiting, or dehydration. Met-
TYPE 2 DIABETES to reduce major adverse cardiovascular formin is associated with vitamin B12
events, based on drug-specific and pa- deficiency, with a recent report from the
Recommendations tient factors (Table 8.1). C* Diabetes Prevention Program Outcomes
c Continuous reevaluation of the med- Study (DPPOS) suggesting that periodic
c Metformin, if not contraindicated ication regimen and adjustment as testing of vitamin B12 levels should be
and if tolerated, is the preferred ini- needed to incorporate patient fac- considered in metformin-treated pa-
tial pharmacologic agent for the tors (Table 8.1) and regimen com- tients, especially in those with anemia
treatment of type 2 diabetes. A plexity is recommended. E or peripheral neuropathy (35).
c For patients with type 2 diabetes
c Long-term use of metformin may be who are not achieving glycemic goals, In patients with metformin contrain-
associated with biochemical vitamin drug intensification, including consid- dications or intolerance, consider an ini-
B12 deficiency, and periodic mea- eration of insulin therapy, should not tial drug from another class depicted in
surement of vitamin B12 levels should be delayed. B Fig. 8.1 under “Dual Therapy” and pro-
be considered in metformin-treated c Metformin should be continued ceed accordingly. When A1C is $9% (75
patients, especially in those with ane- when used in combination with other mmol/mol), consider initiating dual com-
mia or peripheral neuropathy. B agents, including insulin, if not contra- bination therapy (Fig. 8.1) to more expe-
indicated and if tolerated. A ditiously achieve the target A1C level.
c Consider initiating insulin therapy Insulin has the advantage of being effec-
(with or without additional agents) See Section 12 for recommendations tive where other agents may not be and
in patients with newly diagnosed specific for children and adolescents should be considered as part of any com-
type 2 diabetes who are symptom- with type 2 diabetes. The use of metfor- bination regimen when hyperglycemia is
atic and/or have A1C $10% (86 min as first-line therapy was supported by severe, especially if catabolic features
mmol/mol) and/or blood glucose findings from a large meta-analysis, with (weight loss, ketosis) are present. Con-
levels $300 mg/dL (16.7 mmol/L). E selection of second-line therapies based sider initiating combination insulin in-
on patient-specific considerations (30). jectable therapy (Fig. 8.2) when blood
c Consider initiating dual therapy in An ADA/European Association for the Study glucose is $300 mg/dL (16.7 mmol/L) or
patients with newly diagnosed of Diabetes position statement “Manage- A1C is $10% (86 mmol/mol) or if the pa-
type 2 diabetes who have A1C ment of Hyperglycemia in Type 2 Diabe- tient has symptoms of hyperglycemia
$9% (75 mmol/mol). E tes, 2015: A Patient-Centered Approach” (i.e., polyuria or polydipsia). As the pa-
(31) recommended a patient-centered ap- tient’s glucose toxicity resolves, the regi-
c In patients without atherosclerotic proach, including assessment of efficacy, men may, potentially, be simplified.
cardiovascular disease, if mono- hypoglycemia risk, impact on weight, side
therapy or dual therapy does not effects, costs, and patient preferences. Re- Combination Therapy
achieve or maintain the A1C goal nal effects may also be considered when Although there are numerous trials
over 3 months, add an additional selecting glucose-lowering medications for comparing dual therapy with metformin
antihyperglycemic agent based on individual patients. Lifestyle modifications alone, few directly compare drugs as add-
drug-specific and patient factors that improve health (see Section 4 “Lifestyle on therapy. A comparative effectiveness
(Table 8.1). A Management”) should be emphasized meta-analysis (36) suggests that each
along with any pharmacologic therapy. new class of noninsulin agents added to
c A patient-centered approach should initial therapy generally lowers A1C ap-
be used to guide the choice of Initial Therapy proximately 0.7–1.0%. If the A1C target
pharmacologic agents. Consider- Metformin monotherapy should be is not achieved after approximately 3 months
ations include efficacy, hypoglyce- started at diagnosis of type 2 diabetes un- and patient does not have atherosclerotic
mia risk, history of atherosclerotic less there are contraindications. Metfor- cardiovascular disease (ASCVD), consider
cardiovascular disease, impact on min is effective and safe, is inexpensive, a combination of metformin and any one
weight, potential side effects, re- of the preferred six treatment options:
nal effects, delivery method (oral sulfonylurea, thiazolidinedione, DPP-4
versus subcutaneous), cost, and
patient preferences. E

c In patients with type 2 diabetes and
established atherosclerotic cardio-
vascular disease, antihyperglycemic

S76 Pharmacologic Approaches to Glycemic Treatment Diabetes Care Volume 41, Supplement 1, January 2018

Figure 8.1—Antihyperglycemic therapy in type 2 diabetes: general recommendations. *If patient does not tolerate or has contraindications to metformin,
consider agents from another class in Table 8.1. #GLP-1 receptor agonists and DPP-4 inhibitors should not be prescribed in combination. If a patient with
ASCVD is not yet on an agent with evidence of cardiovascular risk reduction, consider adding.

inhibitor, SGLT2 inhibitor, GLP-1 receptor second agent with evidence of cardiovas- dual therapy, proceed to a three-drug
agonist, or basal insulin (Fig. 8.1); the choice cular risk reduction after consideration of combination (Fig. 8.1). Again, if A1C target
of which agent to add is based on drug- drug-specific and patient factors (see p. S77 is not achieved after ;3 months of triple
specific effects and patient factors (Table CARDIOVASCULAR OUTCOMES TRIALS). If A1C target therapy, proceed to combination injectable
8.1). For patients with ASCVD, add a is still not achieved after ;3 months of therapy (Fig. 8.2). Drug choice is based on

Pharmacologic Approaches to Glycemic Treatment S77 Table 8.1—Drug-specific and patient factors to consider when selecting antihyperglycemic treatment in adults with type 2 diabetes

care.diabetesjournals.org *See ref. 31 for description of efficacy. †FDA approved for CVD benefit. CVD, cardiovascular disease; DKA, diabetic ketoacidosis; DKD, diabetic kidney disease; NASH, nonalcoholic steatohepatitis;
RAs, receptor agonists; SQ, subcutaneous; T2DM, type 2 diabetes.

S78 Pharmacologic Approaches to Glycemic Treatment Diabetes Care Volume 41, Supplement 1, January 2018

Figure 8.2—Combination injectable therapy for type 2 diabetes. FBG, fasting blood glucose; hypo, hypoglycemia. Adapted with permission from Inzucchi
et al. (31).

patient preferences (37), as well as various dual therapy, with continuous reevalu- Of note, prices listed are average whole-
patient, disease, and drug characteristics, ation of patient factors to guide treat- sale prices (AWP) (39) and National Aver-
with the goal of reducing blood glucose ment (Table 8.1). age Drug Acquisition Costs (NADAC) (40)
levels while minimizing side effects, espe- and do not account for discounts, re-
cially hypoglycemia. If not already in- Table 8.2 lists drugs commonly used in bates, or other price adjustments often
cluded in the treatment regimen, addition the U.S. Cost-effectiveness models of the involved in prescription sales that affect
of an agent with evidence of cardiovas- newer agents based on clinical utility and the actual cost incurred by the patient.
cular risk reduction should be consid- glycemic effect have been reported (38). While there are alternative means to esti-
ered in patients with ASCVD beyond Table 8.3 provides cost information for mate medication prices, AWP and NADAC
currently approved noninsulin therapies.

Table 8.2—Pharmacology of available glucose-lowering agents in the U.S. for the treatment of type 2 diabetes care.diabetesjournals.org

Class Compound(s) Cellular mechanism(s) Primary physiological action(s) Renal dosing recommendations (63–66)*

Biguanides c Metformin Activates AMP kinase (? other) ↓ Hepatic glucose production c No dose adjustment if eGFR .45;
do not initiate OR assess risk/benefit if currently on metformin if eGFR 30–45;
Sulfonylureas (2nd c Glyburide Closes KATP channels on b-cell ↑ Insulin secretion discontinue if eGFR ,30
generation) c Glipizide plasma membranes
c Glimepiride ↑ Insulin secretion c Avoid use in patients with renal impairment
Closes KATP channels on b-cell c Initiate conservatively at 2.5 mg daily to avoid hypoglycemia
Meglitinides c Repaglinide plasma membranes ↑ Insulin sensitivity c Initiate conservatively at 1 mg daily to avoid hypoglycemia
(glinides) c Nateglinide Activates the nuclear
transcription factor PPAR-g Slows intestinal carbohydrate c Initiate conservatively at 0.5 mg with meals if eGFR ,30
Thiazolidinediones c Pioglitazone Inhibits intestinal a-glucosidase digestion/absorption c Initiate conservatively at 60 mg with meals if eGFR ,30
c Rosiglitazone§ ↑ Insulin secretion (glucose
Inhibits DPP-4 activity, dependent); c No dose adjustment required
a-Glucosidase c Acarbose increasing postprandial incretin ↓ Glucagon secretion (glucose c No dose adjustment required
inhibitors c Miglitol (GLP-1, GIP) concentrations dependent)
c Avoid if eGFR ,30
DPP-4 inhibitors c Sitagliptin c Avoid if eGFR ,25

c Saxagliptin c 100 mg daily if eGFR .50;
c Linagliptin 50 mg daily if eGFR 30–50;
25 mg daily if eGFR ,30

c 5 mg daily if eGFR .50;
2.5 mg daily if eGFR #50

c No dose adjustment required

c Alogliptin c 25 mg daily if eGFR .60;
12.5 mg daily if eGFR 30–60;
Bile acid c Colesevelam Binds bile acids in intestinal ? ↓ Hepatic glucose production; 6.25 mg daily if eGFR ,30
sequestrants tract, increasing hepatic bile ? ↑ Incretin levels
acid production c No specific dose adjustment recommended by manufacturer
Dopamine-2 c Bromocriptine (quick Activates dopaminergic receptors Modulates hypothalamic regulation Pharmacologic Approaches to Glycemic Treatment S79
agonists release)§ of metabolism; c No specific dose adjustment recommended by manufacturer
Inhibits SGLT2 in the proximal ↑ Insulin sensitivity
SGLT2 inhibitors c Canagliflozin nephron Blocks glucose reabsorption by the c No dose adjustment required if eGFR $60;
kidney, increasing glucosuria 100 mg daily if eGFR 45–59;
avoid use and discontinue in patients with eGFR persistently ,45
c Dapagliflozin
c Avoid initiating if eGFR ,60;
GLP-1 receptor c Empagliflozin Activates GLP-1 ↑ Insulin secretion (glucose not recommended with eGFR 30–60;
agonists receptors dependent) contraindicated with eGFR ,30
c Exenatide
c Exenatide extended c Contraindicated with eGFR ,30
c Not recommended with eGFR ,30
release c Not recommended with eGFR ,30

Continued on p. S80

Table 8.2—Continued S80 Pharmacologic Approaches to Glycemic Treatment

Class Compound(s) Cellular mechanism(s) Primary physiological action(s) Renal dosing recommendations (63–66)*

c Liraglutide ↓ Glucagon secretion (glucose c No specific dose adjustment recommended by the manufacturer; limited
c Albiglutide dependent); experience in patients with severe renal impairment
c Lixisenatide Slows gastric emptying;
↑ Satiety c No dose adjustment required for eGFR 15–89 per manufacturer; limited
c Dulaglutide experience in patients with severe renal impairment

c No dose adjustment required for eGFR 60–89;
no dose adjustment required for eGFR 30–59, but patients should be
monitored for adverse effects and changes in kidney function;
clinical experience is limited with eGFR 15–29; patients should be monitored
for adverse effects and changes in kidney function;
avoid if eGFR ,15

c No specific dose adjustment recommended by the manufacturer; limited
experience in patients with severe renal impairment

Amylin mimetics c Pramlintide§ Activates amylin receptors ↓ Glucagon secretion; c No specific dose adjustment recommended by manufacturer
Slows gastric emptying;
↑ Satiety

Insulins c Rapid-acting analogs Activates insulin receptors ↑ Glucose disposal; c Lower insulin doses required with a decrease in eGFR; titrate per clinical
↓ Hepatic glucose production; response
Lispro Suppresses ketogenesis

Aspart

Glulisine

Inhaled insulin

c Short-acting analogs

Human Regular Diabetes Care Volume 41, Supplement 1, January 2018

c Intermediate-acting analogs

Human NPH

c Basal insulin analogs

Glargine

Detemir

Degludec

c Premixed insulin products

NPH/Regular 70/30

70/30 aspart mix

75/25 lispro mix

50/50 lispro mix

*eGFR is given in mL/min/1.73 m2. §Not licensed in Europe for type 2 diabetes. GIP, glucose-dependent insulinotropic peptide; PPAR-g, peroxisome proliferator–activated receptor g.

care.diabetesjournals.org Pharmacologic Approaches to Glycemic Treatment S81

Table 8.3—Median monthly cost of maximum approved daily dose of noninsulin glucose-lowering agents in the U.S.

Dosage strength/product Median AWP Median NADAC Maximum approved

Class Compound(s) (if applicable) (min, max)† (min, max)† daily dose*

Biguanides c Metformin 500 mg (IR) $84 ($4, $93) $2 2,000 mg
850 mg (IR) $108 ($6, $109) $3 2,550 mg
1,000 mg (IR) $87 ($4, $88) $2 2,000 mg
500 mg (ER) $89 ($82, $6,671) $5 ($5, $3,630) 2,000 mg
750 mg (ER) $72 ($65, $92) $5 1,500 mg
1,000 mg (ER) $1,028 ($1,028, $539 ($539, $5,189) 2,000 mg

$7,214) 20 mg
12 mg (micronized)
Sulfonylureas c Glyburide 5 mg $93 ($63, $103) $17 40 mg (IR)
(2nd generation) 6 mg (micronized) $50 ($48, $71) $12 20 mg (XL)
c Glipizide 10 mg (IR) $75 ($67, $97) $4 8 mg
Meglitinides (glinides) 10 mg (XL) $48 $16
Thiazolidinediones c Glimepiride 4 mg $71 ($71, $198) $7 16 mg
a-Glucosidase c Repaglinide 360 mg
c Nateglinide 2 mg $659 ($122, $673) $40
inhibitors c Pioglitazone 120 mg 45 mg
DPP-4 inhibitors c Rosiglitazone $155 $56 8 mg
c Acarbose 45 mg
Bile acid sequestrants c Miglitol 4 mg $348 ($283, $349) $5 300 mg
Dopamine-2 agonists c Sitagliptin 300 mg
SGLT2 inhibitors c Saxagliptin 100 mg $387 $314
c Linagliptin 100 mg 100 mg
GLP-1 receptor c Alogliptin $104 ($104, $106) $25 5 mg
agonists c Colesevelam 100 mg 5 mg
5 mg $241 N/A†† 25 mg
Amylin mimetics c Bromocriptine 5 mg
c Canagliflozin 25 mg $477 $382 3.75 g
c Dapagliflozin $462 $370 3.75 g
c Empagliflozin 625 mg tabs
c Exenatide 1.875 g suspension $457 $367 4.8 mg
c Lixisenatide $449 $357
c Liraglutide 0.8 mg 300 mg
c Exenatide (extended $713 $570 10 mg
300 mg $1,426 $572 25 mg
release) 10 mg 20 mg
c Albiglutide 25 mg $784 $629 20 mg
c Dulaglutide 1.8 mg
c Pramlintide 10 mg pen $512 $411 2 mg**
20 mg pen $517 $413
18 mg/3 mL pen $517 $415
2 mg powder for
$802 $642
suspension or pen $669 N/A††
50 mg pen $968 $775
1.5/0.5 mL pen $747 $600

120 mg pen $626 $500 50 mg**
$811 $648 1.5 mg**
N/A†† 120 mg/injection†††
$2,336

ER and XL, extended release; IR, immediate release. †Calculated for 30-day supply (AWP or NADAC unit price 3 number of doses required to provide
maximum approved daily dose 3 30 days); median AWP or NADAC listed alone when only one product and/or price. *Utilized to calculate median AWP
and NADAC (min, max); generic prices used, if available commercially. ††Not applicable; data not available. **Administered once weekly. †††AWP
and NADAC calculated based on 120 mg three times daily.

were utilized to provide two separate mea- late postprandial hypoglycemia when The empagliflozin and liraglutide trials
sures to allow for a comparison of drug taking a sulfonylurea. Other drugs not demonstrated significant reductions in
prices with the primary goal of highlighting shown in Table 8.1 (e.g., inhaled insulin, cardiovascular death. Exenatide once-
the importance of cost considerations a-glucosidase inhibitors, colesevelam, bro-
when prescribing antihyperglycemic treat- mocriptine, and pramlintide) may be tried weekly did not have statistically sig-
ments. The ongoing Glycemia Reduction in specific situations but considerations
Approaches in Diabetes: A Comparative Ef- include modest efficacy in type 2 diabetes, nificant reductions in major adverse
fectiveness Study (GRADE) will compare frequency of administration, potential for
four drug classes (sulfonylurea, DPP-4 in- drug interactions, cost, and/or side effects. cardiovascular events or cardiovascu-
hibitor, GLP-1 receptor agonist, and basal lar mortality but did have a significant
insulin) when added to metformin therapy Cardiovascular Outcomes Trials reduction in all-cause mortality. In con-
over 4 years on glycemic control and other There are now three large randomized
medical, psychosocial, and health economic controlled trials reporting statistically sig- trast, other GLP-1 receptor agonists
outcomes (41). nificant reductions in cardiovascular events
for two SGLT2 inhibitors (empagliflozin have not shown similar reductions in
Rapid-acting secretagogues (meglitinides) and canagliflozin) and one GLP-1 receptor cardiovascular events (Table 9.4).
may be used instead of sulfonylureas in agonist (liraglutide) where the majority, if Whether the benefits of GLP-1 receptor
patients with sulfa allergies or irregular not all patients, in the trial had ASCVD. agonists are a class effect remains to be
meal schedules or in those who develop definitively established. See ANTIHYPERGLYCEMIC
THERAPIES AND CARDIOVASCULAR OUTCOMES in
Section 9 “Cardiovascular Disease and
Risk Management” and Table 9.4 for a de-
tailed description of these cardiovascular

S82 Pharmacologic Approaches to Glycemic Treatment Diabetes Care Volume 41, Supplement 1, January 2018

Table 8.4—Median cost of insulin products in the U.S. calculated as AWP (39) and NADAC (40) per 1,000 units of specified dosage
form/product

Insulins Compounds Dosage form/product Median AWP Median NADAC
(min, max)* (min, max)*

Rapid-acting c Lispro U-100 vial; $330 $264
analogs c Aspart U-100 3 mL cartridges; $408 $326
U-100 prefilled pen; U-200 prefilled pen $424 $339
c Glulisine U-100 vial; $331 $265
U-100 3 mL cartridges; $410 $330
Short-acting analogs c Inhaled insulin U-100 prefilled pen $426 $341
Intermediate-acting analogs c Human Regular U-100 vial; $306 $245
c Human NPH U-100 prefilled pen $394 $315
Inhalation cartridges $725 ($544, $911) N/A†
Concentrated Human c U-500 Human $135 ($135, $145)
Regular insulin Regular insulin U-100 vial $165 ($165, $178)
$135 ($135, $145)
Basal analogs c Glargine U-100 vial; $165 ($165, $178) $305
U-100 prefilled pen $377
Premixed insulin products c Glargine biosimilar $143
c Detemir U-500 vial; $178 $184
c Degludec U-500 prefilled pen $230
c NPH/Regular 70/30 $239 ($239, $241)
U-100 vial; U-100 prefilled pen; $298
c Lispro 50/50 U-300 prefilled pen $203
$253 $259
c Lispro 75/25 U-100 prefilled pen $323 $285
U-100 vial; U-100 prefilled pen $355
c Aspart 70/30 U-100 prefilled pen; U-200 prefilled pen $134 ($134, $146)
$165 ($165, $178) $305
Premixed insulin/GLP-1 c Degludec/Liraglutide U-100 vial; $377 $278
U-100 prefilled pen $342 $339
receptor agonist products c Glargine/Lixisenatide U-100 vial; $424 $273
U-100 prefilled pen $342 $340
U-100 vial; $424 $275
U-100 prefilled pen $343 $341
U-100 vial; $426 N/A†
U-100 prefilled pen $404
$763
100/3.6 prefilled pen $508
100/33 prefilled pen

*AWP or NADAC calculated as in Table 8.3; median listed alone when only one product and/or price. †Not applicable; data not available.

outcomes trials. Additional large random- avoid using insulin as a threat or de- to reduce the risk of symptomatic and noc-
ized trials of other agents in these classes scribing it as a sign of personal failure turnal hypoglycemia (43–48). Longer-
are ongoing. or punishment. acting basal analogs (U-300 glargine or
degludec) may additionally convey a
Of note, these studies examined the Equipping patients with an algorithm for lower hypoglycemia risk compared with
drugs in combination with metformin self-titration of insulin doses based on self- U-100 glargine when used in combination
(Table 9.4) in the great majority of pa- monitoring of blood glucose improves with oral antihyperglycemic agents (49–
tients for whom metformin was not con- glycemic control in patients with type 2 di- 55). While there is evidence for reduced
traindicated or not tolerated. For patients abetes initiating insulin (42). Comprehen- hypoglycemia with newer, longer-acting
with type 2 diabetes who have ASCVD, on sive education regarding self-monitoring basal insulin analogs, people without a
lifestyle and metformin therapy, it is rec- of blood glucose, diet, and the avoidance history of hypoglycemia are at decreased
ommended to incorporate an agent with of and appropriate treatment of hypogly- risk and could potentially be switched to
strong evidence for cardiovascular risk re- cemia are critically important in any pa- human insulin safely. Thus, due to high
duction especially those with proven ben- tient using insulin. costs of analog insulins, use of human in-
efit on both major adverse cardiovascular sulin may be a practical option for some
events and cardiovascular death after con- Basal Insulin patients, and clinicians should be familiar
sideration of drug-specific patient factors with its use (56). Table 8.4 provides AWP
(Table 8.1). See Fig. 8.1 for additional rec- Basal insulin alone is the most convenient (39) and NADAC (40) information (cost
ommendations on antihyperglycemic initial insulin regimen, beginning at 10 units per 1,000 units) for currently available in-
treatment in adults with type 2 diabetes. per day or 0.1–0.2 units/kg/day, depend- sulin and insulin combination products
ing on the degree of hyperglycemia. Basal in the U.S. There have been substantial
Insulin Therapy insulin is usually prescribed in conjunc- increases in the price of insulin over the
Many patients with type 2 diabetes even- tion with metformin and sometimes one past decade and the cost-effectiveness
tually require and benefit from insulin additional noninsulin agent. When basal of different antihyperglycemic agents is
therapy. The progressive nature of type 2 insulin is added to antihyperglycemic an important consideration in a patient-
diabetes should be regularly and objectively agents in patients with type 2 diabetes, centered approach to care, along with
explained to patients. Providers should long-acting basal analogs (U-100 glargine
or detemir) can be used instead of NPH

care.diabetesjournals.org Pharmacologic Approaches to Glycemic Treatment S83

efficacy, hypoglycemia risk, weight, and pulmonary disease and is not recommended insulin (NPH/Regular 70/30, 70/30 aspart
other patient and drug-specific factors in patients who smoke or who recently stop- mix, 75/25 or 50/50 lispro mix) twice
(Table 8.1) (57). ped smoking. It requires spirometry (FEV1) daily, usually before breakfast and before
testing to identify potential lung disease in all dinner. Each approach has its advan-
Bolus Insulin patients prior to and after starting therapy. tages and disadvantages. For example,
providers may wish to consider regimen
Many individuals with type 2 diabetes Combination Injectable Therapy flexibility when devising a plan for the ini-
may require mealtime bolus insulin dos- If basal insulin has been titrated to an ac- tiation and adjustment of insulin therapy
ing in addition to basal insulin. Rapid- ceptable fasting blood glucose level (or if in people with type 2 diabetes, with rapid-
acting analogs are preferred due to their the dose is .0.5 units/kg/day) and A1C re- acting insulin offering greater flexibility in
prompt onset of action after dosing. In mains above target, consider advancing terms of meal planning than premixed in-
September 2017, the FDA approved a new to combination injectable therapy (Fig. sulin. If one regimen is not effective (i.e.,
faster-acting formulation of insulin aspart. 8.2). When initiating combination inject- basal insulin plus GLP-1 receptor agonist),
The recommended starting dose of meal- able therapy, metformin therapy should consider switching to another regimen to
time insulin is 4 units, 0.1 units/kg, or 10% be maintained while other oral agents achieve A1C targets (i.e., basal insulin plus
of the basal dose. If A1C is ,8% (64 mmol/ may be discontinued on an individual ba- single injection of rapid-acting insulin or pre-
mol) when starting mealtime bolus in- sis to avoid unnecessarily complex or mixed insulin twice daily) (60,61). Regular
sulin, consideration should be given to costly regimens (i.e., adding a fourth anti- human insulin and human NPH/Regular
decreasing the basal insulin dose. hyperglycemic agent). In general, GLP-1 premixed formulations (70/30) are less
receptor agonists should not be discon- costly alternatives to rapid-acting insulin
Premixed Insulin tinued with the initiation of basal insulin. analogs and premixed insulin analogs,
Sulfonylureas, DPP-4 inhibitors, and GLP- respectively, but their pharmacody-
Premixed insulin products contain both a 1 receptor agonists are typically stopped namic profiles may make them less optimal.
basal and prandial component, allowing once more complex insulin regimens be-
coverage of both basal and prandial needs yond basal are used. In patients with sub- Fig. 8.2 outlines these options, as well
with a single injection. NPH/Regular 70/30 optimal blood glucose control, especially as recommendations for further intensifi-
insulin, for example, is composed of 70% those requiring large insulin doses, adjunc- cation, if needed, to achieve glycemic
NPH insulin and 30% regular insulin. The use tive use of a thiazolidinedione or SGLT2 goals. If a patient is still above the A1C
of premixed insulin products has its advan- inhibitor may help to improve control target on premixed insulin twice daily,
tages and disadvantages, as discussed be- and reduce the amount of insulin needed, consider switching to premixed analog in-
low in COMBINATION INJECTABLE THERAPY. though potential side effects should be sulin three times daily (70/30 aspart mix,
considered. Once an insulin regimen is ini- 75/25 or 50/50 lispro mix). In general,
Concentrated Insulin Products tiated, dose titration is important with ad- three times daily premixed analog insu-
justments made in both mealtime and lins have been found to be noninferior
Several concentrated insulin preparations basal insulins based on the blood glucose to basal-bolus regimens with similar rates
are currently available. U-500 regular insu- levels and an understanding of the phar- of hypoglycemia (62). If a patient is still
lin, by definition, is five times as concen- macodynamic profile of each formulation above the A1C target on basal insulin
trated as U-100 regular insulin and has a (pattern control). plus single injection of rapid-acting insulin
delayed onset and longer duration of ac- before the largest meal, advance to a
tion than U-100 regular, possessing both Studies have demonstrated the non- basal-bolus regimen with $2 injections
prandial and basal properties. U-300 glar- inferiority of basal insulin plus a single of rapid-acting insulin before meals. Con-
gine and U-200 degludec are three and injection of rapid-acting insulin at the larg- sider switching patients from one regimen
two times as concentrated as their U-100 est meal relative to basal insulin plus a to another (i.e., premixed analog insulin
formulations and allow higher doses of basal GLP-1 receptor agonist relative to two three times daily to basal-bolus regimen
insulin administration per volume used. daily injections of premixed insulins or vice-versa) if A1C targets are not being
U-300 glargine has a longer duration of ac- (Fig. 8.2). Basal insulin plus GLP-1 recep- met and/or depending on other patient
tion than U-100 glargine. The FDA has also tor agonists are associated with less hy- considerations (60,61). Metformin should
approved a concentrated formulation of poglycemia and with weight loss instead be continued in patients on combination
rapid-acting insulin lispro, U-200 (200 of weight gain but may be less tolerable injectable insulin therapy, if not contra-
units/mL). These concentrated preparations and have a greater cost (58,59). In No- indicated and if tolerated, for further gly-
may be more comfortable for the patient vember 2016, the FDA approved two dif- cemic benefits.
and may improve adherence for patients ferent once-daily fixed-dual combination
with insulin resistance who require large products containing basal insulin plus a References
doses of insulin. While U-500 regular insulin GLP-1 receptor agonist: insulin glargine 1. Peters AL, Laffel L, Eds. American Diabetes
is available in both prefilled pens and vials (a plus lixisenatide and insulin degludec Association/JDRF Type 1 Diabetes Sourcebook.
dedicated syringe was FDA approved in July plus liraglutide. Other options for treat- Alexandria, VA, American Diabetes Association,
2016), other concentrated insulins are avail- ment intensification include adding a sin- 2013
able only in prefilled pens to minimize the gle injection of rapid-acting insulin analog 2. Chiang JL, Kirkman MS, Laffel LMB, Peters AL;
risk of dosing errors. (lispro, aspart, or glulisine) before the Type 1 Diabetes Sourcebook Authors. Type 1 di-
largest meal or stopping the basal insulin abetes through the life span: a position statement
Inhaled Insulin and initiating a premixed (or biphasic) of the American Diabetes Association. Diabetes
Care 2014;37:2034–2054
Inhaled insulin is available for prandial use
with a more limited dosing range. It is contra-
indicated in patients with chronic lung dis-
ease such as asthma and chronic obstructive

S84 Pharmacologic Approaches to Glycemic Treatment Diabetes Care Volume 41, Supplement 1, January 2018

3. Wolpert HA, Atakov-Castillo A, Smith SA, Steil with type 1 diabetes: systematic review and net- 30. Palmer SC, Mavridis D, Nicolucci A, et al. Com-
GM. Dietary fat acutely increases glucose con- work meta-analysis. BMJ 2014;349:g5459 parison of clinical outcomes and adverse events
centrations and insulin requirements in pa- associated with glucose-lowering drugs in patients
tients with type 1 diabetes: implications for 17. Bartley PC, Bogoev M, Larsen J, Philotheou A. with type 2 diabetes: a meta-analysis. JAMA 2016;
carbohydrate-based bolus dose calculation Long-term efficacy and safety of insulin detemir 316:313–324
and intensive diabetes management. Diabetes compared to Neutral Protamine Hagedorn insulin 31. Inzucchi SE, Bergenstal RM, Buse JB, et al.
Care 2013;36:810–816 in patients with type 1 diabetes using a treat-to- Management of hyperglycemia in type 2 diabe-
4. Bell KJ, Toschi E, Steil GM, Wolpert HA. Opti- target basal-bolus regimen with insulin aspart at tes, 2015: a patient-centered approach: update
mized mealtime insulin dosing for fat and protein meals: a 2-year, randomized, controlled trial. Dia- to a position statement of the American Diabe-
in type 1 diabetes: application of a model-based bet Med 2008;25:442–449 tes Association and the European Association
approach to derive insulin doses for open-loop 18. DeWitt DE, Hirsch IB. Outpatient insulin ther- for the Study of Diabetes. Diabetes Care 2015;
diabetes management. Diabetes Care 2016;39: apy in type 1 and type 2 diabetes mellitus: scien- 38:140–149
1631–1634 tific review. JAMA 2003;289:2254–2264 32. Holman RR, Paul SK, Bethel MA, Matthews
5. Bell KJ, Smart CE, Steil GM, Brand-Miller JC, 19. Lane W, Bailey TS, Gerety G, et al.; SWITCH 1. DR, Neil HAW. 10-year follow-up of intensive glu-
King B, Wolpert HA. Impact of fat, protein, and Effect of insulin degludec vs insulin glargine U100 cose control in type 2 diabetes. N Engl J Med 2008;
glycemic index on postprandial glucose control in on hypoglycemia in patients with type 1 diabetes: 359:1577–1589
type 1 diabetes: implications for intensive diabetes the SWITCH 1 Randomized Clinical Trial. JAMA 33. Maruthur NM, Tseng E, Hutfless S, et al. Di-
management in the continuous glucose monitoring 2017;318:33–44 abetes medications as monotherapy or metformin-
era. Diabetes Care 2015;38:1008–1015 20. Home PD, Bergenstal RM, Bolli GB, et al. New based combination therapy for type 2 diabetes: a
6. Yeh H-C, Brown TT, Maruthur N, et al. Compar- insulin glargine 300 units/mL versus glargine systematic review and meta-analysis. Ann Intern
ative effectiveness and safety of methods of in- 100 units/mL in people with type 1 diabetes: a Med 2016;164:740–751
sulin delivery and glucose monitoring for diabetes randomized, phase 3a, open-label clinical trial 34. U.S. Food and Drug Administration. Metformin-
mellitus: a systematic review and meta-analysis. (EDITION 4). Diabetes Care 2015;38:2217–2225 containing drugs: drug safety communication -
Ann Intern Med 2012;157:336–347 revised warnings for certain patients with
7. Bergenstal RM, Klonoff DC, Garg SK, et al.; 21. Bode BW, McGill JB, Lorber DL, Gross JL, reduced kidney function [Internet], 2016. Available
ASPIRE In-Home Study Group. Threshold-based Chang PC, Bregman DB; Affinity 1 Study Group. from http://www.fda.gov/Safety/MedWatch/
insulin-pump interruption for reduction of hypo- Inhaled technosphere insulin compared with in- SafetyInformation/SafetyAlertsforHumanMedical
glycemia. N Engl J Med 2013;369:224–232 jected prandial insulin in type 1 diabetes: a ran- Products/ucm494829.htm?source5govdelivery&
8. Bergenstal RM, Garg S, Weinzimer SA, et al. domized 24-week trial. Diabetes Care 2015;38: utm_medium5email&utm_source5govdelivery.
Safety of a hybrid closed-loop insulin delivery sys- 2266–2273 Accessed 3 October 2016
tem in patients with type 1 diabetes. JAMA 2016; 22. Vella S, Buetow L, Royle P, Livingstone S,
316:1407–1408 Colhoun HM, Petrie JR. The use of metformin in 35. Aroda VR, Edelstein SL, Goldberg RB, et al.;
9. Garg SK, Weinzimer SA, Tamborlane WV, et al. type 1 diabetes: a systematic review of efficacy. Diabetes Prevention Program Research Group.
Glucose outcomes with the in-home use of a hy- Diabetologia 2010;53:809–820 Long-term metformin use and vitamin B12 defi-
brid closed-loop insulin delivery system in adoles- 23. Libman IM, Miller KM, DiMeglio LA, et al.; T1D ciency in the Diabetes Prevention Program Out-
cents and adults with type 1 diabetes. Diabetes Exchange Clinic Network Metformin RCT Study comes Study. J Clin Endocrinol Metab 2016;101:
Technol Ther 2017;19:155–163 Group. Effect of metformin added to insulin on 1754–1761
10. Wood JR, Miller KM, Maahs DM, et al.; T1D glycemic control among overweight/obese ado- 36. Bennett WL, Maruthur NM, Singh S, et al.
Exchange Clinic Network. Most youth with type 1 lescents with type 1 diabetes: a randomized clin- Comparative effectiveness and safety of medica-
diabetes in the T1D Exchange Clinic Registry do not ical trial. JAMA 2015;314:2241–2250 tions for type 2 diabetes: an update including new
meet American Diabetes Association or International 24. Petrie JR, Chaturvedi N, Ford I, et al.; drugs and 2-drug combinations. Ann Intern Med
Society for Pediatric and Adolescent Diabetes clinical REMOVAL Study Group. Cardiovascular and met- 2011;154:602–613
guidelines. Diabetes Care 2013;36:2035–2037 abolic effects of metformin in patients with type 1 37. Vijan S, Sussman JB, Yudkin JS, Hayward RA.
11. Kmietowicz Z. Insulin pumps improve control diabetes (REMOVAL): a double-blind, rando- Effect of patients’ risks and preferences on health
and reduce complications in children with type 1 mised, placebo-controlled trial. Lancet Diabetes gains with plasma glucose level lowering in type 2
diabetes. BMJ 2013;347:f5154 Endocrinol 2017;5:597–609 diabetes mellitus. JAMA Intern Med 2014;174:
25. Dejgaard TF, Frandsen CS, Hansen TS, et al. 1227–1234
12. Phillip M, Battelino T, Atlas E, et al. Nocturnal Efficacy and safety of liraglutide for overweight 38. Institute for Clinical and Economic Review.
glucose control with an artificial pancreas at a di- adult patients with type 1 diabetes and insufficient Controversies in the management of patients with
abetes camp. N Engl J Med 2013;368:824–833 glycaemic control (Lira-1): a randomised, double- type 2 diabetes [Internet], 2014. Available from
13. Nathan DM, Genuth S, Lachin J, et al.; Diabe- blind, placebo-controlled trial. Lancet Diabetes https://icer-review.org/wp-content/uploads/
tes Control and Complications Trial Research Endocrinol 2016;4:221–232 2015/03/CEPAC-T2D-Final-Report-December-22
Group. The effect of intensive treatment of dia- 26. Guo H, Fang C, Huang Y, Pei Y, Chen L, Hu J. .pdf. Accessed 2 November 2017
betes on the development and progression of The efficacy and safety of DPP4 inhibitors in pa-
long-term complications in insulin-dependent tients with type 1 diabetes: a systematic review 39. Truven Health Analytics. Red Book: A Com-
diabetes mellitus. N Engl J Med 1993;329:977– and meta-analysis. Diabetes Res Clin Pract 2016; prehensive, Consistent Drug Pricing Resource
986 121:184–191 [Internet], 2016. Available from: http://www
27. Yang Y, Chen S, Pan H, et al. Safety and effi- .micromedexsolutions.com/micromedex2/librarian.
14. Nathan DM, Cleary PA, Backlund J-YC, et al.; ciency of SGLT2 inhibitor combining with insulin in Accessed 18 July 2017
Diabetes Control and Complications Trial/ subjects with diabetes: systematic review and
Epidemiology of Diabetes Interventions and Com- meta-analysis of randomized controlled trials. 40. Centers for Medicare & Medicaid Services.
plications (DCCT/EDIC) Study Research Group. In- Medicine (Baltimore) 2017;96:e6944 Pharmacy pricing: national average drug acquisi-
tensive diabetes treatment and cardiovascular tion cost [Internet], 2017. Available from https://
disease in patients with type 1 diabetes. N Engl J 28. U.S. Food and Drug Administration. SGLT2 www.medicaid.gov/medicaid/prescription-drugs/
Med 2005;353:2643–2653 inhibitors: drug safety communication - labels to pharmacy-pricing/index.html. Accessed 19 July
15. Diabetes Control and Complications Trial include warnings about too much acid in the blood 2017
(DCCT)/Epidemiology of Diabetes Interventions and serious urinary tract infections [Internet], 2015. 41. Nathan DM, Buse JB, Kahn SE, et al.; GRADE
and Complications (EDIC) Study Research Group. Available from http://www.fda.gov/safety/ Study Research Group. Rationale and design of
Mortality in type 1 diabetes in the DCCT/EDIC medwatch/safetyinformation/safetyalertsforhuma the Glycemia Reduction Approaches in Diabetes:
versus the general population. Diabetes Care nmedicalproducts/ucm475553.htm. Accessed 3 A Comparative Effectiveness Study (GRADE). Di-
2016;39:1378–1383 October 2016 abetes Care 2013;36:2254–2261
16. Tricco AC, Ashoor HM, Antony J, et al. Safety, 42. Blonde L, Merilainen M, Karwe V, Raskin P;
effectiveness, and cost effectiveness of long act- 29. Robertson RP, Davis C, Larsen J, Stratta R, TITRATE Study Group. Patient-directed titra-
ing versus intermediate acting insulin for patients Sutherland DER; American Diabetes Association. tion for achieving glycaemic goals using a once-
Pancreas and islet transplantation in type 1 dia- daily basal insulin analogue: an assessment of
betes. Diabetes Care 2006;29:935

care.diabetesjournals.org Pharmacologic Approaches to Glycemic Treatment S85

two different fasting plasma glucose targets - the in Japanese people with type 2 diabetes using in type 2 diabetes. Diabetes Care 2014;37:2763–
TITRATE study. Diabetes Obes Metab 2009;11: basal insulin and oral antihyperglycaemic drugs: 2773
623–631 glucose control and hypoglycaemia in a random- 59. Eng C, Kramer CK, Zinman B, Retnakaran R.
43. Singh SR, Ahmad F, Lal A, Yu C, Bai Z, Bennett ized controlled trial (EDITION JP 2). Diabetes Obes Glucagon-like peptide-1 receptor agonist and
H. Efficacy and safety of insulin analogues for the Metab 2016;18:366–374 basal insulin combination treatment for the
management of diabetes mellitus: a meta-analysis. 51. Yki-Ja¨rvinen H, Bergenstal RM, Bolli GB, et al. management of type 2 diabetes: a systematic
CMAJ 2009;180:385–397 Glycaemic control and hypoglycaemia with new review and meta-analysis. Lancet 2014;384:
44. Horvath K, Jeitler K, Berghold A, et al. Long- insulin glargine 300 U/ml versus insulin glargine 2228–2234
acting insulin analogues versus NPH insulin (hu- 100 U/ml in people with type 2 diabetes using 60. Dieuzeide G, Chuang L-M, Almaghamsi A,
man isophane insulin) for type 2 diabetes mellitus. basal insulin and oral antihyperglycaemic drugs: Zilov A, Chen J-W, Lavalle-Gonza´lez FJ. Safety
Cochrane Database Syst Rev 2007;2:CD005613 the EDITION 2 randomized 12-month trial including and effectiveness of biphasic insulin aspart 30 in
45. Monami M, Marchionni N, Mannucci E. Long- 6-month extension. Diabetes Obes Metab 2015; people with type 2 diabetes switching from basal-
acting insulin analogues versus NPH human insu- 17:1142–1149 bolus insulin regimens in the A1chieve study. Prim
lin in type 2 diabetes: a meta-analysis. Diabetes 52. Marso SP, McGuire DK, Zinman B, et al. Effi- Care Diabetes 2014;8:111–117
Res Clin Pract 2008;81:184–189 cacy and safety of degludec versus glargine in 61. Mathieu C, Storms F, Tits J, Veneman TF, Colin
46. Owens DR, Traylor L, Mullins P, Landgraf W. type 2 diabetes. N Engl J Med 2017;377:723–732 IM. Switching from premixed insulin to basal-
Patient-level meta-analysis of efficacy and hypo- 53. Rodbard HW, Cariou B, Zinman B, et al.; BEGIN bolus insulin glargine plus rapid-acting insulin:
glycaemia in people with type 2 diabetes initiating Once Long trial investigators. Comparison of insu- the ATLANTIC study. Acta Clin Belg 2013;68:28–
insulin glargine 100U/mL or neutral protamine lin degludec with insulin glargine in insulin-naive 33
Hagedorn insulin analysed according to concomi- subjects with Type 2 diabetes: a 2-year random- 62. Giugliano D, Chiodini P, Maiorino MI,
tant oral antidiabetes therapy. Diabetes Res Clin ized, treat-to-target trial. Diabet Med 2013;30: Bellastella G, Esposito K. Intensification of insulin
Pract 2017;124(Suppl. C):57–65 1298–1304 therapy with basal-bolus or premixed insulin reg-
47. Riddle MC, Rosenstock J, Gerich J; Insulin Glar- 54. Wysham C, Bhargava A, Chaykin L, et al. Effect imens in type 2 diabetes: a systematic review and
gine 4002 Study Investigators. The treat-to-target of insulin degludec vs insulin glargine U100 on meta-analysis of randomized controlled trials. En-
trial: randomized addition of glargine or human hypoglycemia in patients with type 2 diabetes: docrine 2016;51:417–428
NPH insulin to oral therapy of type 2 diabetic the SWITCH 2 Randomized Clinical Trial. JAMA 63. Tuttle KR, Bakris GL, Bilous RW, et al. Diabetic
patients. Diabetes Care 2003;26:3080–3086 2017;318:45–56 kidney disease: a report from an ADA Consensus
48. Hermansen K, Davies M, Derezinski T, 55. Zinman B, Philis-Tsimikas A, Cariou B, et al.; Conference. Diabetes Care 2014;37:2864–2883
Martinez Ravn G, Clauson P, Home P. A 26-week, NN1250-3579 (BEGIN Once Long) Trial Investiga- 64. Neumiller JJ, Alicic RZ, Tuttle KR. Therapeutic
randomized, parallel, treat-to-target trial compar- tors. Insulin degludec versus insulin glargine in considerations for antihyperglycemic agents in di-
ing insulin detemir with NPH insulin as add-on insulin-naive patients with type 2 diabetes: a abetic kidney disease. J Am Soc Nephrol 2017;28:
therapy to oral glucose-lowering drugs in insulin- 1-year, randomized, treat-to-target trial (BEGIN 2263–2274
naive people with type 2 diabetes. Diabetes Care Once Long). Diabetes Care 2012;35:2464–2471 65. U.S. Food and Drug Administration. Cycloset
2006;29:1269–1274 56. Lipska KJ, Hirsch IB, Riddle MC. Human insulin [bromocriptine] prescribing information [In-
49. Bolli GB, Riddle MC, Bergenstal RM, et al.; for type 2 diabetes: an effective, less-expensive ternet], 2017. [Available from https://www
EDITION 3 Study Investigators. New insulin glar- option. JAMA 2017;318:23–24 .accessdata.fda.gov/drugsatfda_docs/label/2017/
gine 300 U/ml compared with glargine 100 U/ml 57. Hua X, Carvalho N, Tew M, Huang ES, Herman 020866s006s007lbl.pdf. Accessed 22 September
in insulin-na¨ıve people with type 2 diabetes on WH, Clarke P. Expenditures and prices of antihy- 2017
oral glucose-lowering drugs: a randomized con- perglycemic medications in the United States: 66. U.S. Food and Drug Administration. Welchol
trolled trial (EDITION 3). Diabetes Obes Metab 2002-2013. JAMA 2016;315:1400–1402 [Colesevelam] prescribing information [Internet],
2015;17:386–394 58. Diamant M, Nauck MA, Shaginian R, et al.; 4B 2014. Available from https://www.accessdata.fda
50. Terauchi Y, Koyama M, Cheng X, et al. New Study Group. Glucagon-like peptide 1 receptor ag- .gov/drugsatfda_docs/label/2011/022362s007lbl
insulin glargine 300 U/ml versus glargine 100 U/ml onist or bolus insulin with optimized basal insulin .pdf. Accessed 22 September 2017

S86 Diabetes Care Volume 41, Supplement 1, January 2018

9. Cardiovascular Disease and Risk American Diabetes Association
Management: Standards of
Medical Care in Diabetesd2018

Diabetes Care 2018;41(Suppl. 1):S86–S104 | https://doi.org/10.2337/dc18-S009

9. CARDIOVASCULAR DISEASE AND RISK MANAGEMENT The American Diabetes Association (ADA) “Standards of Medical Care in Diabetes”
includes ADA’s current clinical practice recommendations and is intended to provide
the components of diabetes care, general treatment goals and guidelines, and tools
to evaluate quality of care. Members of the ADA Professional Practice Committee, a
multidisciplinary expert committee, are responsible for updating the Standards of
Care annually, or more frequently as warranted. For a detailed description of
ADA standards, statements, and reports, as well as the evidence-grading system
for ADA’s clinical practice recommendations, please refer to the Standards of Care
Introduction. Readers who wish to comment on the Standards of Care are invited to
do so at professional.diabetes.org/SOC.

For prevention and management of diabetes complications in children and Suggested citation: American Diabetes Association.
adolescents, please refer to Section 12 “Children and Adolescents.” 9. Cardiovascular disease and risk management:
Standards of Medical Care in Diabetesd2018.
Atherosclerotic cardiovascular disease (ASCVD)ddefined as coronary heart Diabetes Care 2018;41(Suppl. 1):S86–S104
disease, cerebrovascular disease, or peripheral arterial disease presumed to be of
atherosclerotic origindis the leading cause of morbidity and mortality for individuals © 2017 by the American Diabetes Association.
with diabetes and is the largest contributor to the direct and indirect costs of Readers may use this article as long as the work
diabetes. Common conditions coexisting with type 2 diabetes (e.g., hypertension is properly cited, the use is educational and not
and dyslipidemia) are clear risk factors for ASCVD, and diabetes itself confers in- for profit, and the work is not altered. More infor-
dependent risk. Numerous studies have shown the efficacy of controlling individual mation is available at http://www.diabetesjournals
cardiovascular risk factors in preventing or slowing ASCVD in people with diabetes. .org/content/license.
Furthermore, large benefits are seen when multiple cardiovascular risk factors are
addressed simultaneously. Under the current paradigm of aggressive risk factor
modification in patients with diabetes, there is evidence that measures of 10-year
coronary heart disease (CHD) risk among U.S. adults with diabetes have improved
significantly over the past decade (1) and that ASCVD morbidity and mortality have
decreased (2–4).

Therefore, cardiovascular risk factors should be systematically assessed at least
annually in all patients with diabetes. These risk factors include hypertension, dyslipi-
demia, smoking, a family history of premature coronary disease, chronic kidney dis-
ease, and the presence of albuminuria. Modifiable abnormal risk factors should be
treated as described in these guidelines.

HYPERTENSION/BLOOD PRESSURE CONTROL

Hypertension, defined as a sustained blood pressure $140/90 mmHg, is common
among patients with either type 1 or type 2 diabetes. Hypertension is a major risk
factor for both ASCVD and microvascular complications. Moreover, numerous studies
have shown that antihypertensive therapy reduces ASCVD events, heart failure, and
microvascular complications. Please refer to the American Diabetes Association (ADA)

care.diabetesjournals.org Cardiovascular Disease and Risk Management S87

position statement “Diabetes and Hyper- c Lower systolic and diastolic blood patients who have been educated about
tension” for a detailed review of the epi- pressure targets, such as 130/80 added treatment burden, side effects, and
demiology, diagnosis, and treatment of mmHg, may be appropriate for costs, as discussed below.
hypertension (5). individuals at high risk of cardio-
vascular disease, if they can be Additional studies, such as the Systolic
Screening and Diagnosis achieved without undue treat- Blood Pressure Intervention Trial (SPRINT)
ment burden. C and the Hypertension Optimal Treatment
Recommendations (HOT) trial, also examined effects of inten-
c In pregnant patients with diabetes sive versus standard control (Table 9.1),
c Blood pressure should be measured and preexisting hypertension who though the relevance of their results to
at every routine clinical visit. Pa- are treated with antihypertensive people with diabetes is less clear. The
tients found to have elevated blood therapy, blood pressure targets of Action in Diabetes and Vascular Disease:
pressure ($140/90) should have 120–160/80–105 mmHg are sug- Preterax and Diamicron MR Controlled
blood pressure confirmed using gested in the interest of optimiz- Evaluation–Blood Pressure (ADVANCE
multiple readings, including meas- ing long-term maternal health BP) trial did not explicitly test blood pres-
urments on a separate day, to diag- and minimizing impaired fetal sure targets (17); the achieved blood
nose hypertension. B growth. E pressure in the intervention group was
higher than that achieved in the ACCORD
c All hypertensive patients with dia- Randomized clinical trials have demon- BP intensive arm and would be consistent
betes should monitor their blood strated unequivocally that treatment of with a target blood pressure of ,140/90
pressure at home. B hypertension to blood pressure ,140/90 mmHg. Notably, ACCORD BP and SPRINT
mmHg reduces cardiovascular events measured blood pressure using auto-
Blood pressure should be measured by a as well as microvascular complications mated office blood pressure measure-
trained individual and should follow the (9–15). Therefore, patients with type 1 ments, which yields values that are
guidelines established for the general or type 2 diabetes who have hypertension generally lower than typical office blood
population: measurement in the seated should, at a minimum, be treated to blood pressure readings by approximately
position, with feet on the floor and arm pressure targets of ,140/90 mmHg. In- 5–10 mmHg (18), suggesting that imple-
supported at heart level, after 5 min of tensification of antihypertensive ther- menting the ACCORD BP or SPRINT pro-
rest. Cuff size should be appropriate for apy to target blood pressures lower tocols in an outpatient clinic might require
the upper-arm circumference. Elevated than ,140/90 mmHg (e.g., ,130/80 or a systolic blood pressure target higher
values should be confirmed on a separate ,120/80 mmHg) may be beneficial for than ,120 mmHg.
day. Postural changes in blood pressure selected patients with diabetes such as
and pulse may be evidence of autonomic those with a high risk of cardiovascular Meta-analyses of Trials
neuropathy and therefore require adjust- disease. Such intensive blood pressure
ment of blood pressure targets. Orthostatic control has been evaluated in large ran- To clarify optimal blood pressure targets
blood pressure measurements should be domized clinical trials and meta-analyses in patients with diabetes, meta-analyses
checked on initial visit and as indicated. of clinical trials. have stratified clinical trials by mean
baseline blood pressure or mean blood
Home blood pressure self-monitoring Randomized Controlled Trials of Intensive pressure attained in the intervention (or
and 24-h ambulatory blood pressure Versus Standard Blood Pressure Control intensive treatment) arm. Based on these
monitoring may provide evidence of analyses, antihypertensive treatment ap-
white coat hypertension, masked hyper- The Action to Control Cardiovascular Risk pears to be beneficial when mean base-
tension, or other discrepancies between in Diabetes blood pressure (ACCORD BP) line blood pressure is $140/90 mmHg or
office and “true” blood pressure (5). In trial provides the strongest direct assess- mean attained intensive blood pressure
addition to confirming or refuting a diag- ment of the benefits and risks of intensive is $130/80 mmHg (5,9,12–14). Among
nosis of hypertension, home blood pres- blood pressure control among people trials with lower baseline or attained
sure assessment may be useful to monitor with type 2 diabetes (16). In ACCORD BP, blood pressure, antihypertensive treat-
antihypertensive treatment. Studies of indi- compared with standard blood pres- ment reduced the risk of stroke, reti-
viduals without diabetes found that home sure control (target systolic blood pres- nopathy, and albuminuria, but effects
measurements may better correlate with sure ,140 mmHg), intensive blood on other ASCVD outcomes and heart
ASCVD risk than office measurements pressure control (target systolic blood failure were not evident. Taken to-
(6,7). Moreover, home blood pressures pressure ,120 mmHg) did not reduce to- gether, these meta-analyses consis-
may improve patient medication adherence tal major atherosclerotic cardiovascular tently show that treating patients with
and thus help reduce cardiovascular risk (8). events but did reduce the risk of stroke, baseline blood pressure $140 mmHg to
at the expense of increased adverse targets ,140 mmHg is beneficial, while
Treatment Goals events (Table 9.1). The ACCORD BP re- more intensive targets may offer addi-
sults suggest that blood pressure targets tional, though probably less robust, ben-
Recommendations more intensive than ,140/90 mmHg are efits.
not likely to improve cardiovascular out-
c Most patients with diabetes and comes among most people with type 2 di- Individualization of Treatment Targets
hypertension should be treated abetes but may be reasonable in selected
to a systolic blood pressure goal of Patients and clinicians should engage in a
,140 mmHg and a diastolic blood shared decision-making process to deter-
pressure goal of ,90 mmHg. A mine individual blood pressure targets,

S88 Cardiovascular Disease and Risk Management Diabetes Care Volume 41, Supplement 1, January 2018

Table 9.1—Randomized controlled trials of intensive versus standard hypertension treatment strategies

Clinical trial Population Intensive Standard Outcomes

ACCORD BP (16) 4,733 participants with T2D Systolic blood Systolic blood pressure c No benefit in primary end point: composite of
target: 130–140 mmHg nonfatal MI, nonfatal stroke, and CVD death
ADVANCE BP (17) aged 40–79 years with pressure target:
HOT (143)
SPRINT (144) prior evidence of CVD or ,120 mmHg

multiple cardiovascular Achieved (mean) Achieved (mean) c Stroke risk reduced 41% with intensive
systolic/diastolic: control, not sustained through follow-up
risk factors systolic/diastolic: 133.5/70.5 mmHg beyond the period of active treatment

119.3/64.4

mmHg

11,140 participants with T2D Intervention: Control: placebo c Adverse events more common in intensive
group, particularly elevated serum creatinine
aged 55 years and older a single-pill, and electrolyte abnormalities

with prior evidence of CVD fixed-dose c Intervention reduced risk of primary
composite end point of major macrovascular
or multiple cardiovascular combination of and microvascular events (9%), death from
any cause (14%), and death from CVD (18%)

risk factors perindopril and

indapamide

Achieved (mean) Achieved (mean) c 6-year observational follow-up found
systolic/diastolic: reduction in risk of death in intervention group
systolic/diastolic: 141.6/75.2 mmHg attenuated but still significant (142)

136/73 mmHg

18,790 participants, Diastolic blood Diastolic blood pressure c In the overall trial, there was no cardiovascular
including 1,501 with pressure target: target: #90 mmHg benefit with more intensive targets
diabetes #80 mmHg c In the subpopulation with diabetes, an
intensive diastolic target was associated with
a significantly reduced risk (51%) of CVD events

9,361 participants without Systolic blood Systolic blood pressure c Intensive systolic blood pressure target
target: ,140 mmHg lowered risk of the primary composite
diabetes pressure target: outcome 25% (MI, ACS, stroke, heart failure,
Achieved (mean): and death due to CVD)
,120 mmHg 136.2 mmHg
c Intensive target reduced risk of death 27%
Achieved (mean):
121.4 mmHg c Intensive therapy increased risks of electrolyte
abnormalities and AKI

CVD, cardiovascular disease; T2D, type 2 diabetes. Data from this table can also be found in the ADA position statement “Diabetes and Hypertension” (5).

with the acknowledgment that the ben- adults, such as functional limitations, overweight or obese; a Dietary
efits and risks of intensive blood pres- polypharmacy, and multimorbidity, Approaches to Stop Hypertension–
sure targets are uncertain and may vary may be best suited for less intensive style dietary pattern including reduc-
across patients (5). Similar to the factors blood pressure targets. Notably, there ing sodium and increasing potassium
that influence management of hyper- is an absence of high-quality data avail- intake; moderation of alcohol intake;
glycemia, factors that influence blood able to guide blood pressure targets in and increased physical activity. B
pressure treatment targets may include type 1 diabetes.
risks of treatment (e.g., hypotension, Lifestyle management is an important
drug adverse effects), life expectancy, co- Based on current evidence, ADA rec- component of hypertension treatment
morbidities including vascular compli- ommends hypertension diagnosis and because it lowers blood pressure, enhan-
cations, patient attitude and expected treatment as outlined, emphasizing individ- ces the effectiveness of some antihyper-
treatment efforts, and resources and ualization of blood pressure targets. ADA tensive medications, promotes other
support system (19). Specific factors to is aware of hypertension recommendations aspects of metabolic and vascular health,
consider are the absolute risk of car- from other organizations (20a). The ADA and generally leads to few adverse ef-
diovascular events (15,20), risk of pro- Professional Practice Committee continu- fects. Lifestyle therapy consists of reduc-
gressive kidney disease as reflected by ously reviews and considers all studies, par- ing excess body weight through caloric
albuminuria, adverse effects, age, and ticularly high-quality trials including people restriction, restricting sodium intake
overall treatment burden. Patients with diabetes, for potential incorporation (,2,300 mg/day), increasing consump-
who have higher risk of cardiovascular in future recommendations. tion of fruits and vegetables (8–10 serv-
events (particularly stroke) or albumin- ings per day) and low-fat dairy products
uria and who are able to attain intensive Treatment Strategies (2–3 servings per day), avoiding excessive
blood pressure control relatively easily Lifestyle Intervention alcohol consumption (no more than
and without substantial adverse effects 2 servings per day in men and no more
may be best suited for intensive blood Recommendation than 1 serving per day in women) (21),
pressure targets. In contrast, patients and increasing activity levels (22).
with conditions more common in older c For patients with blood pressure
.120/80 mmHg, lifestyle inter-
vention consists of weight loss if

care.diabetesjournals.org Cardiovascular Disease and Risk Management S89

These lifestyle interventions are rea- Initial Number of Antihypertensive Medications. analysis of randomized clinical trials found a
sonable for individuals with diabetes small benefit of evening versus morning
and mildly elevated blood pressure Initial treatment for people with diabetes dosing of antihypertensive medications
(systolic .120 mmHg or diastolic .80 depends on the severity of hypertension with regard to blood pressure control but
mmHg) and should be initiated along with (Fig. 9.1). Those with blood pressure be- had no data on clinical effects (35). In two
pharmacologic therapy when hypertension tween 140/90 mmHg and 159/99 mmHg subgroup analyses of a single subsequent
is diagnosed (Fig. 9.1) (22). A lifestyle ther- may begin with a single drug. For patients randomized controlled trial, moving at
apy plan should be developed in collabo- with blood pressure $160/100 mmHg, least one antihypertensive medication
ration with the patient and discussed as initial pharmacologic treatment with to bedtime significantly reduced cardio-
part of diabetes management. two antihypertensive medications is rec- vascular events, but results were based
ommended in order to more effectively on a small number of events (36).
Pharmacologic Interventions achieve adequate blood pressure control
(23,24). Single-pill antihypertensive com- Hyperkalemia and AKI. Treatment with ACE
Recommendations binations may improve medication ad- inhibitors or ARBs can cause AKI and hyper-
herence in some patients (25). kalemia, while diuretics can cause AKI and
c Patients with confirmed office-based Classes of Antihypertensive Medications. Ini- either hypokalemia or hyperkalemia (de-
blood pressure $140/90 mmHg tial treatment for hypertension should pending on mechanism of action) (37,38).
should, in addition to lifestyle ther- include any of the drug classes demon- Detection and management of these ab-
apy, have prompt initiation and timely strated to reduce cardiovascular events normalities is important because AKI and
titration of pharmacologic therapy to in patients with diabetes: ACE inhibitors hyperkalemia each increase the risks of
achieve blood pressure goals. A (26,27), angiotensin receptor blockers cardiovascular events and death (39).
(ARBs) (26,27), thiazide-like diuretics Therefore, serum creatinine and potassium
c Patients with confirmed office-based (28), or dihydropyridine calcium channel should be monitored during treatment with
blood pressure $160/100 mmHg blockers (29). For patients with albumin- an ACE inhibitor, ARB, or diuretic, particu-
should, in addition to lifestyle ther- uria (urine albumin-to-creatinine ratio larly among patients with reduced glomer-
apy, have prompt initiation and [UACR] $30 mg/g), initial treatment ular filtration who are at increased risk of
timely titration of two drugs or a sin- should include an ACE inhibitor or ARB hyperkalemia and AKI (37,38,40).
gle-pill combination of drugs dem- in order to reduce the risk of progressive
onstrated to reduce cardiovascular kidney disease (5) (Fig. 9.1). In the ab- Resistant Hypertension
events in patients with diabetes. A sence of albuminuria, risk of progressive
kidney disease is low, and ACE inhibitors Recommendation
c Treatment for hypertension should and ARBs have not been found to afford
include drug classes demonstrated superior cardioprotection when compared c Patients with hypertension who are
to reduce cardiovascular events in pa- with thiazide-like diuretics or dihydro- not meeting blood pressure targets
tients with diabetes (ACE inhibitors, pyridine calcium channel blockers(30). on three classes of antihypertensive
angiotensin receptor blockers, thiazide- b-Blockers may be used for the treatment medications (including a diuretic)
like diuretics, or dihydropyridine calcium of prior myocardial infarction (MI), ac- should be considered for mineralocor-
channel blockers). A tive angina, or heart failure but have not ticoid receptor antagonist therapy. B
been shown to reduce mortality as blood
c Multiple-drug therapy is generally pressure-lowering agents in the absence Resistant hypertension is defined as
required to achieve blood pressure of these conditions (11,31). blood pressure $140/90 mmHg despite
targets. However, combinations of Multiple-Drug Therapy. Multiple-drug ther- a therapeutic strategy that includes ap-
ACE inhibitors and angiotensin re- apy is often required to achieve blood propriate lifestyle management plus a di-
ceptor blockers and combinations pressure targets (Fig. 9.1), particularly in uretic and two other antihypertensive
of ACE inhibitors or angiotensin re- the setting of diabetic kidney disease. drugs belonging to different classes at
ceptor blockers with direct renin in- However, the use of both ACE inhibitors adequate doses. Prior to diagnosing resis-
hibitors should not be used. A and ARBs in combination, or the combina- tant hypertension, a number of other
tion of an ACE inhibitor or ARB and a direct conditions should be excluded, including
c An ACE inhibitor or angiotensin re- renin inhibitor, is not recommended given medication nonadherence, white coat
ceptor blocker, at the maximumly the lack of added ASCVD benefit and in- hypertension, and secondary hyperten-
tolerated dose indicated for blood creased rate of adverse eventsdnamely, sion. In general, barriers to medication
pressure treatment, is the recom- hyperkalemia, syncope, and acute kidney adherence (such as cost and side effects)
mended first-line treatment for hy- injury (AKI) (32–34). Titration of and/or should be identified and addressed
pertension in patients with diabetes addition of further blood pressure medi- (Fig. 9.1). Mineralocorticoid receptor an-
and urinary albumin-to-creatinine cations should be made in a timely fash- tagonists are effective for management of
ratio $300 mg/g creatinine A or ion to overcome clinical inertia in resistant hypertension in patients with
30–299 mg/g creatinine B. If one achieving blood pressure targets. type 2 diabetes when added to existing
class is not tolerated, the other Bedtime Dosing. Growing evidence suggests treatment with a ACE inhibitor or ARB,
should be substituted B. that there is an association between the thiazide-like diuretic, and dihydropyridine
absence of nocturnal blood pressure dip- calcium channel blocker (41). Miner-
c For patients treated with an ACE in- ping and the incidence of ASCVD. A meta- alocorticoid receptor antagonists
hibitor, angiotensin receptor blocker, also reduce albuminuria and have
or diuretic, serum creatinine/estimated
glomerular filtration rate and serum
potassium levels should be monitored
at least annually. B

S90 Cardiovascular Disease and Risk Management Diabetes Care Volume 41, Supplement 1, January 2018

Figure 9.1—Recommendations for the treatment of confirmed hypertension in people with diabetes. *An ACE inhibitor (ACEi) or ARB is suggested to treat
hypertension for patients with UACR 30–299 mg/g creatinine and strongly recommended for patients with UACR $300 mg/g creatinine. **Thiazide-like
diuretic; long-acting agents shown to reduce cardiovascular events, such as chlorthalidone and indapamide, are preferred. ***Dihydropyridine calcium
channel blocker. BP, blood pressure. This figure can also be found in the ADA position statement “Diabetes and Hypertension” (5).

additional cardiovascular benefits monitoring for serum creatinine and po- Pregnancy and Antihypertensive Medications.
(42–45). However, adding a mineralocor- tassium in these patients, and long-term
ticoid receptor antagonist to a regimen outcome studies are needed to better Since there is a lack of randomized con-
including an ACE inhibitor or ARB may evaluate the role of mineralocorticoid re- trolled trials of antihypertensive therapy
increase the risk for hyperkalemia, em- ceptor antagonists in blood pressure in pregnant women with diabetes, rec-
phasizing the importance of regular management. ommendations for the management of
hypertension in pregnant women with

care.diabetesjournals.org Cardiovascular Disease and Risk Management S91

diabetes should be similar to those for all LIPID MANAGEMENT In adults with diabetes, it is reasonable to
pregnant women. The American College obtain a lipid profile (total cholesterol,
of Obstetricians and Gynecologists (ACOG) Lifestyle Intervention LDL cholesterol, HDL cholesterol, and tri-
has recommended that women with mild to glycerides) at the time of diagnosis, at the
moderate gestational hypertension (systolic Recommendations initial medical evaluation, and at least ev-
blood pressure ,160 mmHg or diastolic ery 5 years thereafter in patients under
blood pressure ,110 mmHg) do not need c Lifestyle modification focusing on the age of 40 years. In younger patients
to be treated with antihypertensive med- weight loss (if indicated); the reduc- with longer duration of disease (such as
ications as there is no benefit identified tion of saturated fat, trans fat, and those with youth-onset type 1 diabetes),
that clearly outweighs potential risks of cholesterol intake; increase of die- more frequent lipid profiles may be rea-
therapy (46). A 2014 Cochrane systematic tary n-3 fatty acids, viscous fiber, sonable. A lipid panel should also be ob-
review of antihypertensive therapy for and plant stanols/sterols intake; tained immediately before initiating
mild to moderate chronic hypertension and increased physical activity statin therapy. Once a patient is taking a
that included 49 trials and over 4,700 should be recommended to im- statin, LDL cholesterol levels should be
women did not find any conclusive evi- prove the lipid profile in patients assessed 4–12 weeks after initiation of
dence for or against blood pressure treat- with diabetes. A statin therapy, after any change in dose,
ment to reduce the risk of preeclampsia and on an individual basis (e.g., to moni-
for the mother or effects on perinatal c Intensify lifestyle therapy and opti- tor for medication adherence and effi-
outcomes such as preterm birth, small- mize glycemic control for patients cacy). In cases where patients are
for-gestational-age infants, or fetal death with elevated triglyceride levels adherent but the LDL cholesterol level is
(47). For pregnant women who require ($150 mg/dL [1.7 mmol/L]) and/ not responding, clinical judgment is rec-
antihypertensive therapy, systolic blood or low HDL cholesterol (,40 mg/dL ommended to determine the need for
pressure levels of 120–160 mmHg and di- [1.0 mmol/L] for men, ,50 mg/dL and timing of lipid panels. In individual
astolic blood pressure levels of 80–105 [1.3 mmol/L] for women). C patients, the highly variable LDL choles-
mmHg are suggested to optimize mater- terol–lowering response seen with statins
nal health without risking fetal harm. Lifestyle intervention, including weight is poorly understood (50). Clinicians
Lower targets (systolic blood pressure loss, increased physical activity, and med- should attempt to find a dose or alterna-
110–119 mmHg and diastolic blood pres- ical nutrition therapy, allows some pa- tive statin that is tolerable, if side effects
sure 65–79 mmHg) may contribute to im- tients to reduce ASCVD risk factors. occur. There is evidence for benefit from
proved long-term maternal health; however, Nutrition intervention should be tailored even extremely low, less than daily statin
they may be associated with impaired fetal according to each patient’s age, diabetes doses (51).
growth. Pregnant women with hypertension type, pharmacologic treatment, lipid lev-
and evidence of end-organ damage from els, and medical conditions. Statin Treatment
cardiovascular and/or renal disease may
be considered for lower blood pressure Recommendations should focus on re- Recommendations
targets to avoid progression of these con- ducing saturated fat, cholesterol, and trans
ditions during pregnancy. fat intake and increasing plant stanols/ c For patients of all ages with diabe-
sterols, n-3 fatty acids, and viscous fiber tes and atherosclerotic cardiovas-
During pregnancy, treatment with ACE (such as in oats, legumes, and citrus) in- cular disease, high-intensity statin
inhibitors, ARBs, and spironolactone are take. Glycemic control may also beneficially therapy should be added to lifestyle
contraindicated as they may cause fetal modify plasma lipid levels, particularly in therapy. A
damage. Antihypertensive drugs known to patients with very high triglycerides and
be effective and safe in pregnancy include poor glycemic control. See Section 4 c For patients with diabetes aged
methyldopa, labetalol, and long-acting “Lifestyle Management” for additional ,40 years with additional athero-
nifedipine, while hydralzine may be consid- nutrition information. sclerotic cardiovascular disease
ered in the acute management of hyperten- risk factors, the patient and provider
sion in pregnancy or severe preeclampsia Ongoing Therapy and Monitoring With should consider using moderate-
(46). Diuretics are not recommended for Lipid Panel intensity statin in addition to lifestyle
blood pressure control in pregnancy but therapy. C
may be used during late-stage pregnancy Recommendations
if needed for volume control (46,48). c For patients with diabetes aged 40–
ACOG also recommends that postpartum c In adults not taking statins or other 75 years A and .75 years B without
patients with gestational hypertension, pre- lipid-lowering therapy, it is reasonable atherosclerotic cardiovascular dis-
eclampsia, and superimposed preeclampsia to obtain a lipid profile at the time of ease, use moderate-intensity statin
have their blood pressures observed for 72 h diabetes diagnosis, at an initial medi- in addition to lifestyle therapy.
in the hospital and for 7–10 days postpar- cal evaluation, and every 5 years
tum. Long-term follow-up is recommended thereafter if under the age of 40 years, c In clinical practice, providers may
for these women as they have increased life- or more frequently if indicated. E need to adjust the intensity of statin
time cardiovascular risk (49). See Section 13 therapy based on individual patient
“Management of Diabetes in Pregnancy” c Obtain a lipid profile at initiation of response to medication (e.g., side
for additional information. statins or other lipid-lowering ther- effects, tolerability, LDL cholesterol
apy, 4–12 weeks after initiation or a levels, or percent LDL reduction on
change in dose, and annually thereafter statin therapy). For patients who do
as it may help to monitor the response not tolerate the intended intensity
to therapy and inform adherence. E

S92 Cardiovascular Disease and Risk Management Diabetes Care Volume 41, Supplement 1, January 2018

of statin, the maximally tolerated Table 9.2—Recommendations for statin and combination treatment in adults with
statin dose should be used. E
c For patients with diabetes and ath- diabetes
erosclerotic cardiovascular disease, if
LDL cholesterol is $70 mg/dL on Age ASCVD Recommended statin intensity^and
maximally tolerated statin dose,
consider adding additional LDL- combination treatment*
lowering therapy (such as ezetimibe
or PCSK9 inhibitor) after evaluating ,40 years No None†
the potential for further athero- Yes High
sclerotic cardiovascular disease
risk reduction, drug-specific ad- c If LDL cholesterol $70 mg/dL despite maximally tolerated statin
verse effects, and patient preferen- dose, consider adding additional LDL-lowering therapy (such as
ces. Ezetimibe may be preferred ezetimibe or PCSK9 inhibitor)#
due to lower cost. A
c Statin therapy is contraindicated in $40 years No Moderate‡
pregnancy. B Yes High

Initiating Statin Therapy Based on Risk c If LDL cholesterol $70 mg/dL despite maximally tolerated statin
dose, consider adding additional LDL-lowering therapy (such as
Patients with type 2 diabetes have an in- ezetimibe or PCSK9 inhibitor)
creased prevalence of lipid abnormalities,
contributing to their high risk of ASCVD. ^*In addition to lifestyle therapy. For patients who do not tolerate the intended intensity of statin,
Multiple clinical trials have demonstrated
the beneficial effects of statin therapy on the maximally tolerated statin dose should be used. †Moderate-intensity statin may be considered
ASCVD outcomes in subjects with and based on risk-benefit profile and presence of ASCVD risk factors. ASCVD risk factors include LDL cholesterol
without CHD (52,53). Subgroup analyses $100 mg/dL (2.6 mmol/L), high blood pressure, smoking, chronic kidney disease, albuminuria, and
of patients with diabetes in larger trials family history of premature ASCVD. ‡High-intensity statin may be considered based on risk-benefit
(54–58) and trials in patients with diabe- profile and presence of ASCVD risk factors. #Adults aged ,40 years with prevalent ASCVD were not
tes (59,60) showed significant primary well represented in clinical trials of non-statin–based LDL reduction. Before initiating combination
and secondary prevention of ASCVD lipid-lowering therapy, consider the potential for further ASCVD risk reduction, drug-specific adverse
events and CHD death in patients with
diabetes. Meta-analyses, including data effects, and patient preferences.
from over 18,000 patients with diabetes
from 14 randomized trials of statin therapy death and nonfatal MI) are greatest in The Risk Calculator
(mean follow-up 4.3 years), demonstrate a people with high baseline ASCVD risk
9% proportional reduction in all-cause (known ASCVD and/or very high LDL cho- The American College of Cardiology/
mortality and 13% reduction in vascular lesterol levels), but the overall benefits of American Heart Association ASCVD risk
mortality for each mmol/L (39 mg/dL) re- statin therapy in people with diabetes at calculator is generally a useful tool to esti-
duction in LDL cholesterol (61). moderate or even low risk for ASCVD are mate 10-year ASCVD risk (my.americanheart
convincing (62,63). The relative benefit of .org). However, as diabetes itself confers
Accordingly, statins are the drugs of lipid-lowering therapy has been uniform increased risk for ASCVD and risk calcula-
choice for LDL cholesterol lowering and across most subgroups tested (53,61), in- tors in general do not account for the
cardioprotection. Table 9.2 shows recom- cluding subgroups that varied with re- duration of diabetes or the presence of
mended lipid-lowering strategies, and Ta- spect to age and other risk factors. other complications such as albuminuria,
ble 9.3 shows the two statin dosing the risk calculator has limited use for as-
intensities that are recommended for Risk Stratification sessing cardiovascular risk in individuals
use in clinical practice: high-intensity with diabetes.
statin therapy will achieve approxi- Two broad groups of patients exist for
mately a 50% reduction in LDL choles- management of cardiovascular risk: those Recently, risk scores and other cardio-
terol, and moderate-intensity statin with documented ASCVD (as defined vascular biomarkers have been devel-
regimens achieve 30–50% reductions in above) and those without; treatment is oped for risk stratification of secondary
LDL cholesterol. Low-dose statin therapy often referred to as “secondary” and “pri- prevention patients (i.e., those who are
is generally not recommended in patients mary” prevention, respectively. Because already high risk because they have
with diabetes but is sometimes the only risk is higher in patients with ASCVD, ASCVD) but are not yet in widespread
dose of statin that a patient can tolerate. more intensive therapy is indicated and use (67,68). With newer, more expensive
For patients who do not tolerate the has been shown to be of benefit in mul- lipid-lowering therapies now available,
intended intensity of statin, the maximally tiple large randomized cardiovascular use of these risk assessments may help
tolerated statin dose should be used. outcomes trials (61,64–66). target these new therapies to “higher
risk” ASCVD patients in the future.
As in those without diabetes, absolute
reductions in ASCVD outcomes (CHD Table 9.3—High-intensity and moderate-intensity statin therapy*

High-intensity statin therapy (lowers LDL Moderate-intensity statin therapy
cholesterol by $50%) (lowers LDL cholesterol by 30% to 50%)

Atorvastatin 40–80 mg Atorvastatin 10–20 mg
Rosuvastatin 20–40 mg Rosuvastatin 5–10 mg
Simvastatin 20–40 mg
Pravastatin 40–80 mg
Lovastatin 40 mg

Fluvastatin XL 80 mg
Pitavastatin 2–4 mg

*Once-daily dosing. XL, extended release.

care.diabetesjournals.org Cardiovascular Disease and Risk Management S93

Primary Prevention (Patients Without ASCVD) Association and American Diabetes Asso- diabetes (27% of participants), the com-
ciation” (69) for additional discussion. bination of moderate-intensity simvasta-
For primary prevention, moderate-dose tin (40 mg) and ezetimibe (10 mg)
statin therapy is recommended for those Secondary Preventions (Patients With showed a significant reduction of major
40 years and older (55,62,63), though adverse cardiovascular events with an ab-
high-intensity therapy may be considered ASCVD) solute risk reduction of 5% (40% vs. 45%)
on an individual basis in the context of ad- and relative risk reduction of 14% (RR
ditional ASCVD risk factors. The evidence is High-intensity statin therapy is recommen- 0.86 [95% CI 0.78–0.94]) over moderate-
strong for patients with diabetes aged 40– ded for all patients with diabetes and intensity simvastatin (40 mg) alone (65).
75 years, an age-group well represented ASCVD. This recommendation is based on
in statin trials showing benefit. the Cholesterol Treatment Trialists’ Collab- Statins and PCSK9 Inhibitors
oration involving 26 statin trials, of which
The evidence is lower for patients 5 compared high-intensity versus moderate- Placebo-controlled trials evaluating the
aged .75 years; relatively few older pa- intensity statins. Together, they found re- addition of the PCSK9 inhibitors evolo-
tients with diabetes have been enrolled in ductions in nonfatal cardiovascular events cumab and alirocumab to maximally
primary prevention trials. However, het- with more intensive therapy, in patients tolerated doses of statin therapy in par-
erogeneity by age has not been seen in with and without diabetes (53,57,64). ticipants who were at high risk for ASCVD
the relative benefit of lipid-lowering ther- demonstrated an average reduction in
apy in trials that included older partici- Over the past few years, there have LDL cholesterol ranging from 36 to 59%.
pants (53,60,61), and because older age been multiple large randomized trials in- These agents have been approved as ad-
confers higher risk, the absolute benefits vestigating the benefits of adding nonsta- junctive therapy for patients with ASCVD
are actually greater (53,65). Moderate- tin agents to statin therapy, including or familial hypercholesterolemia who are
intensity statin therapy is recommended three that evaluated further lowering of receiving maximally tolerated statin ther-
in patients with diabetes that are 75 years LDL cholesterol with ezetimibe (65), apy but require additional lowering of LDL
or older. However, the risk-benefit profile PCSK9 inhibitors (66), and, cholesteryl es- cholesterol (71,72).
should be routinely evaluated in this pop- ter transfer protein [CETP] inhibitors, an
ulation, with downward titration of dose investigational class of drugs with some The effects of PCSK9 inhibition on
performed as needed. See Section 11 recent supportive data (70). Each trial ASCVD outcomes was investigated in
“Older Adults” for more details on clinical found a significant benefit in the reduc- the Further Cardiovascular Outcomes Re-
considerations for this population. tion of ASCVD events that was directly search With PCSK9 Inhibition in Subjects
related to the degree of further LDL cho- With Elevated Risk (FOURIER) trial, which
Age <40 Years and/or Type 1 Diabetes. Very lesterol lowering. These three large trials enrolled 27,564 patients with prior
little clinical trial evidence exists for pa- comprised over 75,000 patients and ASCVD and an additional high-risk feature
tients with type 2 diabetes under the age 250,000 patient-years of follow-up, and who were receiving their maximally toler-
of 40 years or for patients with type 1 di- approximately one-third of participants ated statin therapy (two-thirds were on
abetes of any age. In the Heart Protection had diabetes. For patients with ASCVD high-intensity statin) but who still had an
Study (lower age limit 40 years), the sub- who are on high-intensity (and maximally LDL cholesterol $70 mg/dL or a non-HDL
group of ;600 patients with type 1 dia- tolerated) statin therapy and have an LDL cholesterol $100 mg/dL (66). Patients
betes had a proportionately similar, cholesterol $70 mg/dL, the addition of were randomized to receive subcutane-
although not statistically significant, re- nonstatin LDL-lowering therapy is recom- ous injections of evolocumab (either
duction in risk as patients with type 2 di- mended after considering the potential for 140 mg every 2 weeks or 420 mg every
abetes (55). Even though the data are not further ASCVD risk reduction, drug-specific month based on patient preference) ver-
definitive, similar statin treatment ap- adverse effects, and patient preferences. sus placebo. Evolocumab reduced LDL
proaches should be considered for pa- cholesterol by 59% from a median of
tients with type 1 or type 2 diabetes, Combination Therapy for LDL 92 to 30 mg/dL in the treatment arm.
particularly in the presence of other car- Cholesterol Lowering
diovascular risk factors. Patients below During the median follow-up of 2.2
the age of 40 have lower risk of devel- Statins and Ezetimibe years, the composite outcome of cardio-
oping a cardiovascular event over a vascular death, MI, stroke, hospitalization
10-year horizon; however, their lifetime The IMProved Reduction of Outcomes: for angina, or revascularization occurred
risk of developing cardiovascular disease Vytorin Efficacy International Trial in 11.3% vs. 9.8% of the placebo and evo-
and suffering an MI, stroke, or cardiovas- (IMPROVE-IT) was a randomized con- locumab groups, respectively, represent-
cular death is high. For patients under the trolled trial in 18,144 patients comparing ing a 15% relative risk reduction (P ,
age of 40 years and/or who have type 1 the addition of ezetimibe to simvastatin 0.001). The combined end point of cardio-
diabetes with other ASCVD risk factors, therapy versus simvastatin alone. Individuals vascular death, MI, or stroke was reduced
we recommend that the patient and were $50 years of age, had experienced a by 20%, from 7.4 to 5.9% (P , 0.001).
health care provider discuss the relative recent acute coronary syndrome (ACS), Importantly, similar benefits were seen
benefits and risks and consider the use and were treated for an average of in prespecified subgroup of patients
of moderate-intensity statin therapy. 6 years. Overall, the addition of ezetimibe with diabetes, comprising 11,031 patients
Please refer to “Type 1 Diabetes Mellitus led to a 6.4% relative benefit and a 2% ab- (40% of the trial) (73).
and Cardiovascular Disease: A Scientific solute reduction in major adverse cardiovas-
Statement From the American Heart cular events, with the degree of benefit Statins and CETP Inhibitors
being directly proportional to the change
in LDL cholesterol, which was 70 mg/dL in Inhibition of CETP increases HDL choles-
the statin group on average and 54 mg/dL in terol and further reduces LDL cholesterol.
the combination group (65). In those with

S94 Cardiovascular Disease and Risk Management Diabetes Care Volume 41, Supplement 1, January 2018

This class of drugs is not likely to be avail- dyslipidemia in individuals with type 2 di- (1.7–4.5 mmol/L) to statin therapy plus
able for clinical use, but studies pro- abetes. However, the evidence for the extended-release niacin or placebo. The
vide further insight into the effects of use of drugs that target these lipid frac- trial was halted early due to lack of effi-
LDL cholesterol lowering on cardiovascular tions is substantially less robust than that cacy on the primary ASCVD outcome (first
events. for statin therapy (78). In a large trial in event of the composite of death from
patients with diabetes, fenofibrate failed CHD, nonfatal MI, ischemic stroke, hospi-
A total of four trials have been con- to reduce overall cardiovascular out- talization for an ACS, or symptom-driven
ducted, three of which failed to show comes (79). coronary or cerebral revascularization)
benefit (74–76). Of these, one showed and a possible increase in ischemic stroke
harm and two were stopped after approx- Other Combination Therapy in those on combination therapy (82).
imately 2 years and thus did not have
sufficient time or power to identify the Recommendations The much larger Heart Protection
benefit. The final study, the Randomized Study 2–Treatment of HDL to Reduce
Evaluation of the Effects of Anacetrapib c Combination therapy (statin/fibrate) the Incidence of Vascular Events (HPS2-
Through Lipid-modification (REVEAL) trial has not been shown to improve ath- THRIVE) trial also failed to show a benefit
enrolled 30,449 patients with ASCVD (70). erosclerotic cardiovascular disease of adding niacin to background statin
All patients received intensive atorvasta- outcomes and is generally not rec- therapy (83). A total of 25,673 patients
tin therapy and were randomized to ana- ommended. A with prior vascular disease were random-
cetrapib or placebo. ized to receive 2 g of extended-release
c Combination therapy (statin/niacin) niacin and 40 mg of laropiprant (an antag-
During the median follow-up of 4.1 has not been shown to provide addi- onist of the prostaglandin D2 receptor
years, the primary outcome (coronary tional cardiovascular benefit above DP1 that has been shown to improve ad-
death, MI, or coronary revascularization) statin therapy alone, may increase herence to niacin therapy) versus a
was significantly reduced with the addi- the risk of stroke with additional matching placebo daily and followed
tion of anacetrapib from 11.8 to 10.8%, side effects, and is generally not for a median follow-up period of 3.9
with a hazard ratio (HR) of 0.91 (P 5 recommended. A years. There was no significant difference
0.004). The relative difference in risk in the rate of coronary death, MI, stroke,
was similar across multiple prespecified Statin and Fibrate or coronary revascularization with the ad-
subgroups, including among 11,320 pa- dition of niacin–laropiprant versus pla-
tients with diabetes (37% of the trial). Combination therapy (statin and fibrate) cebo (13.2% vs. 13.7%; rate ratio, 0.96;
The benefit appeared to be related to is associated with an increased risk for P 5 0.29). Niacin–laropiprant was associ-
the reduction in LDL (and more broadly abnormal transaminase levels, myositis, ated with an increased incidence of new-
non-HDL) as opposed to the raising of and rhabdomyolysis. The risk of rhabdo- onset diabetes (absolute excess, 1.3
HDL. The mean achieved LDL cholesterol myolysis is more common with higher percentage points; P , 0.001) and distur-
was 63 mg/dL vs. 53 mg/dL at the trial doses of statins and renal insufficiency bances in diabetes control among those
midpoint in the placebo and anacetrapib and appears to be higher when statins with diabetes. In addition, there was an
groups, respectively. This study reaffirms are combined with gemfibrozil (com- increase in serious adverse events associ-
the benefit of further lowering of LDL pared with fenofibrate) (80). ated with the gastrointestinal system,
cholesterol on reducing cardiovascular musculoskeletal system, skin, and, unex-
events. In the ACCORD study, in patients with pectedly, infection and bleeding.
type 2 diabetes who were at high risk for
Treatment of Other Lipoprotein ASCVD, the combination of fenofibrate Therefore, combination therapy with a
Fractions or Targets and simvastatin did not reduce the rate statin and niacin is not recommended
of fatal cardiovascular events, nonfatal given the lack of efficacy on major ASCVD
Recommendation MI, or nonfatal stroke as compared with outcomes and side effects.
simvastatin alone. Prespecified subgroup
c For patients with fasting triglyceride analyses suggested heterogeneity in Diabetes With Statin Use
levels $500 mg/dL (5.7 mmol/L), treatment effects with possible benefit Several studies have reported a modestly
evaluate for secondary causes of for men with both a triglyceride level increased risk of incident diabetes with
hypertriglyceridemia and consider $204 mg/dL (2.3 mmol/L) and an HDL statin use (84,85), which may be limited
medical therapy to reduce the risk cholesterol level #34 mg/dL (0.9 mmol/L) to those with diabetes risk factors. An
of pancreatitis. C (81). analysis of one of the initial studies
suggested that although statin use was
Hypertriglyceridemia should be ad- Statin and Niacin associated with diabetes risk, the cardio-
dressed with dietary and lifestyle changes vascular event rate reduction with statins
including abstinence from alcohol (77). The Atherothrombosis Intervention in far outweighed the risk of incident diabe-
Severe hypertriglyceridemia (.1,000 Metabolic Syndrome With Low HDL/High tes even for patients at highest risk for
mg/dL) may warrant pharmacologic ther- Triglycerides: Impact on Global Health diabetes (86). The absolute risk increase
apy (fibric acid derivatives and/or fish oil) Outcomes (AIM-HIGH) trial randomized was small (over 5 years of follow-up,
to reduce the risk of acute pancreatitis. over 3,000 patients (about one-third 1.2% of participants on placebo devel-
with diabetes) with established ASCVD, oped diabetes and 1.5% on rosuvastatin
Low levels of HDL cholesterol, often low LDL cholesterol levels (,180 mg/dL developed diabetes) (86). A meta-analysis
associated with elevated triglyceride [4.7 mmol/L]), low HDL cholesterol levels
levels, are the most prevalent pattern of (men ,40 mg/dL [1.0 mmol/L] and
women ,50 mg/dL [1.3 mmol/L]), and
triglyceride levels of 150–400 mg/dL

care.diabetesjournals.org Cardiovascular Disease and Risk Management S95

of 13 randomized statin trials with 91,140 Risk Reduction greater than the number of episodes of
participants showed an odds ratio of 1.09 Aspirin has been shown to be effective in bleeding induced, although these compli-
for a new diagnosis of diabetes, so that reducing cardiovascular morbidity and cations do not have equal effects on long-
(on average) treatment of 255 patients mortality in high-risk patients with previ- term health (94).
with statins for 4 years resulted in one ous MI or stroke (secondary prevention).
additional case of diabetes while simulta- Its net benefit in primary prevention Treatment Considerations
neously preventing 5.4 vascular events among patients with no previous cardio- In 2010, a position statement of the ADA,
among those 255 patients (85). vascular events is more controversial the American Heart Association, and the
both for patients with diabetes and for American College of Cardiology Foun-
Statins and Cognitive Function patients without diabetes (89,90). Previ- dation recommended that low-dose
A recent systematic review of the U.S. ous randomized controlled trials of aspi- (75–162 mg/day) aspirin for primary pre-
Food and Drug Administration’s (FDA’s) rin specifically in patients with diabetes vention is reasonable for adults with di-
postmarketing surveillance databases, failed to consistently show a significant abetes and no previous history of vascular
randomized controlled trials, and cohort, reduction in overall ASCVD end points, disease who are at increased ASCVD risk
case-control, and cross-sectional studies raising questions about the efficacy of as- and who are not at increased risk for
evaluating cognition in patients receiving pirin for primary prevention in people bleeding (95). This now out-of-date state-
statins found that published data do not re- with diabetes, although some sex differ- ment included sex-specific recommenda-
veal an adverse effect of statins on cognition ences were suggested (91–93). tions for use of aspirin therapy as primary
(87). In addition, no change in cognitive prevention in persons with diabetes (95).
function has been reported in studies with The Antithrombotic Trialists’ Collabora- However, since that time, multiple recent
the addition of ezetimibe (65) or PCSK9 tion published an individual patient–level well-conducted studies and meta-analyses
inhibitors (66,88) to statin therapy, includ- meta-analysis (89) of the six large trials of have reported a risk of heart disease and
ing among patients treated to very low LDL aspirin for primary prevention in the gen- stroke that is equivalent if not higher in
cholesterol levels. Therefore, a concern that eral population. These trials collectively women compared with men with diabe-
statins or other lipid-lowering agents might enrolled over 95,000 participants, includ- tes, including among nonelderly adults.
cause cognitive dysfunction or dementia is ing almost 4,000 with diabetes. Overall, Thus, current recommendations for using
not currently supported by evidence and they found that aspirin reduced the risk aspirin as primary prevention include both
should not deter their use in individuals of serious vascular events by 12% (RR men and women aged $50 years with
with diabetes at high risk for ASCVD (87). 0.88 [95% CI 0.82–0.94]). The largest re- diabetes and at least one additional major
duction was for nonfatal MI, with little risk factor (family history of premature
ANTIPLATELET AGENTS effect on CHD death (RR 0.95 [95% CI ASCVD, hypertension, dyslipidemia,
0.78–1.15]) or total stroke. There was smoking, or chronic kidney disease/
Recommendations some evidence of a difference in aspirin albuminuria) who are not at increased
effect by sex: aspirin significantly reduced risk of bleeding (e.g., older age, anemia,
c Use aspirin therapy (75–162 mg/day) ASCVD events in men but not in women. renal disease) (96–99). While risk calcu-
as a secondary prevention strategy Conversely, aspirin had no effect on lators such as those from the American
in those with diabetes and a history stroke in men but significantly reduced College of Cardiology/American Heart As-
of atherosclerotic cardiovascular stroke in women. However, there was sociation (my.americanheart.org) may
disease. A no heterogeneity of effect by sex in the be a useful tool to estimate 10-year
risk of serious vascular events (P 5 0.9). ASCVD risk, diabetes itself confers in-
c For patients with atherosclerotic creased risk for ASCVD. As a result, such
cardiovascular disease and docu- Sex differences in the effects of aspirin risk calculators have limited utility in help-
mented aspirin allergy, clopidogrel have not been observed in studies of sec- ing to assess the potential benefits of as-
(75 mg/day) should be used. B ondary prevention (89). In the six trials pirin therapy in individuals with diabetes.
examined by the Antithrombotic Trialists’ Noninvasive imaging techniques such as
c Dual antiplatelet therapy (with low- Collaboration, the effects of aspirin on coronary computed tomography angiog-
dose aspirin and a P2Y12 inhibitor) major vascular events were similar for pa- raphy may potentially help further tai-
is reasonable for a year after an acute tients with or without diabetes: RR 0.88 lor aspirin therapy, particularly in those
coronary syndrome A and may have (95% CI 0.67–1.15) and RR 0.87 (95% CI at low risk (100), but are not generally
benefits beyond this period. B 0.79–0.96), respectively. The CI was wider recommended. Sex differences in the
for those with diabetes because of antiplatelet effect of aspirin have been sug-
c Aspirin therapy (75–162 mg/day) smaller numbers. gested in the general population (101);
may be considered as a primary pre- however, further studies are needed to
vention strategy in those with type 1 Aspirin appears to have a modest ef- investigate the presence of such differen-
or type 2 diabetes who are at in- fect on ischemic vascular events, with the ces in individuals with diabetes.
creased cardiovascular risk. This absolute decrease in events depending
includes most men and women with on the underlying ASCVD risk. The main Aspirin Use in People <50 Years of Age
diabetes aged $50 years who have adverse effect is an increased risk of gas- Aspirin is not recommended for those at
at least one additional major risk trointestinal bleeding. The excess risk may low risk of ASCVD (such as men and women
factor (family history of premature be as high as 5 per 1,000 per year in real- aged ,50 years with diabetes with no
atherosclerotic cardiovascular dis- world settings. In adults with ASCVD
ease, hypertension, dyslipidemia, risk .1% per year, the number of ASCVD
smoking, or albuminuria) and are events prevented will be similar to or
not at increased risk of bleeding. C

S96 Cardiovascular Disease and Risk Management Diabetes Care Volume 41, Supplement 1, January 2018

other major ASCVD risk factors) as the low adding ticagrelor to aspirin significantly disease, after lifestyle management
benefit is likely to be outweighed by the reduces the risk of recurrent ischemic and metformin, the antihyperglyce-
risks of bleeding. Clinical judgment should events including cardiovascular and coro- mic agent canagliflozin may be con-
be used for those at intermediate risk nary heart disease death (108). More sidered to reduce major adverse
(younger patients with one or more risk studies are needed to investigate the cardiovascular events, based on
factors or older patients with no risk fac- longer-term benefits of these therapies drug-specific and patient factors
tors) until further research is available. after ACS among patients with diabetes. (see Table 8.1). C
Patients’ willingness to undergo long-
term aspirin therapy should also be con- CORONARY HEART DISEASE Cardiac Testing
sidered (102). Aspirin use in patients Candidates for advanced or invasive car-
aged ,21 years is generally contraindi- Recommendations diac testing include those with 1) typical
cated due to the associated risk of Reye or atypical cardiac symptoms and 2) an ab-
syndrome. Screening normal resting electrocardiogram (ECG).
c In asymptomatic patients, routine Exercise ECG testing without or with echo-
Aspirin Dosing cardiography may be used as the initial test.
Average daily dosages used in most clini- screening for coronary artery dis- In adults with diabetes $40 years of age,
cal trials involving patients with diabetes ease is not recommended as it measurement of coronary artery calcium
ranged from 50 mg to 650 mg but were does not improve outcomes as long is also reasonable for cardiovascular risk
mostly in the range of 100–325 mg/day. as atherosclerotic cardiovascular dis- assessment. Pharmacologic stress echo-
There is little evidence to support any ease risk factors are treated. A cardiography or nuclear imaging should
specific dose, but using the lowest possi- c Consider investigations for coronary be considered in individuals with diabetes
ble dose may help to reduce side effects artery disease in the presence of any in whom resting ECG abnormalities pre-
(103). In the U.S., the most common low- of the following: atypical cardiac clude exercise stress testing (e.g., left
dose tablet is 81 mg. Although platelets symptoms (e.g., unexplained dyspnea, bundle branch block or ST-T abnormali-
from patients with diabetes have altered chest discomfort); signs or symptoms ties). In addition, individuals who require
function, it is unclear what, if any, effect of associated vascular disease includ- stress testing and are unable to exercise
that finding has on the required dose of ing carotid bruits, transient ischemic should undergo pharmacologic stress
aspirin for cardioprotective effects in the attack, stroke, claudication, or periph- echocardiography or nuclear imaging.
patient with diabetes. Many alternate eral arterial disease; or electrocardio-
pathways for platelet activation exist gram abnormalities (e.g., Q waves). E Screening Asymptomatic Patients
that are independent of thromboxane The screening of asymptomatic patients
A2 and thus not sensitive to the effects Treatment with high ASCVD risk is not recommended
of aspirin (104). “Aspirin resistance” has c In patients with known atheroscle- (109), in part because these high-risk pa-
been described in patients with diabetes tients should already be receiving inten-
when measured by a variety of ex vivo rotic cardiovascular disease, con- sive medical therapydan approach that
and in vitro methods (platelet aggregom- sider ACE inhibitor or angiotensin provides similar benefit as invasive revas-
etry, measurement of thromboxane B2) receptor blocker therapy to reduce cularization (110,111). There is also some
(101), but other studies suggest no impair- the risk of cardiovascular events. B evidence that silent MI may reverse over
ment in aspirin response among patients c In patients with prior myocardial in- time, adding to the controversy concern-
with diabetes (105). A recent trial suggested farction, b-blockers should be con- ing aggressive screening strategies (112).
that more frequent dosing regimens of aspi- tinued for at least 2 years after the In prospective studies, coronary artery
rin may reduce platelet reactivity in individ- event. B calcium has been established as an in-
uals with diabetes (106); however, these c In patients with type 2 diabetes with dependent predictor of future ASCVD events
observations alone are insufficient to em- stable congestive heart failure, in patients with diabetes and is consistently
pirically recommend that higher doses of metformin may be used if estimated superior to both the UK Prospective Diabetes
aspirin be used in this group at this time. glomerular filtration rate remains Study (UKPDS) risk engine and the Framing-
It appears that 75–162 mg/day is optimal. .30 mL/min but should be avoided ham Risk Score in predicting risk in this
in unstable or hospitalized patients population (113–115). However, a random-
Indications for P2Y12 Use with congestive heart failure. B ized observational trial demonstrated no
A P2Y12 receptor antagonist in combina- c In patients with type 2 diabetes and clinical benefit to routine screening of
tion with aspirin should be used for at established atherosclerotic cardio- asymptomatic patients with type 2 dia-
least 1 year in patients following an vascular disease, antihyperglycemic betes and normal ECGs (116). Despite
ACS and may have benefits beyond this therapy should begin with lifestyle abnormal myocardial perfusion imaging
period. Evidence supports use of either management and metformin and in more than one in five patients, cardiac
ticagrelor or clopidogrel if no percutane- subsequently incorporate an agent outcomes were essentially equal (and
ous coronary intervention was performed proven to reduce major adverse car- very low) in screened versus unscreened
and clopidogrel, ticagrelor, or prasugrel diovascular events and cardiovascular patients. Accordingly, indiscriminate
if a percutaneous coronary intervention mortality (currently empagliflozin screening is not considered cost-effective.
was performed (107). In patients with di- and liraglutide), after considering Studies have found that a risk factor–
abetes and prior MI (1–3 years before), drug-specific and patient factors
(see Table 8.1). A
c In patients with type 2 diabetes and es-
tablished atherosclerotic cardiovascular

Table 9.4—CVOTs completed after issuance of the FDA 2008 guidance GLP-1 receptor agonists SGLT2 inhibitors care.diabetesjournals.org
DPP-4 inhibitors

SAVOR-TIMI 53 EXAMINE TECOS ELIXA LEADER SUSTAIN-6 EXSCEL EMPA-REG CANVAS CANVAS-R
(129) (145) (132) (140) (138) (139)* (141) OUTCOME (133) (135) (135)
(n 5 14,671) (n 5 6,068) (n 5 14,752)
(n 5 16,492) (n 5 5,380) (n 5 9,340) (n 5 3,297) (n 5 7,020) (n 5 4,330) (n 5 5,812)

Intervention Saxagliptin/ Alogliptin/ Sitagliptin/ Lixisenatide/ Liraglutide/ Semaglutide/ Exenatide QW/ Empagliflozin/ Canagliflozin/placebo
Main inclusion criteria placebo placebo placebo placebo placebo placebo placebo placebo
Type 2 Type 2 Type 2 Type 2 diabetes and preexisting
Type 2 diabetes Type 2 diabetes Type 2 Type 2 diabetes Type 2 diabetes CVD at $30 years of age or $2
and history of and ACS diabetes and diabetes and diabetes and diabetes and with or without and preexisting cardiovascular risk factors at $50
or multiple within 15–90 preexisting history of ACS preexisting preexisting preexisting CVD CVD with BMI
risk factors for days before (,180 days) CVD, kidney CVD, HF, or #45 kg/m2 and years of age
CVD randomization CVD disease, or HF CKD at $50 6.5–10.0
5.5–11.0 at $50 years of years of age or 62 eGFR $30
A1C inclusion criteria (%) $6.5 6.5–11.0 6.5–8.0 60.3 cardiovascular 12 mL/min/1.73 m2 7.0–10.5
Age (years)†† 65.1 61.0 65.4 9.3 age or risk at $60 3.2 63.3
Diabetes duration (years)†† 10.3 7.1 11.6 2.1 cardiovascular years of age 74 7.0–10.0 13.5
Median follow-up (years) 2.1 1.5 3.0 93 77 63.1
Statin use (%) 78 91 80 66 risk at $60 $7.0 5.7 2.1
Metformin use (%) 70 66 82 years of age 64.6 73.1/16.2 57% .10 75
Prior CVD/CHF (%) 78/13 74/18 100/22 13.9 8.0 3.1 77
Mean baseline A1C (%) 8.0 100/28 7.2 7.7 $7.0 2.1 77
Mean difference in A1C 8.0 64.3 73 20.53^ 74 65.6/14.4
20.3^ 20.3^ 20.3^ 73 8.2
between groups at 20.3^ 12.8 60/24 2010/2017 99/10
end of treatment (%) 2010/2013 2008/2015 2010/2015 8.7 3-point MACE 8.1
Year started/reported 3-point MACE 2009/2013 4-point MACE 4-point MACE 3.8
Primary outcome§ 3-point MACE 0.91 20.3^‡
0.98 1.02 72 (0.83–1.00)
(0.89–1.08) (0.89–1.17) Individual 2010/2015
3-point MACE Expanded 76 components 3-point MACE
of MACE (see
0.99 MACE 81/18 0.86
(0.89–1.10) below) (0.74–0.99)
1.00 8.7 4-point MACE
(0.90–1.11)
20.4^ ^20.7 or 21.0 † 0.89 20.58^
(0.78–1.01)
2010/2016 2013/2016 2009/2017

3-point MACE 3-point MACE 3-point MACE Progression to Cardiovascular Disease and Risk Management S97
albuminuria**

1.00 0.96 (95% 0.87 0.74 0.86 0.73
(0.89–1.12) UL #1.16) (0.78–0.97) (0.58–0.95) (0.75–0.97)§ (0.47–0.77)

Key secondary outcome§ Expanded MACE 4-point MACE Expanded Expanded All-cause and 40% reduction in
MACE MACE cardiovascular composite eGFR,

mortality (see renal replacement,
below) renal death
0.60
1.02 0.95 0.88 0.74 (0.47–0.77)
(0.94–1.11) (95% UL # 1.14) (0.81–0.96) (0.62–0.89)

Continued on p. S98

Table 9.4—Continued S98 Cardiovascular Disease and Risk Management

DPP-4 inhibitors GLP-1 receptor agonists SGLT2 inhibitors

SAVOR-TIMI 53 EXAMINE TECOS ELIXA LEADER SUSTAIN-6 EXSCEL EMPA-REG CANVAS CANVAS-R
(129) (145) (132) (140) (138) (139)* (141) OUTCOME (133) (135) (135)
(n 5 14,671) (n 5 6,068) (n 5 14,752)
(n 5 16,492) (n 5 5,380) (n 5 9,340) (n 5 3,297) (n 5 7,020) (n 5 4,330) (n 5 5,812)

Cardiovascular death§ 1.03 0.85 1.03 0.98 0.78 0.98 0.88 0.62 0.96 (0.77–1.18)¶
(0.87–1.22) (0.66–1.10) (0.89–1.19) (0.78–1.22) (0.66–0.93) (0.65–1.48) (0.76–1.02) (0.49–0.77) 0.87 (0.72–1.06)#
MI§
0.95 1.08 0.95 1.03 0.86 0.74 0.97 0.87 0.85 0.85
Stroke§ (0.80–1.12) (0.88–1.33) (0.81–1.11) (0.87–1.22) (0.73–1.00) (0.51–1.08) (0.85–1.10) (0.70–1.09) (0.65–1.11) (0.61–1.19)

HF hospitalization§ 1.11 0.91 0.97 1.12 0.86 0.61 0.85 1.18 0.97 0.82
(0.88–1.39) (0.55–1.50) (0.79–1.19) (0.79–1.58) (0.71–1.06) (0.38–0.99) (0.70–1.03) (0.89–1.56) (0.70–1.35) (0.57–1.18)
Unstable angina
hospitalization§ 1.27 1.19 1.00 0.96 0.87 1.11 0.94 0.65 0.77 HR 0.56
(1.07–1.51) (0.90–1.58) (0.83–1.20) (0.75–1.23) (0.73–1.05) (0.77–1.61) (0.78–1.13) (0.50–0.85) (0.55–1.08) (0.38–0.83)
All-cause mortality§
1.19 0.90 0.90 1.11 0.98 0.82 1.05 0.99 d
Worsening (0.89–1.60) (0.60–1.37) (0.70–1.16) (0.47–2.62) (0.76–1.26) (0.47–1.44) (0.94–1.18) (0.74–1.34)
nephropathy§| 0.87 (0.74–1.01)‡‡
1.11 0.88 1.01 0.94 0.85 1.05 0.86 0.68 0.90 (0.76–1.07)##
(0.96–1.27) (0.71–1.09) (0.90–1.14) (0.78–1.13) (0.74–0.97) (0.74–1.50) (0.77–0.97) (0.57–0.82) 0.60 (0.47–0.77)

1.08 d d d 0.78 0.64 d 0.61
(0.88–1.32) (0.67–0.92) (0.46–0.88) (0.53–0.70)

d, not assessed/reported; CANVAS-R, CANVAS-Renal; CHF, congestive heart failure; CVD, cardiovascular disease; eGFR, estimated glomerular filtration rate; MACE, major adverse cardiac event; UL, upper limit. Data from Diabetes Care Volume 41, Supplement 1, January 2018
this table was adapted from Cefalu et al. (146) in the January 2018 issue of Diabetes Care. *Powered to rule out an HR of 1.8; superiority hypothesis not prespecified. **On the basis of prespecified outcomes, the
renal outcomes are not viewed as statistically significant. ††Age was reported as means in all trials except EXAMINE, which reported medians; diabetes duration was reported as means in all but four trials,with SAVOR-
TIMI 58, EXAMINE, and EXSCEL reporting medians and EMPA-REG OUTCOME reporting as percentage of population with diabetes duration .10 years. †A1C change of 0.66% with 0.5 mg and 1.05% with 1 mg dose of
semaglutide. ‡A1C change of 0.30 in EMPA-REG OUTCOME is based on pooled results for both doses (i.e., 0.24% for 10 mg and 0.36% for 25 mg of empagliflozin). §Outcomes reported as HR (95% CI). |Worsening
nephropathy is defined as the new onset of UACR .300 mg/g creatinine or a doubling of the serum creatinine level and an estimated glomerular filtration rate of #45 mL/min/1.73 m2, the need for continuous
renal-replacement therapy, or death from renal disease in EMPA-REG OUTCOME, LEADER, and SUSTAIN-6 and as doubling of creatinine level, initiation of dialysis, renal transplantation, or creatinine .6.0 mg/dL (530 mmol/L)
in SAVOR-TIMI 53. Worsening nephropathy was a prespecified exploratory adjudicated outcome in SAVOR-TIMI 53, LEADER, and SUSTAIN-6 but not in EMPA-REG OUTCOME. ¶Truncated data set (prespecified

in treating hierarchy as the principal data set for analysis for superiority of all-cause mortality and cardiovascular death in the CANVAS Program).^Significant difference in A1C between groups (P , 0.05). #Nontruncated data

set. ‡‡Truncated integrated data set (refers to pooled data from CANVAS after 20 November 2012 plus CANVAS-R; prespecified in treating hierarchy as the principal data set for analysis for superiority of

all-cause mortality and cardiovascular death in the CANVAS Program). ##Nontruncated integrated data (refers to pooled data from CANVAS, including before 20 November 2012 plus CANVAS-R).

care.diabetesjournals.org Cardiovascular Disease and Risk Management S99

based approach to the initial diagnostic Recent studies have also examined the SGLT2 inhibitors (particularly the preven-
evaluation and subsequent follow-up for relationship between dipeptidyl pep- tion of heart failure), are being followed
coronary artery disease fails to identify tidase 4 (DPP-4) inhibitors and heart up with new outcomes trials in patients
which patients with type 2 diabetes will failure and have had mixed results. with established heart failure, both with
have silent ischemia on screening tests The Saxagliptin Assessment of Vascular and without diabetes, to determine their
(117,118). Any benefit of newer nonin- Outcomes Recorded in Patients with Di- efficacy in treatment of heart failure.
vasive coronary artery disease screening abetes Mellitus–Thrombolysis in Myocar-
methods, such as computed tomography dial Infarction 53 (SAVOR-TIMI 53) study Antihyperglycemic Therapies and
and computed tomography angiography, showed that patients treated with Cardiovascular Outcomes
to identify patient subgroups for different saxagliptin (a DPP-4 inhibitor) were more In 2008, the FDA issued a guidance for
treatment strategies remains unproven. likely to be hospitalized for heart failure industry to perform cardiovascular out-
Although asymptomatic patients with di- than were those given placebo (3.5% vs. comes trials for all new medications for
abetes with higher coronary disease bur- 2.8%, respectively) (129). Two other re- the treatment for type 2 diabetes amid
den have more future cardiac events cent multicenter, randomized, double- concerns of increased cardiovascular risk
(113,119,120), the role of these tests be- blind, noninferiority trials, Examination of (137). Previously approved diabetes med-
yond risk stratification is not clear. Their rou- Cardiovascular Outcomes with Alogliptin ications were not subject to the guidance.
tine use leads to radiation exposure and may versus Standard of Care (EXAMINE) and Recently published cardiovascular outcomes
result in unnecessary invasive testing such as Trial Evaluating Cardiovascular Outcomes trials have provided additional data on car-
coronary angiography and revascularization with Sitagliptin (TECOS), did not show asso- diovascular outcomes in patients with type 2
procedures. The ultimate balance of bene- ciations between DPP-4 inhibitor use and diabetes with cardiovascular disease or at
fit, cost, and risks of such an approach in heart failure. The FDA reported that the hos- high risk for cardiovascular disease (see Table
asymptomaticpatients remains controversial, pital admission rate for heart failure in 9.4). Cardiovascular outcomes trials of
particularly in the modern setting of aggres- EXAMINE was 3.9% for patients randomly DPP-4 inhibitors have all, so far, not shown
sive ASCVD risk factor control. assigned to alogliptin compared with cardiovascular benefits relative to placebo.
3.3% for those randomly assigned to However, results from other new agents
Lifestyle and Pharmacologic placebo (130). Alogliptin had no effect have provided a mix of results.
Interventions on the composite end point of cardiovas-
Intensive lifestyle intervention focusing cular death and hospital admission for EMPA-REG OUTCOME trial was a ran-
on weight loss through decreased caloric heart failure in the post hoc analysis (HR domized, double-blind trial that assessed
intake and increased physical activity as 1.00 [95% CI 0.82–1.21]) (131). TECOS the effect of empagliflozin, a SGLT2 inhib-
performed in the Action for Health in Di- showed no difference in the rate of heart itor, versus placebo on cardiovascular
abetes (Look AHEAD) trial may be con- failure hospitalization for the sitagliptin outcomes in 7,020 patients with type 2
sidered for improving glucose control, group (3.1%; 1.07 per 100 person-years) diabetes and existing cardiovascular dis-
fitness, and some ASCVD risk factors compared with the placebo group (3.1%; ease. Study participants had a mean age
(121). Patients at increased ASCVD risk 1.09 per 100 person-years) (132). of 63 years, 57% had diabetes for more
should receive aspirin and a statin and than 10 years, and 99% had established
ACE inhibitor or ARB therapy if the patient A benefit on the incidence of heart fail- cardiovascular disease.EMPA-REGOUTCOME
has hypertension, unless there are con- ure has been observed with the use of showed that over a median follow-up of
traindications to a particular drug class. some sodium–glucose cotransporter 3.1 years, treatment reduced the compos-
While clear benefit exists for ACE inhibitor 2 (SGLT2) inhibitors. In the BI 10773 ite outcome of MI, stroke, and cardiovas-
or ARB therapy in patients with diabetic (Empagliflozin) Cardiovascular Outcome cular death by 14% (absolute rate 10.5%
kidney disease or hypertension, the bene- Event Trial in Type 2 Diabetes Mellitus vs. 12.1% in the placebo group, HR in the
fits in patients with ASCVD in the absence Patients (EMPA-REG OUTCOME), the ad- empagliflozin group 0.86; 95% CI 0.74–
of these conditions are less clear, espe- dition of empagliflozin to standard care 0.99; P = 0.04 for superiority) and cardio-
cially when LDL cholesterol is concomi- led to a significant 35% reduction in the vascular death by 38% (absolute rate
tantly controlled (122,123). In patients hospitalization for heart failure compared 3.7% vs. 5.9%, HR 0.62; 95% CI 0.49–
with prior MI, active angina, or heart fail- with placebo (133). Although the majority 0.77; P , 0.001) (133). The FDA recently
ure, b-blockers should be used (124). of patients in the study did not have heart added a new indication for empagliflozin,
failure at baseline, this benefit was con- to reduce the risk of major adverse car-
Diabetes and Heart Failure sistent in patients with and without a diovascular death in adults with type 2
As many as 50% of patients with type 2 prior history of heart failure (134). Simi- diabetes and cardiovascular disease.
diabetes may develop heart failure (125). larly, in the Canagliflozin Cardiovascu-
Data on the effects of glucose-lowering lar Assessment Study (CANVAS), there A second large cardiovascular out-
agents on heart failure outcomes have was a 33% reduction in hospitalization comes trial program of an SGLT2 inhibi-
demonstrated that thiazolidinediones for heart failure with canagliflozin versus tor, canagliflozin, has been reported
have a strong and consistent relation- placebo (135). Although heart failure hos- (135). The CANVAS Program integrated
ship with increased risk of heart failure pitalizations were prospectively adjudicated data from two trials, including the CANVAS
(126–128). Therefore, thiazolidinedione in both trials, the type(s) of heart failure trial that started in 2009 before the ap-
use should be avoided in patients with events prevented were not characterized. proval of canagliflozin and the CANVAS-R
symptomatic heart failure. These preliminary findings, which strongly trial that started in 2014 after the approval
suggest heart failure–related benefits of of canagliflozin. Combining both these trials,
10,142 participants with type 2 diabetes and

S100 Cardiovascular Disease and Risk Management Diabetes Care Volume 41, Supplement 1, January 2018

high cardiovascular risk were randomized to stroke and cardiovascular death, in adults significant (141). A total of 14,752 pa-
canagliflozin or placebo and were followed with type 2 diabetes and established car- tients with type 2 diabetes (of whom
for an average 3.6 years. The mean age of diovascular disease. 10,782 [73.1%] had previous cardiovascu-
patients was 63 years and 66% had a history lar disease) were randomized to receive
of cardiovascular disease. The combined Results from a moderate-sized trial of an- extended-release exenatide 2 mg or pla-
analysis of the two trials found that other GLP-1 receptor agonist, semaglutide, cebo and followed for a median of 3.2
canagliflozin significantly reduced the com- were consistent with the LEADER trial years. The primary end point of cardio-
posite outcome of cardiovascular death, (139). Semaglutide, a once-weekly GLP-1 vascular death, MI, or stroke occurred
MI, or stroke versus placebo (occurring in receptor agonist, has not yet been ap- in 839 patients (11.4%; 3.7 events per
26.9 vs. 31.5 participants per 1,000 patient- proved by the FDA for the treatment of 100 person-years) in the exenatide group
years; HR 0.86 [95% CI 0.75–0.97]; type 2 diabetes. The preapproval Trial to and in 905 patients (12.2%; 4.0 events per
P , 0.001 for noninferiority; P 5 0.02 Evaluate Cardiovascular and Other Long- 100 person-years) in the placebo group
for superiority). The specific estimates term Outcomes with Semaglutide in Sub- (HR 0.91 [95% CI 0.83–1.00]; P , 0.001
for canagliflozin versus placebo on the jects With Type 2 Diabetes (SUSTAIN-6) for noninferiority) but was not superior to
primary composite cardiovascular out- was the initial randomized trial powered placebo with respect to the primary end
come were HR 0.88 (0.75–1.03) for the to test noninferiority of semaglutide for point (P 5 0.06 for superiority). However,
CANVAS trial, and 0.82 (0.66–1.01) for the purpose of initial regulatory approval. all-cause mortality was lower in the exe-
the CANVAS-R, with no heterogeneity In this study, 3,297 patients with type 2 di- natide group (HR 0.86 [95% CI 0.77–0.97].
found between trials. In the combined abetes were randomized to receive once- The incidence of acute pancreatitis, pan-
analysis, there was not a statistically sig- weekly semaglutide (0.5 mg or 1.0 mg) or creatic cancer, medullary thyroid carci-
nificant difference in cardiovascular death placebo for 2 years. The primary outcome noma, and serious adverse events did
(HR 0.87 [95% CI 0.72–1.06]). The initial (the first occurrence of cardiovascular not differ significantly between the two
CANVAS trial was partially unblinded prior death, nonfatal MI, or nonfatal stroke) groups.
to completion because of the need to file occurred in 108 patients (6.6%) in the
interim cardiovascular outcome data for semaglutide group vs. 146 patients (8.9%) In summary, there are now large
regulatory approval of the drug (136). Of in the placebo group (HR 0.74 [95% CI randomized controlled trials reporting
note, there was an increased risk of am- 0.58–0.95]; P , 0.001). More patients dis- statistically significant reductions in car-
putation with canaglifozin (6.3 vs. 3.4 par- continued treatment in the semaglutide diovascular events for two of the FDA-
ticipants per 1,000 patient-years; HR 1.97 group because of adverse events, mainly approved SGLT2 inhibitors (empagliflozin
[95% CI 1.41–2.75]) (135). gastrointestinal. and canagliflozin) and one of the FDA-
approved GLP-1 receptor agonists (liraglutide)
The Liraglutide Effect and Action in Di- The Evaluation of Lixisenatide in Acute where the majority, if not all, patients in
abetes: Evaluation of Cardiovascular Coronary Syndrome (ELIXA) trial studied the trial had ASCVD. The empagliflozin
Outcome ResultsdA Long Term Evalua- the once-daily GLP-1 receptor agonist and liraglutide trials further demon-
tion (LEADER) trial was a randomized, lixisenatide on cardiovascular outcomes strated significant reductions in cardio-
double-blind trial that assessed the effect in patients with type 2 diabetes who vascular death. Once-weekly exenatide
of liraglutide, a glucagon-like peptide had had a recent acute coronary event did not have statistically significant re-
1 (GLP-1) receptor agonist, versus placebo (140). A total of 6,068 patients with type 2 ductions in major adverse cardiovascu-
on cardiovascular outcomes in 9,340 pa- diabetes with a recent hospitalization for lar events or cardiovascular mortality
tients with type 2 diabetes at high risk for MI or unstable angina within the previ- but did have a significant reduction in
cardiovascular disease or with cardiovascu- ous 180 days were randomized to re- all-cause mortality. In contrast, other
lar disease. Study participants with a mean ceive lixisenatide or placebo in addition GLP-1 receptor agonists have not
age of 64 years and a mean duration of to standard care and were followed for shown similar reductions in cardiovas-
diabetes of nearly 13 years. Over 80% of a median of approximately 2.1 years. cular events (Table 9.4). Whether the
study participants had established cardio- The primary outcome of cardiovascular benefits of GLP-1 receptor agonists are a
vascular disease. After a median follow-up death, MI, stroke, or hospitalization for class effect remains to be definitively
of 3.8 years, LEADER showed that the pri- unstable angina occurred in 406 patients established. Additional large randomized
mary composite outcome (MI, stroke, or (13.4%) in the lixisenatide group vs. trials of other agents in these classes are
cardiovascular death) occurred in fewer 399 (13.2%) in the placebo group (HR ongoing.
participants in the treatment group 1.02 [95% CI 0.89–1.17]), which demon-
(13.0%) when compared with the placebo strated the noninferiority of lixisenatide Of note, these studies examined the
group (14.9%) (HR 0.87; 95% CI 0.78–0.97; to placebo (P , 0.001) but did not show drugs in combination with metformin (Ta-
P , 0.001 for noninferiority; P = 0.01 for superiority (P 5 0.81). ble 9.4) in the great majority of patients
superiority). Deaths from cardiovascular for whom metformin was not contraindi-
causes in the were significantly reduced Most recently, the Exenatide Study of cated or was tolerated. For patients with
in the liraglutide group (4.7%) compared Cardiovascular Event Lowering (EXSCEL) type 2 diabetes who have ASCVD, on life-
to the placebo group (6.0%) (HR 0.78; trial also reported results with the once- style and metformin therapy, it is recom-
95% CI 0.66–0.93; P = 0.007) (138). The weekly GLP-1 receptor agonist extended- mended to incorporate an agent with
FDA recently approved use of liraglutide release exenatide and found that major strong evidence for cardiovascular risk
to reduce the risk of major adverse car- adverse cardiovascular events were nu- reduction, especially those with proven
diovascular events, including heart attack, merically lower with use of extended- benefit on both major adverse cardiovas-
release exenatide compared with placebo, cular events and cardiovascular death,
although this difference was not statistically

care.diabetesjournals.org Cardiovascular Disease and Risk Management S101

after consideration of drug-specific pa- 15. Xie X, Atkins E, Lv J, et al. Effects of intensive for the treatment of hypertension in older adults
tient factors (Table 8.1). See Fig. 8.1 for blood pressure lowering on cardiovascular and with and without glucose disorders: a report from
additional recommendations on antihy- renal outcomes: updated systematic review and the ALLHAT study. J Clin Hypertens (Greenwich)
meta-analysis. Lancet 2016;387:435–443 2004;6:116–125
perglycemic treatment in adults with 16. Cushman WC, Evans GW, Byington RP, et al.; 29. Weber MA, Bakris GL, Jamerson K, et al.;
ACCORD Study Group. Effects of intensive blood- ACCOMPLISH Investigators. Cardiovascular events
type 2 diabetes. pressure control in type 2 diabetes mellitus. N during differing hypertension therapies in pa-
Engl J Med 2010;362:1575–1585 tients with diabetes. J Am Coll Cardiol 2010;56:
References 17. Patel A, MacMahon S, Chalmers J, et al.; 77–85
ADVANCE Collaborative Group. Effects of a fixed 30. Bangalore S, Fakheri R, Toklu B, Messerli FH.
1. Ali MK, Bullard KM, Saaddine JB, Cowie CC, combination of perindopril and indapamide on Diabetes mellitus as a compelling indication for
Imperatore G, Gregg EW. Achievement of goals macrovascular and microvascular outcomes in pa- use of renin angiotensin system blockers: system-
in U.S. diabetes care, 1999–2010. N Engl J Med tients with type 2 diabetes mellitus (the ADVANCE atic review and meta-analysis of randomized tri-
2013;368:1613–1624 trial): a randomised controlled trial. Lancet 2007; als. BMJ 2016;352:i438
2. Rawshani A, Rawshani A, Franze´n S, et al. Mor- 370:829–840
tality and cardiovascular disease in type 1 and type 18. Bakris GL. The implications of blood pressure 31. Carlberg B, Samuelsson O, Lindholm LH. Ate-
2 diabetes. N Engl J Med 2017;376:1407–1418 measurement methods on treatment targets for nolol in hypertension: is it a wise choice? Lancet
3. Gregg EW, Li Y, Wang J, et al. Changes in di- blood pressure. Circulation 2016;134:904–905 2004;364:1684–1689
abetes-related complications in the United States, 19. Association AD; American Diabetes Associa- 32. Yusuf S, Teo KK, Pogue J, et al.; ONTARGET
1990–2010. N Engl J Med 2014;370:1514–1523 tion. Glycemic targets. Sec. 6. In Standards of Investigators. Telmisartan, ramipril, or both in pa-
4. Ford ES, Ajani UA, Croft JB, et al. Explaining the Medical Care in Diabetesd2017. Diabetes Care tients at high risk for vascular events. N Engl J Med
decrease in U.S. deaths from coronary disease, 2017;40(Suppl. 1):S48–S56 2008;358:1547–1559
1980–2000. N Engl J Med 2007;356:2388–2398 20. Blood Pressure Lowering Treatment Trialists’ 33. Fried LF, Emanuele N, Zhang JH, et al.; VA
5. de Boer IH, Bangalore S, Benetos A, et al. Di- Collaboration. Blood pressure-lowering treat- NEPHRON-D Investigators. Combined angiotensin
abetes and hypertension: a position statement by ment based on cardiovascular risk: a meta-analysis inhibition for the treatment of diabetic nephrop-
the American Diabetes Association. Diabetes Care of individual patient data. Lancet 2014;384: athy. N Engl J Med 2013;369:1892–1903
2017;40:1273–1284 591–598 34. Makani H, Bangalore S, Desouza KA, Shah A,
6. Bobrie G, Gene` s N, Vaur L, et al. Is “isolated 20a. Whelton PK, Carey RM, Aronow WS, et al. Messerli FH. Efficacy and safety of dual blockade
home” hypertension as opposed to “isolated of- 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ of the renin-angiotensin system: meta-analysis of
fice” hypertension a sign of greater cardiovascular ASH/ASPC/NMA/PCNA guideline for the preven- randomised trials. BMJ 2013;346:f360
risk? Arch Intern Med 2001;161:2205–2211 tion, detection, evaluation, and management
7. Sega R, Facchetti R, Bombelli M, et al. Prognos- of high blood pressure in adults. J Am Coll Cardiol. 35. Zhao P, Xu P, Wan C, Wang Z. Evening versus
tic value of ambulatory and home blood pressures In press morning dosing regimen drug therapy for hyper-
compared with office blood pressure in the tension. Cochrane Database Syst Rev 2011 (10):
general population: follow-up results from the 21. Sacks FM, Svetkey LP, Vollmer WM, et al.; CD004184
Pressioni Arteriose Monitorate e Loro Associa- DASH-Sodium Collaborative Research Group. Ef- 36. Hermida RC, Ayala DE, Mojo´ n A, Ferna´ndez
zioni (PAMELA) study. Circulation 2005;111: fects on blood pressure of reduced dietary sodium JR. Influence of time of day of blood pressure-
1777–1783 and the Dietary Approaches to Stop Hypertension lowering treatment on cardiovascular risk in hy-
8. Omboni S, Gazzola T, Carabelli G, Parati G. (DASH) diet. N Engl J Med 2001;344:3–10 pertensive patients with type 2 diabetes. Diabetes
Clinical usefulness and cost effectiveness of 22. James PA, Oparil S, Carter BL, et al. 2014 Care 2011;34:1270–1276
home blood pressure telemonitoring: meta-analysis evidence-based guideline for the management of 37. Nilsson E, Gasparini A, A¨rnlo¨v J, et al. Inci-
of randomized controlled studies. J Hypertens 2013; high blood pressure in adults: report from the panel dence and determinants of hyperkalemia and
31:455–467; discussion 467–468 members appointed to the Eighth Joint National hypokalemia in a large healthcare system. Int J
9. Emdin CA, Rahimi K, Neal B, Callender T, Committee (JNC 8). JAMA 2014;311:507–520 Cardiol 2017;245:277–284
Perkovic V, Patel A. Blood pressure lowering in 23. Bakris GL, Weir MR; Study of Hypertension 38. Bandak G, Sang Y, Gasparini A, et al. Hyper-
type 2 diabetes: a systematic review and meta- and the Efficacy of Lotrel in Diabetes (SHIELD) kalemia after initiating renin-angiotensin system
analysis. JAMA 2015;313:603–615 Investigators. Achieving goal blood pressure in blockade: the Stockholm Creatinine Measure-
10. Arguedas JA, Leiva V, Wright JM. Blood pres- patients with type 2 diabetes: conventional versus ments (SCREAM) project. J Am Heart Assoc
sure targets for hypertension in people with di- fixed-dose combination approaches. J Clin Hyper- 2017;6:e005428
abetes mellitus. Cochrane Database Syst Rev tens (Greenwich) 2003;5:202–209 39. Hughes-Austin JM, Rifkin DE, Beben T, et al.
2013;10:CD008277 24. Feldman RD, Zou GY, Vandervoort MK, Wong The relation of serum potassium concentra-
11. Ettehad D, Emdin CA, Kiran A, et al. Blood CJ, Nelson SAE, Feagan BG. A simplified approach tion with cardiovascular events and mortality in
pressure lowering for prevention of cardiovascu- to the treatment of uncomplicated hypertension: community-living individuals. Clin J Am Soc Nephrol
lar disease and death: a systematic review and a cluster randomized, controlled trial. Hyperten- 2017;12:245–252
meta-analysis. Lancet 2016;387:957–967 sion 2009;53:646–653 40. James MT, Grams ME, Woodward M, et al.;
12. Brunstro¨m M, Carlberg B. Effect of antihy- 25. Bangalore S, Kamalakkannan G, Parkar S, CKD Prognosis Consortium. A meta-analysis of the
pertensive treatment at different blood pressure Messerli FH. Fixed-dose combinations improve association of estimated GFR, albuminuria, diabe-
levels in patients with diabetes mellitus: system- medication compliance: a meta-analysis. Am J tes mellitus, and hypertension with acute kidney
atic review and meta-analyses. BMJ 2016;352: Med 2007;120:713–719 injury. Am J Kidney Dis 2015;66:602–612
i717 26. Catala´-Lo´ pez F, Mac´ıas Saint-Gerons D, 41. Iliescu R, Lohmeier TE, Tudorancea I, Laffin L,
13. Bangalore S, Kumar S, Lobach I, Messerli FH. Gonza´lez-Bermejo D, et al. Cardiovascular and re- Bakris GL. Renal denervation for the treatment of
Blood pressure targets in subjects with type 2 nal outcomes of renin-angiotensin system block- resistant hypertension: review and clinical perspec-
diabetes mellitus/impaired fasting glucose: obser- ade in adult patients with diabetes mellitus: tive. Am J Physiol Renal Physiol 2015;309:F583–
vations from traditional and bayesian random- a systematic review with network meta-analyses. F594
effects meta-analyses of randomized trials. PLoS Med 2016;13:e1001971 42. Bakris GL, Agarwal R, Chan JC, et al.; Miner-
Circulation 2011;123:2799–2810 27. Palmer SC, Mavridis D, Navarese E, et al. alocorticoid Receptor Antagonist Tolerability
14. Thomopoulos C, Parati G, Zanchetti A. Effects Comparative efficacy and safety of blood pressure- Study–Diabetic Nephropathy (ARTS-DN) Study
of blood-pressure-lowering treatment on out- lowering agents in adults with diabetes and kidney Group. Effect of finerenone on albuminuria in pa-
come incidence in hypertension: 10 - should blood disease: a network meta-analysis. Lancet 2015;385: tients with diabetic nephropathy: a randomized
pressure management differ in hypertensive pa- 2047–2056 clinical trial. JAMA 2015;314:884–894
tients with and without diabetes mellitus? Over- 28. Barzilay JI, Davis BR, Bettencourt J, et al.; 43. Williams B, MacDonald TM, Morant S, et al.;
view and meta-analyses of randomized trials. J ALLHAT Collaborative Research Group. Cardiovas- British Hypertension Society’s PATHWAY Studies
Hypertens 2017;35:922–944 cular outcomes using doxazosin vs. chlorthalidone Group. Spironolactone versus placebo, bisoprolol,
and doxazosin to determine the optimal treatment

S102 Cardiovascular Disease and Risk Management Diabetes Care Volume 41, Supplement 1, January 2018

for drug-resistant hypertension (PATHWAY-2): to New Targets (TNT) study. Diabetes Care 2006; 71. Moriarty PM, Jacobson TA, Bruckert E, et al.
a randomised, double-blind, crossover trial. Lancet 29:1220–1226 Efficacy and safety of alirocumab, a monoclonal
2015;386:2059–2068 58. Sever PS, Poulter NR, Dahlo¨f B, et al. Reduc- antibody to PCSK9, in statin-intolerant patients:
44. Filippatos G, Anker SD, Bo¨hm M, et al. A tion in cardiovascular events with atorvastatin design and rationale of ODYSSEY ALTERNATIVE,
randomized controlled study of finerenone vs. in 2,532 patients with type 2 diabetes: Anglo- a randomized phase 3 trial. J Clin Lipidol 2014;8:
eplerenone in patients with worsening chronic Scandinavian Cardiac Outcomes Trial–lipid-lower- 554–561
heart failure and diabetes mellitus and/or chronic ing arm (ASCOT-LLA). Diabetes Care 2005;28: 72. Zhang X-L, Zhu Q-Q, Zhu L, et al. Safety and
kidney disease. Eur Heart J 2016;37:2105–2114 1151–1157 efficacy of anti-PCSK9 antibodies: a meta-analysis
45. Bomback AS, Klemmer PJ. Mineralocorticoid 59. Knopp RH, d’Emden M, Smilde JG, Pocock SJ. of 25 randomized, controlled trials. BMC Med
receptor blockade in chronic kidney disease. Efficacy and safety of atorvastatin in the preven- 2015;13:123
Blood Purif 2012;33:119–124 tion of cardiovascular end points in subjects with 73. Sabatine MS, Leiter LA, Wiviott SD, et al. Car-
46. American College of Obstetricians and Gyne- type 2 diabetes: the Atorvastatin Study for Pre- diovascular safety and efficacy of the PCSK9 in-
cologists; Task Force on Hypertension in Preg- vention of Coronary Heart Disease Endpoints in Non- hibitor evolocumab in patients with and without
nancy. Hypertension in pregnancy. Report of the Insulin-Dependent Diabetes Mellitus (ASPEN). diabetes and the effect of evolocumab on glycae-
American College of Obstetricians and Gynecolo- Diabetes Care 2006;29:1478–1485 mia and risk of new-onset diabetes: a prespecified
gists’ Task Force on Hypertension in Pregnancy. 60. Colhoun HM, Betteridge DJ, Durrington PN, analysis of the FOURIER randomised controlled
Obstet Gynecol 2013;122:1122–1131 et al.; CARDS investigators. Primary prevention of trial [article online]. Lancet Diabetes Endocrinol
47. Abalos E, Duley L, Steyn DW. Antihyperten- cardiovascular disease with atorvastatin in type 2 2017. Available from http://www.thelancet
sive drug therapy for mild to moderate hyperten- diabetes in the Collaborative Atorvastatin Diabe- .com/journals/landia/article/PIIS2213-8587(17)
sion during pregnancy. Cochrane Database Syst tes Study (CARDS): multicentre randomised 30313-3/abstract. Accessed 28 September 2017
Rev 2014;2:CD002252 placebo-controlled trial. Lancet 2004;364:685–696 74. Barter PJ, Caulfield M, Eriksson M, et al.;
48. Al-Balas M, Bozzo P, Einarson A. Use of diu- 61. Kearney PM, Blackwell L, Collins R, et al.; Cho- ILLUMINATE Investigators. Effects of torcetrapib
retics during pregnancy. Can Fam Physician 2009; lesterol Treatment Trialists’ (CTT) Collaborators. in patients at high risk for coronary events. N Engl
55:44–45 Efficacy of cholesterol-lowering therapy in J Med 2007;357:2109–2122
49. Irgens HU, Reisaeter L, Irgens LM, Lie RT. Long 18,686 people with diabetes in 14 randomised 75. Schwartz GG, Olsson AG, Abt M, et al.; dal-
term mortality of mothers and fathers after pre- trials of statins: a meta-analysis. Lancet 2008; OUTCOMES Investigators. Effects of dalcetrapib in
eclampsia: population based cohort study. BMJ 371:117–125 patients with a recent acute coronary syndrome.
2001;323:1213–1217 62. Taylor F, Huffman MD, Macedo AF, et al. Sta- N Engl J Med 2012;367:2089–2099
50. Chasman DI, Posada D, Subrahmanyan L, tins for the primary prevention of cardiovascular 76. Lincoff AM, Nicholls SJ, Riesmeyer JS, et al.;
Cook NR, Stanton VP Jr, Ridker PM. Pharmacoge- disease. Cochrane Database Syst Rev 2013 (1): ACCELERATE Investigators. Evacetrapib and car-
netic study of statin therapy and cholesterol re- CD004816 diovascular outcomes in high-risk vascular dis-
duction. JAMA 2004;291:2821–2827 63. Carter AA, Gomes T, Camacho X, Juurlink DN, ease. N Engl J Med 2017;376:1933–1942
51. Meek C, Wierzbicki AS, Jewkes C, et al. Daily Shah BR, Mamdani MM. Risk of incident diabetes 77. Berglund L, Brunzell JD, Goldberg AC, et al.;
and intermittent rosuvastatin 5 mg therapy in among patients treated with statins: population Endocrine society. Evaluation and treatment of
statin intolerant patients: an observational study. based study [published correction appears in BMJ hypertriglyceridemia: an Endocrine Society clini-
Curr Med Res Opin 2012;28:371–378 2013;347:f4356]. BMJ 2013;346:f2610 cal practice guideline. J Clin Endocrinol Metab
52. Mihaylova B, Emberson J, Blackwell L, et al.; 64. Cannon CP, Braunwald E, McCabe CH, et al.; 2012;97:2969–2989
Cholesterol Treatment Trialists’ (CTT) Collabora- Pravastatin or Atorvastatin Evaluation and In- 78. Singh IM, Shishehbor MH, Ansell BJ. High-
tors. The effects of lowering LDL cholesterol fection Therapy-Thrombolysis in Myocardial In- density lipoprotein as a therapeutic target: a sys-
with statin therapy in people at low risk of vascu- farction 22 Investigators. Intensive versus tematic review. JAMA 2007;298:786–798
lar disease: meta-analysis of individual data from moderate lipid lowering with statins after acute 79. Keech A, Simes RJ, Barter P, et al.; FIELD study
27 randomised trials. Lancet 2012;380:581–590 coronary syndromes. N Engl J Med 2004;350: investigators. Effects of long-term fenofibrate
53. Baigent C, Keech A, Kearney PM, et al.; Cho- 1495–1504 therapy on cardiovascular events in 9795 people
lesterol Treatment Trialists’ (CTT) Collaborators. 65. Cannon CP, Blazing MA, Giugliano RP, et al.; with type 2 diabetes mellitus (the FIELD study):
Efficacy and safety of cholesterol-lowering treat- IMPROVE-IT Investigators. Ezetimibe added to randomised controlled trial. Lancet 2005;366:
ment: prospective meta-analysis of data from statin therapy after acute coronary syndromes. 1849–1861
90,056 participants in 14 randomised trials of sta- N Engl J Med 2015;372:2387–2397 80. Jones PH, Davidson MH. Reporting rate of
tins. Lancet 2005;366:1267–1278 66. Sabatine MS, Giugliano RP, Keech AC, et al.; rhabdomyolysis with fenofibrate 1 statin versus
54. Pyo˘ra¨la¨ K, Pedersen TR, Kjekshus J, Faergeman O, FOURIER Steering Committee and Investigators. gemfibrozil 1 any statin. Am J Cardiol 2005;95:
Olsson AG, Thorgeirsson G. Cholesterol lowering Evolocumab and clinical outcomes in patients 120–122
with simvastatin improves prognosis of diabetic with cardiovascular disease. N Engl J Med 2017; 81. Ginsberg HN, Elam MB, Lovato LC, et al.;
patients with coronary heart disease. A subgroup 376:1713–1722 ACCORD Study Group. Effects of combination lipid
analysis of the Scandinavian Simvastatin Survival 67. Bohula EA, Morrow DA, Giugliano RP, et al. therapy in type 2 diabetes mellitus. N Engl J Med
Study (4S). Diabetes Care 1997;20:614–620 Atherothrombotic risk stratification and ezetimibe 2010;362:1563–1574
55. Collins R, Armitage J, Parish S, Sleigh P, Peto R; for secondary Prevention. J Am Coll Cardiol 2017; 82. Boden WE, Probstfield JL, Anderson T, et al.;
Heart Protection Study Collaborative Group. 69:911–921 AIM-HIGH Investigators. Niacin in patients with
MRC/BHF Heart Protection Study of cholesterol- 68. Bohula EA, Bonaca MP, Braunwald E, et al. low HDL cholesterol levels receiving intensive sta-
lowering with simvastatin in 5963 people with Atherothrombotic risk stratification and the effi- tin therapy. N Engl J Med 2011;365:2255–2267
diabetes: a randomised placebo-controlled trial. cacy and safety of vorapaxar in patients with 83. Landray MJ, Haynes R, Hopewell JC, et al.;
Lancet 2003;361:2005–2016 stable ischemic heart disease and previous HPS2-THRIVE Collaborative Group. Effects of
56. Goldberg RB, Mellies MJ, Sacks FM, et al.; The myocardial infarction. Circulation 2016;134: extended-release niacin with laropiprant in high-
Care Investigators. Cardiovascular events and 304–313 risk patients. N Engl J Med 2014;371:203–212
their reduction with pravastatin in diabetic 69. de Ferranti SD, de Boer IH, Fonseca V, et al. 84. Rajpathak SN, Kumbhani DJ, Crandall J,
and glucose-intolerant myocardial infarction Type 1 diabetes mellitus and cardiovascular dis- Barzilai N, Alderman M, Ridker PM. Statin therapy
survivors with average cholesterol levels: sub- ease: a scientific statement from the American and risk of developing type 2 diabetes: a meta-
group analyses in the Cholesterol And Recur- Heart Association and American Diabetes Associ- analysis. Diabetes Care 2009;32:1924–1929
rent Events (CARE) trial. Circulation 1998;98: ation. Circulation 2014;130:1110–1130 85. Sattar N, Preiss D, Murray HM, et al. Statins
2513–2519 70. Bowman L, Hopewell JC, Chen F, et al.; HPS3/ and risk of incident diabetes: a collaborative
57. Shepherd J, Barter P, Carmena R, et al. Effect TIMI55–REVEAL Collaborative Group. Effects meta-analysis of randomised statin trials. Lancet
of lowering LDL cholesterol substantially below of anacetrapib in patients with atherosclerotic 2010;375:735–742
currently recommended levels in patients with vascular disease. N Engl J Med 2017;377:1217– 86. Ridker PM, Pradhan A, MacFadyen JG, Libby
coronary heart disease and diabetes: the Treating 1227 P, Glynn RJ. Cardiovascular benefits and diabetes

care.diabetesjournals.org Cardiovascular Disease and Risk Management S103

risks of statin therapy in primary prevention: an 99. Peters SAE, Huxley RR, Woodward M. Diabe- diabetes: the PREDICT study. Eur Heart J 2008;
analysis from the JUPITER trial. Lancet 2012;380: tes as a risk factor for stroke in women com- 29:2244–2251
565–571 pared with men: a systematic review and 114. Raggi P, Shaw LJ, Berman DS, Callister TQ.
87. Richardson K, Schoen M, French B, et al. Sta- meta-analysis of 64 cohorts, including 775,385 Prognostic value of coronary artery calcium
tins and cognitive function: a systematic review. individuals and 12,539 strokes. Lancet 2014;383: screening in subjects with and without diabetes.
Ann Intern Med 2013;159:688–697 1973–1980 J Am Coll Cardiol 2004;43:1663–1669
88. Giugliano RP, Mach F, Zavitz K, et al.; 100. Dimitriu-Leen AC, Scholte AJHA, van 115. Anand DV, Lim E, Hopkins D, et al. Risk strat-
EBBINGHAUS Investigators. Cognitive function Rosendael AR, et al. Value of coronary computed ification in uncomplicated type 2 diabetes: pro-
in a randomized trial of evolocumab. N Engl J tomography angiography in tailoring aspirin ther- spective evaluation of the combined use of
Med 2017;377:633–643 apy for primary prevention of atherosclerotic coronary artery calcium imaging and selective
89. Baigent C, Blackwell L, Collins R, et al.; Antith- events in patients at high risk with diabetes melli- myocardial perfusion scintigraphy. Eur Heart J
rombotic Trialists’ (ATT) Collaboration. Aspirin in tus. Am J Cardiol 2016;117:887–893 2006;27:713–721
the primary and secondary prevention of vascular 101. Larsen SB, Grove EL, Neergaard-Petersen S, 116. Young LH, Wackers FJT, Chyun DA, et al.;
disease: collaborative meta-analysis of individual Wu¨rtz M, Hvas A-M, Kristensen SD. Determinants DIAD Investigators. Cardiac outcomes after
participant data from randomised trials. Lancet of reduced antiplatelet effect of aspirin in patients screening for asymptomatic coronary artery dis-
2009;373:1849–1860 with stable coronary artery disease. PLoS One ease in patients with type 2 diabetes: the DIAD
90. Perk J, De Backer G, Gohlke H, et al.; European 2015;10:e0126767 study: a randomized controlled trial. JAMA 2009;
Association for Cardiovascular Prevention & Re- 102. Mora S, Ames JM, Manson JE. Low-dose as- 301:1547–1555
habilitation (EACPR); ESC Committee for Practice pirin in the primary prevention of cardiovascular 117. Wackers FJT, Young LH, Inzucchi SE, et al.;
Guidelines (CPG). European Guidelines on cardio- disease: shared decision making in clinical prac- Detection of Ischemia in Asymptomatic Dia-
vascular disease prevention in clinical practice tice. JAMA 2016;316:709–710 betics Investigators. Detection of silent myocar-
(version 2012). The Fifth Joint Task Force of the 103. Campbell CL, Smyth S, Montalescot G, dial ischemia in asymptomatic diabetic subjects:
European Society of Cardiology and Other Socie- Steinhubl SR. Aspirin dose for the prevention of the DIAD study. Diabetes Care 2004;27:1954–
ties on Cardiovascular Disease Prevention in Clin- cardiovascular disease: a systematic review. JAMA 1961
ical Practice (constituted by representatives of 2007;297:2018–2024 118. Scognamiglio R, Negut C, Ramondo A,
nine societies and by invited experts). Eur Heart 104. Dav`ı G, Patrono C. Platelet activation and Tiengo A, Avogaro A. Detection of coronary ar-
J 2012;33:1635–1701 atherothrombosis. N Engl J Med 2007;357: tery disease in asymptomatic patients with
91. Belch J, MacCuish A, Campbell I, et al. The 2482–2494 type 2 diabetes mellitus. J Am Coll Cardiol
prevention of progression of arterial disease and 105. Zaccardi F, Rizzi A, Petrucci G, et al. In vivo 2006;47:65–71
diabetes (POPADAD) trial: factorial randomised platelet activation and aspirin responsiveness in 119. Hadamitzky M, Hein F, Meyer T, et al. Prog-
placebo controlled trial of aspirin and antioxidants type 1 diabetes. Diabetes 2016;65:503–509 nostic value of coronary computed tomographic
in patients with diabetes and asymptomatic pe- 106. Bethel MA, Harrison P, Sourij H, et al. Ran- angiography in diabetic patients without known
ripheral arterial disease. BMJ 2008;337:a1840 domized controlled trial comparing impact on coronary artery disease. Diabetes Care 2010;33:
92. Zhang C, Sun A, Zhang P, et al. Aspirin for platelet reactivity of twice-daily with once-daily 1358–1363
primary prevention of cardiovascular events in aspirin in people with type 2 diabetes. Diabet 120. Choi E-K, Chun EJ, Choi S-I, et al. Assessment
patients with diabetes: A meta-analysis. Diabetes Med 2016;33:224–230 of subclinical coronary atherosclerosis in
Res Clin Pract 2010;87:211–218 107. Vandvik PO, Lincoff AM, Gore JM, et al. Pri- asymptomatic patients with type 2 diabetes
93. De Berardis G, Sacco M, Strippoli GFM, et al. mary and secondary prevention of cardiovascular mellitus with single photon emission computed
Aspirin for primary prevention of cardiovascular disease: Antithrombotic Therapy and Prevention tomography and coronary computed tomogra-
events in people with diabetes: meta-analysis of of Thrombosis, 9th ed: American College of phy angiography. Am J Cardiol 2009;104:890–
randomised controlled trials. BMJ 2009;339: Chest Physicians Evidence-Based Clinical Practice 896
b4531 Guidelines [published correction appears in Chest 121. Wing RR, Bolin P, Brancati FL, et al.; Look
94. Pignone M, Earnshaw S, Tice JA, Pletcher MJ. 2012;141:1129]. Chest 2012;141(Suppl.):e637S– AHEAD Research Group. Cardiovascular effects
Aspirin, statins, or both drugs for the primary pre- e668S of intensive lifestyle intervention in type 2 diabe-
vention of coronary heart disease events in men: 108. Bhatt DL, Bonaca MP, Bansilal S, et al. Re- tes. N Engl J Med 2013;369:145–154
a cost-utility analysis. Ann Intern Med 2006;144: duction in ischemic events with ticagrelor in di- 122. Braunwald E, Domanski MJ, Fowler SE, et al.;
326–336 abetic patients with prior myocardial infarction in PEACE Trial Investigators. Angiotensin-converting-
95. Pignone M, Alberts MJ, Colwell JA, et al.; PEGASUS-TIMI 54. J Am Coll Cardiol 2016;67: enzyme inhibition in stable coronary artery dis-
American Diabetes Association; American Heart 2732–2740 ease. N Engl J Med 2004;351:2058–2068
Association; American College of Cardiology 109. Bax JJ, Young LH, Frye RL, Bonow RO, 123. Telmisartan Randomised AssessmeNt Study
Foundation. Aspirin for primary prevention of car- Steinberg HO, Barrett EJ; ADA. Screening for cor- in ACE iNtolerant subjects with cardiovascular Dis-
diovascular events in people with diabetes: a po- onary artery disease in patients with diabetes. Di- ease (TRANSCEND) Investigators, Yusuf S, Teo K,
sition statement of the American Diabetes abetes Care 2007;30:2729–2736 et al. Effects of the angiotensin-receptor blocker
Association, a scientific statement of the Ameri- 110. Boden WE, O’Rourke RA, Teo KK, et al.; telmisartan on cardiovascular events in high-risk
can Heart Association, and an expert consensus COURAGE Trial Research Group. Optimal medical patients intolerant to angiotensin-converting en-
document of the American College of Cardiology therapy with or without PCI for stable coronary zyme inhibitors: a randomised controlled trial.
Foundation. Diabetes Care 2010;33:1395–1402 disease. N Engl J Med 2007;356:1503–1516 Lancet 2008;372:1174–1183
96. Huxley RR, Peters SAE, Mishra GD, WoodwardM. 111. BARI 2D Study Group, Frye RL, August P, 124. Kezerashvili A, Marzo K, De Leon J. Beta
Risk of all-cause mortality and vascular events in et al. A randomized trial of therapies for type 2 blocker use after acute myocardial infarction in
women versus men with type 1 diabetes: a sys- diabetes and coronary artery disease. N Engl J the patient with normal systolic function: when
tematic review and meta-analysis. Lancet Diabe- Med 2009;360:2503–2015 is it “ok” to discontinue? Curr Cardiol Rev 2012;8:
tes Endocrinol 2015;3:198–206 112. Wackers FJT, Chyun DA, Young LH, et al.; 77–84
97. Peters SAE, Huxley RR, Woodward M. Diabe- Detection of Ischemia in Asymptomatic Diabetics 125. Kannel WB, Hjortland M, Castelli WP. Role of
tes as risk factor for incident coronary heart dis- (DIAD) Investigators. Resolution of asymptomatic diabetes in congestive heart failure: the Framing-
ease in women compared with men: a systematic myocardial ischemia in patients with type 2 dia- ham study. Am J Cardiol 1974;34:29–34
review and meta-analysis of 64 cohorts including betes in the Detection of Ischemia in Asymptom- 126. Dormandy JA, Charbonnel B, Eckland DJA,
858,507 individuals and 28,203 coronary events. atic Diabetics (DIAD) study. Diabetes Care 2007; et al.; PROactive Investigators. Secondary prevention
Diabetologia 2014;57:1542–1551 30:2892–2898 of macrovascular events in patients with type 2
98. Kalyani RR, Lazo M, Ouyang P, et al. Sex differ- 113. Elkeles RS, Godsland IF, Feher MD, et al.; diabetes in the PROactive Study (PROspective
ences in diabetes and risk of incident coronary PREDICT Study Group. Coronary calcium mea- pioglitAzone Clinical Trial In macroVascular Events):
artery disease in healthy young and middle-aged surement improves prediction of cardiovascular a randomised controlled trial. Lancet 2005;366:
adults. Diabetes Care 2014;37:830–838 events in asymptomatic patients with type 2 1279–1289

S104 Cardiovascular Disease and Risk Management Diabetes Care Volume 41, Supplement 1, January 2018

127. Singh S, Loke YK, Furberg CD. Long-term risk 134. Fitchett D, Butler J, van de Borne P, et al. outcomes in patients with type 2 diabetes. N
of cardiovascular events with rosiglitazone: Effects of empagliflozin on risk for cardiovascular Engl J Med 2016;375:1834–1844
a meta-analysis. JAMA 2007;298:1189–1195 death and heart failure hospitalization across the 140. Pfeffer MA, Claggett B, Diaz R, et al.; ELIXA
128. Lincoff AM, Wolski K, Nicholls SJ, Nissen SE. spectrum of heart failure risk in the EMPA-REG Investigators. Lixisenatide in patients with type 2
Pioglitazone and risk of cardiovascular events in OUTCOME trial [article online]. Eur Heart J 2017. diabetes and acute coronary syndrome. N Engl J
patients with type 2 diabetes mellitus: a meta- Available from https://academic.oup.com/ Med 2015;373:2247–2257
analysis of randomized trials. JAMA 2007;298: eurheartj/article/doi/10.1093/eurheartj/ehx511/ 141. Holman RR, Bethel MA, Mentz RJ, et al.;
1180–1188 4096345/Effects-of-empagliflozin-on-risk-for. Ac- EXSCEL Study Group. Effects of once-weekly exe-
129. Scirica BM, Bhatt DL, Braunwald E, et al.; cessed 29 September 2017 natide on cardiovascular outcomes in type 2 di-
SAVOR-TIMI 53 Steering Committee and Investi- 135. Neal B, Perkovic V, Mahaffey KW, et al.; abetes. N Engl J Med 2017;377:1228–1239
gators. Saxagliptin and cardiovascular outcomes CANVAS Program Collaborative Group. Cana- 142. Zoungas S, Chalmers J, Neal B, et al.;
in patients with type 2 diabetes mellitus. N Engl J gliflozin and cardiovascular and renal events ADVANCE-ON Collaborative Group. Follow-up
Med 2013;369:1317–1326 in type 2 diabetes. N Engl J Med 2017;377:644– of blood-pressure lowering and glucose control
130. U.S. Food and Drug Administration. Diabe- 657 in type 2 diabetes. N Engl J Med 2014;371:
tes medications containing saxagliptin and alog- 136. Neal B, Perkovic V, Matthews DR, et al. Ra- 1392–1406
liptin: drug safety communication - risk of heart tionale, design and baseline characteristics of the 143. Hansson L, Zanchetti A, Carruthers SG, et al.;
failure [Internet]. Available from https://www CANagliflozin cardioVascular Assessment Study- HOT Study Group. Effects of intensive blood-
.fda.gov/safety/medwatch/safetyinformation/ Renal (CANVAS-R): a randomized, placebo controlled pressure lowering and low-dose aspirin in pa-
safetyalertsforhumanmedicalproducts/ucm494252 trial. Diabetes Obes Metab 2017;19:387–393 tients with hypertension: principal results of the
.htm. Accessed 13 October 2017 137. U.S. Food and Drug Administration. Guid- Hypertension Optimal Treatment (HOT) rando-
131. Zannad F, Cannon CP, Cushman WC, et al.; ance for industry diabetes mellitus: evaluating mised trial. Lancet 1998;351:1755–1762
EXAMINE Investigators. Heart failure and mortal- cardiovascular risk in new antidiabetic therapies 144. Wright JT Jr, Williamson JD, Whelton PK,
ity outcomes in patients with type 2 diabetes tak- to treat type 2 diabetes. Silver Spring, MD, 2008. et al.; SPRINT Research Group. A randomized trial
ing alogliptin versus placebo in EXAMINE: Available from http://www.fda.gov/downloads/ of intensive versus standard blood-pressure con-
a multicentre, randomised, double-blind trial. Drugs/GuidanceComplianceRegulatoryInformation/ trol. N Engl J Med 2015;373:2103–2116
Lancet 2015;385:2067–2076 Guidances/ucm071627.pdf). Accessed 3 November 145. White WB, Cannon CP, Heller SR, et al.;
132. Green JB, Bethel MA, Armstrong PW, et al.; 2017 EXAMINE Investigators. Alogliptin after acute cor-
TECOS Study Group. Effect of sitagliptin on cardio- 138. Marso SP, Daniels GH, Brown-Frandsen K, onary syndrome in patients with type 2 diabetes.
vascular outcomes in type 2 diabetes. N Engl J et al.; LEADER Steering Committee; LEADER Trial N Engl J Med 2013;369:1327–1335
Med 2015;373:232–242 Investigators. Liraglutide and cardiovascular out- 146. Cefalu WT, Kaul S, Gerstein HC, et al. Cardio-
133. Zinman B, Wanner C, Lachin JM, et al.; comes in type 2 diabetes. N Engl J Med 2016;375: vascular outcomes trials in type 2 diabetes: where
EMPA-REG OUTCOME Investigators. Empagliflozin, 311–322 do we go from here? Reflections from a Diabetes
cardiovascular outcomes, and mortality in type 2 139. Marso SP, Bain SC, Consoli A, et al.; SUSTAIN- Care Editors’ Expert Forum. Diabetes Care. In
diabetes. N Engl J Med 2015;373:2117–2128 6 Investigators. Semaglutide and cardiovascular press

Diabetes Care Volume 41, Supplement 1, January 2018 S105

10. Microvascular Complications American Diabetes Association
and Foot Care: Standards of
Medical Care in Diabetesd2018

Diabetes Care 2018;41(Suppl. 1):S105–S118 | https://doi.org/10.2337/dc18-S010

The American Diabetes Association (ADA) “Standards of Medical Care in Diabetes” 10. MICROVASCULAR COMPLICATIONS AND FOOT CARE
includes ADA’s current clinical practice recommendations and is intended to provide
the components of diabetes care, general treatment goals and guidelines, and tools
to evaluate quality of care. Members of the ADA Professional Practice Committee, a
multidisciplinary expert committee, are responsible for updating the Standards of
Care annually, or more frequently as warranted. For a detailed description of ADA
standards, statements, and reports, as well as the evidence-grading system for ADA’s
clinical practice recommendations, please refer to the Standards of Care Introduc-
tion. Readers who wish to comment on the Standards of Care are invited to do so at
professional.diabetes.org/SOC.

DIABETIC KIDNEY DISEASE Suggested citation: American Diabetes Association.
10. Microvascular complications and foot care:
Recommendations Standards of Medical Care in Diabetesd2018.
Diabetes Care 2018;41(Suppl. 1):S105–S118
Screening
c At least once a year, assess urinary albumin (e.g., spot urinary albumin–to–creatinine © 2017 by the American Diabetes Association.
Readers may use this article as long as the work
ratio) and estimated glomerular filtration rate in patients with type 1 diabetes is properly cited, the use is educational and not
with duration of $5 years, in all patients with type 2 diabetes, and in all patients for profit, and the work is not altered. More infor-
with comorbid hypertension. B mation is available at http://www.diabetesjournals
.org/content/license.
Treatment
c Optimize glucose control to reduce the risk or slow the progression of diabetic

kidney disease. A
c Optimize blood pressure control to reduce the risk or slow the progression of

diabetic kidney disease. A
c For people with nondialysis-dependent diabetic kidney disease, dietary protein

intake should be approximately 0.8 g/kg body weight per day (the recommended
daily allowance). For patients on dialysis, higher levels of dietary protein intake
should be considered. B
c In nonpregnant patients with diabetes and hypertension, either an ACE
inhibitor or an angiotensin receptor blocker is recommended for those
with modestly elevated urinary albumin–to–creatinine ratio (30–299 mg/g
creatinine) B and is strongly recommended for those with urinary albumin–to–
creatinine ratio $300 mg/g creatinine and/or estimated glomerular filtration
rate ,60 mL/min/1.73 m2. A
c Periodically monitor serum creatinine and potassium levels for the development
of increased creatinine or changes in potassium when ACE inhibitors, angiotensin
receptor blockers, or diuretics are used. B

S106 Microvascular Complications and Foot Care Diabetes Care Volume 41, Supplement 1, January 2018

c Continued monitoring of urinary collections are more burdensome and over time as the prevalence of diabetes in-
albumin–to–creatinine ratio in pa- add little to prediction or accuracy. Mea- creases in the U.S. (3,4,11,12).
tients with albuminuria treated with surement of a spot urine sample for albumin
an ACE inhibitor or an angiotensin re- alone (whether by immunoassay or by using An active urinary sediment (containing
ceptor blocker is reasonable to assess a sensitive dipstick test specific for albu- red or white blood cells or cellular casts),
the response to treatment and pro- minuria) without simultaneously measur- rapidly increasing albuminuria or nephrotic
gression of diabetic kidney disease. E ing urine creatinine (Cr) is less expensive but syndrome, rapidly decreasing eGFR, or the
susceptible to false-negative and false- absence of retinopathy (in type 1 diabe-
c An ACE inhibitor or an angiotensin positive determinations as a result of varia- tes) may suggest alternative or additional
receptor blocker is not recom- tion in urine concentration due to hydration. causes of kidney disease. For patients
mended for the primary prevention with these features, referral to a nephrol-
of diabetic kidney disease in pa- Normal UACR is generally defined ogist for further diagnosis, including the
tients with diabetes who have nor- as ,30 mg/g Cr, and increased urinary possibility of kidney biopsy, should be
mal blood pressure, normal urinary albumin excretion is defined as $30 considered. It is rare for patients with
albumin–to–creatinine ratio (,30 mg/g Cr. However, UACR is a continuous type 1 diabetes to develop kidney disease
mg/g creatinine), and normal esti- measurement, and differences within the without retinopathy. In type 2 diabetes,
mated glomerular filtration rate. B normal and abnormal ranges are associ- retinopathy is only moderately sensitive
ated with renal and cardiovascular out- and specific for CKD caused by diabetes,
c When estimated glomerular filtration comes (7–9). Furthermore, because of as confirmed by kidney biopsy (13).
rate is ,60 mL/min/1.73 m2, evalu- biological variability in urinary albumin
ate and manage potential complica- excretion, two of three specimens of Stage 1–2 CKD has been defined by
tions of chronic kidney disease. E UACR collected within a 3- to 6-month evidence of kidney damage (usually albu-
period should be abnormal before con- minuria) with eGFR $60 mL/min/1.73 m2,
c Patients should be referred for sidering a patient to have albuminuria. while stages 3–5 CKD have been de-
evaluation for renal replacement Exercise within 24 h, infection, fever, fined by progressively lower ranges of
treatment if they have an estimated congestive heart failure, marked hyper- eGFR (14) (Table 10.1). More recently,
glomerular filtration rate ,30 glycemia, menstruation, and marked Kidney Disease: Improving Global Out-
mL/min/1.73 m2. A hypertension may elevate UACR inde- comes (KDIGO) recommended a more
pendently of kidney damage. comprehensive CKD staging that incor-
c Promptly refer to a physician expe- porates albuminuria and is more closely
rienced in the care of kidney disease eGFR should be calculated from serum associated with risks of cardiovascular
for uncertainty about the etiology of Cr using a validated formula. The Chronic disease (CVD) and CKD progression (2).
kidney disease, difficult management Kidney Disease Epidemiology Collabora- It has not been determined whether ap-
issues, and rapidly progressing kidney tion (CKD-EPI) equation is generally pre- plication of the more complex system aids
disease. B ferred (2). eGFR is routinely reported by clinical care or improves health outcomes.
laboratories with serum Cr, and eGFR cal-
Epidemiology of Diabetic Kidney culators are available from http://www Acute Kidney Injury
Disease .nkdep.nih.gov. An eGFR ,60 mL/min/ Acute kidney injury (AKI) is usually diag-
Chronic kidney disease (CKD) is diagnosed 1.73 m2 is generally considered abnormal, nosed by a rapid increase in serum Cr,
by the persistent presence of elevated though optimal thresholds for clinical di- which is also reflected as a rapid decrease
urinary albumin excretion (albuminuria), agnosis are debated (10). in eGFR, over a relatively short period of
low estimated glomerular filtration rate time. People with diabetes are at higher
(eGFR), or other manifestations of kidney Urinary albumin excretion and eGFR risk of AKI than those without diabetes
damage (1,2). Diabetic kidney disease, or each vary within people over time, and (15). Other risk factors for AKI include
CKD attributed to diabetes, occurs in 20– abnormal results should be confirmed to preexisting CKD, the use of medications
40% of patients with diabetes (1,3–5). Di- stage CKD (1,2). that cause kidney injury (e.g., nonsteroi-
abetic kidney disease typically develops dal anti-inflammatory drugs), and the use
after diabetes duration of 10 years in Diagnosis of Diabetic Kidney Disease of medications that alter renal blood flow
type 1 diabetes, but may be present at Diabetic kidney disease is usually a clinical and intrarenal hemodynamics. In particu-
diagnosis of type 2 diabetes. Diabetic kid- diagnosis made based on the presence of lar, many antihypertensive medications
ney disease can progress to end-stage re- albuminuria and/or reduced eGFR in the (e.g., diuretics, ACE inhibitors, and angio-
nal disease (ESRD) requiring dialysis or absence of signs or symptoms of other tensin receptor blockers [ARBs]) can re-
kidney transplantation and is the leading primary causes of kidney damage. The typ- duce intravascular volume, renal blood
cause of ESRD in the United States (6). In ical presentation of diabetic kidney disease flow, and/or glomerular filtration. There
addition, among people with type 1 or is considered to include a long-standing is a concern that sodium–glucose cotrans-
2 diabetes, the presence of CKD markedly duration of diabetes, retinopathy, albumin- porter 2 (SGLT2) inhibitors may promote
increases cardiovascular risk (7). uria without hematuria, and gradually pro- AKI through volume depletion, particu-
gressive kidney disease. However, signs of larly when combined with diuretics or
Assessment of Albuminuria and CKD may be present at diagnosis or without other medications that reduce glomeru-
Estimated Glomerular Filtration Rate retinopathy in type 2 diabetes, and reduced lar filtration. However, existing evidence
Screening for albuminuria can be most eGFR without albuminuria has been fre- from clinical trials and observational stud-
easily performed by urinary albumin–to– quently reported in type 1 and type 2 di- ies suggests that SGLT2 inhibitors do not
creatinine ratio (UACR) in a random spot abetes and is becoming more common
urine collection (1,2). Timed or 24-h

care.diabetesjournals.org Microvascular Complications and Foot Care S107

Table 10.1—CKD stages and corresponding focus of kidney-related care Focus of kidney-related care
CKD stage†

Stage eGFR Evidence of Diagnose Evaluate and treat Evaluate and Prepare for renal
(mL/min/1.73 m2) kidney cause of risk factors for CKD treat CKD replacement therapy
kidney injury
damage* progression** complications***

No clinical $60 d ! ! !
evidence of $90 1 ! ! !
CKD 60–89 1 ! ! !
30–59 1/2 !
1 15–29 1/2 !
2 ,15 1/2 !
3
4
5

†CKD stages 1 and 2 are defined by evidence of kidney damage (1), while CKD stages 3–5 are defined by reduced eGFR with or without evidence of kidney
damage (1/2). *Kidney damage is most often manifest as albuminuria (UACR $30 mg/g Cr) but can also include glomerular hematuria, other
abnormalities of the urinary sediment, radiographic abnormalities, and other presentations. **Risk factors for CKD progression include elevated blood
pressure, glycemia, and albuminuria. ***See Table 10.2.

significantly increase AKI (16,17). Timely to therapy and disease progression and Comorbidities” for further information
identification and treatment of AKI are may aid in assessing adherence to ACE in- on immunization).
important because AKI is associated hibitor or ARB therapy. In addition, in clin-
with increased risks of progressive CKD ical trials of ACE inhibitors or ARB therapy Interventions
and other poor health outcomes (18). in type 2 diabetes, reducing albuminuria
from levels $300 mg/g Cr has been asso- Nutrition
Surveillance ciated with improved renal and cardiovas-
Albuminuria and eGFR should be moni- cular outcomes, leading some to suggest For people with nondialysis-dependent di-
tored regularly to enable timely diagnosis that medications should be titrated to min- abetic kidney disease, dietary protein intake
of diabetic kidney disease, monitor pro- imize UACR. However, this approach has should be approximately 0.8 g/kg body
gression of diabetic kidney disease, not been formally evaluated in prospec- weight per day (the recommended daily al-
detect superimposed kidney diseases in- tive trials. In type 1 diabetes, remission of lowance) (1). Compared with higher levels
cluding AKI, assess risk of CKD compli- albuminuria may occur spontaneously of dietary protein intake, this level slowed
cations, dose drugs appropriately, and and cohort studies evaluating associa- GFR decline with evidence of a greater ef-
determine whether nephrology referral tions of change in albuminuria with clini- fect over time. Higher levels of dietary pro-
is needed. Among people with existing cal outcomes have reported inconsistent tein intake (.20% of daily calories from
kidney disease, albuminuria and eGFR results (22,23). protein or .1.3 g/kg/day) have been as-
may change due to progression of dia- sociated with increased albuminuria, more
betic kidney disease, development of a The prevalence of CKD complications rapid kidney function loss, and CVD mor-
separate superimposed cause of kidney correlates with eGFR. When eGFR is tality and therefore should be avoided.
disease, AKI, or other effects of medica- ,60 mL/min/1.73 m2, screening for com- Reducing the amount of dietary protein
tions, as noted above. Serum potassium plications of CKD is indicated (Table 10.2). below the recommended daily allowance
should also be monitored for patients Early vaccination against hepatitis B virus of 0.8 g/kg/day is not recommended be-
treated with ACE inhibitors, ARBs, and di- is indicated in patients likely to progress cause it does not alter glycemic measures,
uretics because these medications can to ESRD (see Section 3 “Comprehensive cardiovascular risk measures, or the
cause hyperkalemia or hypokalemia, Medical Evaluation and Assessment of course of GFR decline. In dialysis, protein-
which are associated with cardiovascular energy wasting is common, and in-
risk and mortality (19–21). For patients creased dietary protein intake may be
with eGFR ,60 mL/min/1.73 m2, appro-
priate medication dosing should be veri- Table 10.2—Selected complications of CKD
fied, exposure to nephrotoxins (e.g.,
nonsteroidal anti-inflammatory drugs Complication Medical and laboratory evaluation
and iodinated contrast) should be mini-
mized, and potential CKD complications Elevated blood pressure Blood pressure, weight
should be evaluated.
Volume overload History, physical examination, weight
The need for annual quantitative assess-
ment of albumin excretion after diagnosis Electrolyte abnormalities Serum electrolytes
of albuminuria, institution of ACE inhibitors
or ARB therapy, and achieving blood pres- Metabolic acidosis Serum electrolytes
sure control is a subject of debate. Contin-
ued surveillance can assess both response Anemia Hemoglobin; iron testing if indicated

Metabolic bone disease Serum calcium, phosphate, PTH, vitamin 25(OH)D

Complications of CKD generally become prevalent when eGFR falls below 60 mL/min/1.73 m2 (stage

3 CKD or greater) and become more common and severe as CKD progresses. Evaluation of elevated

blood pressure and volume overload should occur at every possible clinical contact; laboratory
evaluations are generally indicated every 6–12 months for stage 3 CKD, every 3–5 months for stage
4 CKD, and every 1–3 months for stage 5 CKD, or as indicated to evaluate symptoms or changes in
therapy. PTH, parathyroid hormone; 25(OH)D, 25-hydroxyvitamin D.

S108 Microvascular Complications and Foot Care Diabetes Care Volume 41, Supplement 1, January 2018

necessary to help preserve muscle mass tubular glucose reabsorption, weight, sys- Disease and Risk Management” for fur-
and function. temic blood pressure, intraglomerular ther discussion). All of these trials included
pressure, and albuminuria and slow GFR large numbers of people with kidney dis-
For some patients with diabetes, restric- loss through mechanisms that appear ease (for example, the baseline prevalence
tion of dietary sodium may be useful to independent of glycemia (17,38–40). of albuminuria in EMPA-REG OUTCOME
control blood pressure and reduce cardio- Glucagon-like peptide 1 receptor agonists was 53%), and some of the cardiovascular
vascular risk (24), and restriction of dietary and dipeptidyl peptidase 4 inhibitors also outcomes trials (CANVAS and LEADER)
potassium may be necessary to control have direct effects on the kidney and have were enriched with patients with kidney
serum potassium concentration (15,19– been reported to improve renal outcomes disease through eligibility criteria based on
21). These interventions may be most im- compared with placebo (41–44). albuminuria or reduced eGFR. The glucose-
portant for patients with reduced eGFR, lowering effects of SGLT2 inhibitors are
for whom urinary excretion of sodium and A number of large cardiovascular out- blunted with eGFR (17,45). However, the
potassium may be impaired. Recommen- comes trials in patients with type 2 diabetes cardiovascular benefits of empagliflozin,
dations for dietary sodium and potassium at high risk for cardiovascular disease or with canagliflozin, and liraglutide were similar
intake should be individualized on the basis existing cardiovascular disease (EMPA-REG among participants with and without kid-
of comorbid conditions, medication use, OUTCOME [BI 10773 (Empagliflozin) Car- ney disease at baseline (40,41,45,46).
blood pressure, and laboratory data. diovascular Outcome Event Trial in Type 2
Diabetes Mellitus Patients], CANVAS With reduced eGFR, drug dosing may
Glycemia [Canagliflozin Cardiovascular Assessment require modification (1). The U.S. Food
Study], LEADER [Liraglutide Effect and and Drug Administration (FDA) revised
Intensive glycemic control with the goal Action in Diabetes: Evaluation of Cardio- guidance for the use metformin in dia-
of achieving near-normoglycemia has been vascular Outcome ResultsdA Long Term betic kidney disease in 2016 (47), recom-
shown in large prospective randomized Evaluation], and SUSTAIN-6 [Trial to Eval- mending use of eGFR instead of serum Cr
studies to delay the onset and progression uate Cardiovascular and Other Long-term to guide treatment and expanding the
of albuminuria and reduced eGFR in patients Outcomes With Semaglutide in Subjects With pool of patients with kidney disease for
with type 1 diabetes (25,26) and type 2 di- Type 2 Diabetes]) examined kidney effects whom metformin treatment should be
abetes (1,27–32). Insulin alone was used as secondary outcomes (40,41,44,45). considered. Revised FDA guidance states
to lower blood glucose in the Diabetes Specifically, compared with placebo, that metformin is contraindicated in patients
Control and Complications Trial (DCCT)/ empagliflozin reduced the risk of incident with an eGFR ,30 mL/min/ 1.73 m2, eGFR
Epidemiology of Diabetes Interventions or worsening nephropathy (a composite of should be monitored while taking metfor-
and Complications (EDIC) study of type 1 progression to UACR .300 mg/g Cr, dou- min, the benefits and risks of continuing
diabetes, while a variety of agents were bling of serum Cr, ESRD, or death from treatment should be reassessed when
used in clinical trials of type 2 diabetes, ESRD) by 39% and the risk of doubling of eGFR falls ,45 mL/min/1.73 m2, metfor-
supporting the conclusion that glycemic serum Cr accompanied by eGFR #45 mL/ min should not be initiated for patients
control itself helps prevent diabetic kidney min/1.73 m2 by 44%; canagliflozin reduced with an eGFR ,45 mL/min/1.73 m2, and
disease and its progression. The effects of the risk of progression of albuminuria by metformin should be temporarily discon-
glucose-lowering therapies on diabetic 27% and the risk of reduction in eGFR, tinued at the time of or before iodinated
kidney disease have helped define A1C ESRD, or death from ESRD by 40%; liraglu- contrast imaging procedures in patients
targets (see Table 6.2). tide reduced the risk of new or worsening with eGFR 30–60 mL/min/ 1.73 m2. Other
nephropathy (a composite of persistent glucose-lowering medications also re-
The presence of diabetic kidney dis- UACR .300 mg/g Cr, doubling of serum quire dose adjustment or discontinuation
ease affects the risks and benefits of in- Cr, ESRD, or death from ESRD) by 22%; and at low eGFR (see Table 8.2) (1).
tensive glycemic control and a number of semaglutide reduced the risk of new or
specific glucose-lowering medications. In worsening nephropathy (a composite of Cardiovascular Disease and Blood Pressure
the Action to Control Cardiovascular Risk persistent UACR .300 mg/g Cr, doubling
in Diabetes (ACCORD) trial of type 2 di- of serum Cr, or ESRD) by 36% (each P , Hypertension is a strong risk factor for the
abetes, adverse effects of intensive glyce- 0.01). Additional trials with primary kid- development and progression of diabetic
mic control (hypoglycemia and mortality) ney outcomes are needed to definitively kidney disease (48). Antihypertensive ther-
were increased among patients with determine whether specific glucose-low- apy reduces the risk of albuminuria (49–
kidney disease at baseline (33,34). More- ering drugs improve renal outcomes. 52), and among patients with type 1 or
over, there is a lag time of at least 2 years 2 diabetes with established diabetic kid-
in type 2 diabetes to over 10 years in type 1 Patients with diabetic kidney disease ney disease (eGFR ,60 mL/min/1.73 m2
diabetes for the effects of intensive glucose are at high risk of cardiovascular events, and UACR $300 mg/g Cr), ACE inhibitor
control to manifest as improved eGFR out- and some SGLT2 inhibitors and glucagon- or ARB therapy reduces the risk of pro-
comes (31,35,36). Therefore, in some pa- like peptide 1 receptor agonists have gression to ESRD (53–55). Moreover, an-
tients with prevalent diabetic kidney demonstrated cardiovascular benefits. tihypertensive therapy reduces risks of
disease and substantial comorbidity, target Namely, in EMPA-REG OUTCOME, CANVAS, cardiovascular events (49).
A1C levels may be less intensive (1,37). and LEADER, empagliflozin, canagliflozin,
and liraglutide, respectively, each re- Blood pressure levels ,140/90 mmHg
Specific Glucose-Lowering Medications duced cardiovascular events, evaluated are generally recommended to reduce
as primary outcomes, compared with CVD mortality and slow CKD progression
Some glucose-lowering medications also placebo (see Section 9 “Cardiovascular among people with diabetes (52). Lower
have effects on the kidney that are direct, blood pressure targets (e.g., ,130/80
i.e., not mediated through glycemia. For
example, SGLT2 inhibitors reduce renal

care.diabetesjournals.org Microvascular Complications and Foot Care S109

mmHg) may be considered for patients are effective for management of resistant c If there is no evidence of retinopathy
based on individual anticipated benefits hypertension, have been shown to reduce for one or more annual eye exam and
and risks. Patients with diabetic kidney dis- albuminuria in short-term studies of dia- glycemia is well controlled, then
ease are at increased risk of CKD progression betic kidney disease, and may have addi- exams every 1–2 years may be con-
(particularly those with albuminuria) and tional cardiovascular benefits (65–67). sidered. If any level of diabetic ret-
CVD and therefore may be suitable in There has been, however, an increase in inopathy is present, subsequent
some cases for lower blood pressure targets. hyperkalemic episodes in those on dual dilated retinal examinations should
therapy, and larger, longer trials with clin- be repeated at least annually by an
ACE inhibitors or ARBs are the pre- ical outcomes are needed before recom- ophthalmologist or optometrist. If
ferred first-line agent for blood pressure mending such therapy. retinopathy is progressing or sight-
treatment among patients with diabetes, threatening, then examinations will
hypertension, eGFR ,60 mL/min/1.73 m2, Referral to a Nephrologist be required more frequently. B
and UACR $300 mg/g Cr because of their
proven benefits for prevention of CKD Consider referral to a physician expe- c While retinal photography may
progression (53–56). In general, ACE inhibi- rienced in the care of kidney disease serve as a screening tool for reti-
tors and ARBs are considered to have similar when there is uncertainty about the eti- nopathy, it is not a substitute for a
benefits (57,58) and risks. In the setting of ology of kidney disease, difficult man- comprehensive eye exam. E
lower levels of albuminuria (30–299 mg/g agement issues (anemia, secondary
Cr), ACE inhibitor or ARB therapy has been hyperparathyroidism, metabolic bone c Women with preexisting type 1 or
demonstrated to reduce progression to disease, resistant hypertension, or elec- type 2 diabetes who are planning
more advanced albuminuria ($300 mg/g trolyte disturbances), or advanced kidney pregnancy or who are pregnant
Cr) and cardiovascular events but not pro- disease (eGFR ,30 mL/min/1.73 m2) re- should be counseled on the risk of
gression to ESRD (56,59). While ACE inhib- quiring discussion of renal replacement development and/or progression of
itors or ARBs are often prescribed for therapy for ESRD. The threshold for re- diabetic retinopathy. B
albuminuria without hypertension, clini- ferral may vary depending on the fre-
cal trials have not been performed in quency with which a provider encounters c Eye examinations should occur be-
this setting to determine whether this im- patients with diabetes and kidney dis- fore pregnancy or in the first trimes-
proves renal outcomes. ease. Consultation with a nephrologist ter in patients with preexisting
when stage 4 CKD develops (eGFR #30 type 1 or type 2 diabetes, and then
Absent kidney disease, ACE inhibitors mL/min/1.73 m2) has been found to re- patients should be monitored every
or ARBs are useful to control blood pres- duce cost, improve quality of care, and trimester and for 1 year postpartum
sure but may not be superior to alterna- delay dialysis (68). However, other spe- as indicated by the degree of reti-
tive proven classes of antihypertensive cialists and providers should also educate nopathy. B
therapy, including thiazide-like diuretics their patients about the progressive na-
and dihydropyridine calcium channel ture of diabetic kidney disease, the kidney Treatment
blockers (60). In a trial of people with preservation benefits of proactive treat- c Promptly refer patients with any
type 2 diabetes and normal urine albumin ment of blood pressure and blood glu-
excretion, an ARB reduced or suppressed cose, and the potential need for renal level of macular edema, severe
the development of albuminuria but in- replacement therapy. nonproliferative diabetic retinopa-
creased the rate of cardiovascular events thy (a precursor of proliferative
(61). In a trial of people with type 1 di- DIABETIC RETINOPATHY diabetic retinopathy), or any prolif-
abetes exhibiting neither albuminuria nor erative diabetic retinopathy to an
hypertension, ACE inhibitors or ARBs did Recommendations ophthalmologist who is knowledge-
not prevent the development of diabetic able and experienced in the man-
glomerulopathy assessed by kidney bi- c Optimize glycemic control to re- agement of diabetic retinopathy. A
opsy (62). Therefore, ACE inhibitors or duce the risk or slow the progres- c The traditional standard treatment,
ARBs are not recommended for patients sion of diabetic retinopathy. A panretinal laser photocoagulation
without hypertension to prevent the de- therapy, is indicated to reduce the
velopment of diabetic kidney disease. c Optimize blood pressure and serum risk of vision loss in patients with
lipid control to reduce the risk or high-risk proliferative diabetic retinop-
Two clinical trials studied the combina- slow the progression of diabetic ret- athy and, in some cases, severe non-
tions of ACE inhibitors and ARBs and found inopathy. A proliferative diabetic retinopathy. A
no benefits on CVD or diabetic kidney dis- c Intravitreous injections of anti–
ease, and the drug combination had higher Screening vascular endothelial growth factor
adverse event rates (hyperkalemia and/or c Adults with type 1 diabetes should ranibizumab are not inferior to tradi-
AKI) (63,64). Therefore, the combined use tional panretinal laser photocoagula-
of ACE inhibitors and ARBs should be have an initial dilated and compre- tion and are also indicated to reduce
avoided. hensive eye examination by an oph- the risk of vision loss in patients with
thalmologist or optometrist within proliferative diabetic retinopathy. A
Mineralocorticoid receptor antagonists 5 years after the onset of diabetes. B c Intravitreous injections of anti–
(spironolactone, eplerenone, and finere- c Patients with type 2 diabetes should vascular endothelial growth factor
none) in combination with ACE inhibitors have an initial dilated and compre- are indicated for central-involved di-
or ARBs remain an area of great interest. hensive eye examination by an oph- abetic macular edema, which occurs
Mineralocorticoid receptor antagonists thalmologist or optometrist at the
time of the diabetes diagnosis. B