Frequency and Determinants of Postprocedural Hemodynamic ...

Frequency and Determinants of Postprocedural
Hemodynamic Instability After Carotid
Angioplasty and Stenting

Adnan I. Qureshi, MD; Andreas R. Luft, MD; Mudit Sharma, BS; Vallabh Janardhan, MD;
Demetrius K. Lopes, MD; Jehanzeb Khan, MD; Lee R. Guterman, PhD, MD; L. Nelson Hopkins, MD

Background and Purpose—Hemodynamic instability can occur acutely after carotid angioplasty and stent placement
(CAS). We performed this study to determine the frequency of hemodynamic instability in a series of patients who
underwent CAS and to analyze factors associated with development of postprocedural hemodynamic events.

Methods—We reviewed medical records and angiograms in a series of 51 patients (mean age 68.3Ϯ8.9 years) who
underwent CAS for symptomatic (nϭ29) or asymptomatic (nϭ22) carotid artery stenosis. Any episodes of hypotension
(systolic blood pressure Ͻ90 mm Hg), hypertension (systolic blood pressure Ͼ160 mm Hg), or bradycardia (heart rate
Ͻ60 bpm) that occurred in the acute postprocedural period were recorded. The effect of demographic, clinical,
intraprocedural, and angiographic factors on subsequent development of hemodynamic instability was analyzed by
logistic regression.

Results—The frequency of postprocedural hemodynamic complications in our patient series was as follows: hypotension,
22.4%; hypertension, 38.8%; and bradycardia, 27.5%. Intraprocedural hypotension (odds ratio [OR] 14.6, Pϭ0.024) and
history of myocardial infarction (OR 14.1, Pϭ0.04) independently predicted postprocedural hypotension. Postproce-
dural hypertension was predicted by intraprocedural hypertension (OR 7.6, Pϭ0.01) and previous ipsilateral carotid
endarterectomy (OR 7.6, Pϭ0.02). Postprocedural bradycardia was associated with intraprocedural hypotension (OR 74,
Pϭ0.001) and intraprocedural bradycardia (OR 12, Pϭ0.008). All events had resolved at the conclusion of the intensive
care unit monitoring period (mean 25.7 hours, range 18 to 43 hours).

Conclusions—Postprocedural hemodynamic instability is frequent after CAS and supports the need for monitoring in
settings suited to expeditious management of cardiovascular emergencies. Patients who have evidence of hemodynamic
instability during the procedure are at highest risk. (Stroke. 1999;30:2086-2093.)

Key Words: angioplasty Ⅲ carotid artery Ⅲ hypotension Ⅲ stents

H emodynamic instability consisting of hypertension, hy- and predisposing factors of hemodynamic instability in the
potension, or bradycardia after carotid endarterectomy postprocedural period have not been investigated.
(CEA) is well recognized.1–7 Acute hypertension is attributed
to transient dysfunction of adventitial baroreceptors in the We performed this study to determine the frequency and
endarterectomized carotid artery segments, although metabol- characteristics of hemodynamic instability in the acute post-
ic factors such as renin and vasopressin have also been procedural period after CAS. We also evaluated the effect of
implicated.3,5,7 Bradycardia, sometimes associated with hy- factors related to both the patient’s medical condition and the
potension, is also reported to occur frequently in the early procedure itself on postprocedural hemodynamic status.
postoperative period.7 Presumably, bradycardia and hypoten-
sion are related to increased activity from the carotid sinus Subjects and Methods
nerve or carotid baroreceptors due to increased compliance
and stretch in the arterial wall after removal of atheromatous We reviewed the medical records and angiograms of patients who
plaque.1,6 Recently, carotid angioplasty and stent placement underwent CAS from September 1997 through October 1998.
(CAS) has been introduced as an alternative to CEA for the Patients were identified by a registry maintained by the Department
treatment of carotid artery stenosis.8–11 Although CAS in- of Neurosurgery. Only those patients who underwent angioplasty for
volves extensive manipulation in the vicinity of both the carotid artery stenosis were included. Patients with stenosis related to
adventitial baroreceptors and the carotid sinus, the frequency dissection or radiation therapy were not included.

A written informed consent was obtained from each patient. The
study was approved by the local institutional review committee.

Received March 11, 1999; final revision received June 24, 1999; accepted July 22, 1999.
From the Department of Neurosurgery and Toshiba Stroke Research Center, School of Medicine and Biomedical Sciences, University at Buffalo, State
University of New York, Buffalo, NY.
Correspondence to Adnan I. Qureshi, MD, SUNYAB Department of Neurosurgery, 3 Gates Circle, Buffalo, NY 14209-1194.
© 1999 American Heart Association, Inc.

Stroke is available at http://www.strokeaha.org

Downloaded from http://stroke.ahaj2o0u8rn6als.org/ by guest on April 11, 2016

Qureshi et al Hemodynamic Instability After Carotid Angioplasty and Stenting 2087

Protocol for Angioplasty and Stent Placement Data Collection

Patients were started on aspirin (325 mg/d) and ticlopidine (250 mg Records Review
BID) or clopidogrel (75 mg/d) 72 hours before the procedure. On the From the medical records review, the following information was
day of the procedure, each patient was evaluated by a physician from collected for each patient: age; sex; race; current smoking status or
the endovascular team, and a complete physical examination was alcohol use; history of angina pectoris, myocardial infarction (MI),
documented, including measurements of heart rate, respiratory rate, coronary artery disease (including previous bypass or angioplasty),
and blood pressure. Each patient was given dexamethasone (10 mg) hypertension, diabetes mellitus, hyperlipidemia, cardiac dysrhyth-
and nimodipine (60 mg) orally on the morning of the procedure for mia, valvular heart disease, cardiac failure, peripheral vascular
potential neuroprotective effect. Patients taking hypertensive medi- disease, renal insufficiency, pulmonary disease, or CEA; and medi-
cations received their morning doses; additional doses were withheld cations used before admission (including antiplatelet agents). Any
until after the procedure. clinical symptoms were noted for documentation of symptomatic
ipsilateral or contralateral carotid artery disease. The following
Heart and respiratory rates were monitored continuously through- procedural details were collected: diameter of largest balloon used
out the procedure. Blood pressure was measured at 5-minute inter- and diameter and type of stent placed.
vals at the start of the procedure and every minute during balloon
inflations and stent placement. The blood pressure was measured in Blood pressure, heart rate, and respiratory rate measurements
1 arm (usually the left) by an automated-cuff-inflation sphygmoma- obtained before, during, and after the procedure were recorded. Any
nometer (Omnicare 24, Hewlett Packard). The same arm was used episodes of bradycardia (defined as heart rate Ͻ60 bpm), hypoten-
consistently throughout the procedure and during the postoperative sion (defined as systolic blood pressure Ͻ90 mm Hg), and hyper-
period. tension (defined as systolic blood pressure Ͼ160 mm Hg), as well as
medications used to treat these conditions, were recorded. Although
Baseline angiography was performed, and the lumen diameters of recordings were made more frequently, blood pressure, heart rate,
the stenotic and adjacent arterial segments were measured. The and respiratory rate values (for purposes of analysis) were collected
angioplasty was performed with a large-lumen guide catheter, a at 30-minute intervals during the procedure and at 2-hour intervals
flexible guidewire, and a balloon catheter. An intravenous heparin for 18 to 43 hours (mean 25.7 hours) afterward. The lengths of stay
bolus (5000 to 7500 U) was given to achieve an activated clotting in the hospital and in the intensive care unit were recorded for each
time of Ϸ300 seconds. If there was evidence of thrombosis or patient.
ulceration in the stenotic segment, urokinase (150 000 to 250 000 U)
was infused through a microcatheter (around the guidewire) proxi- Evaluation of Angiograms
mal to the lesion before introduction of the balloon.12 A 6F to 9F Digital subtraction angiography images were collected retrospec-
guide catheter (2 to 3 mm in diameter) was advanced through the tively from the digital tape storage system of our angiography unit
sheath into the common carotid artery. The flexible, angulated-tip (Toshiba Medical Systems). Ipsilateral and contralateral diagnostic
guidewire (0.14 to 0.18 inches in diameter) was advanced through images of the extracranial carotid arteries and the intracranial
the guide catheter, navigated across the stenosis, and positioned in circulation, as well as images of the stented artery, were obtained.
the distal arterial segment. With the guidewire across the stenosis, The extracranial carotid artery images were transferred to a computer
the deflated balloon catheter was advanced over the wire and workstation running NIH Image software (http://rsb.info.nih.gov/
positioned at the stenosis. The positions of the guidewire and balloon nih-image/) for subsequent measurements. After adjustments for
catheter were inspected periodically under fluoroscopy. Intravenous magnification were made, diameters of the stenosis, the distal
atropine was given for prophylaxis in some patients at the physi- internal carotid artery (ICA), and the stented lesion were measured.
cian’s discretion before inflation of balloon. Intravenous atropine All measurements were made on lateral views. The measurements
was given either as a 0.5- (nϭ6), 1.0- (nϭ10), or 1.5-mg (nϭ2) dose. were performed in user-defined regions of interest (ROIs) after
The balloon was inflated for a period lasting from a few seconds to adjustment for image scale. NIH Image measurement functionality
1 minute at 6 to 12 atm with a mixture of saline and contrast material was used (rulers that were applied manually after the ROIs were
used to fluoroscopically visualize the inflation. If the stenosis was magnified 4-fold). The degree of stenosis was calculated according
adequately dilated, the balloon catheter was removed. If the dilation to the primary NASCET method, ie, diameter of stenosis/distal ICA
was inadequate, the balloon catheter was replaced by a larger one. diameter.13 This method was chosen to ensure comparability with
After predilatation, a stent was placed across the dilated segment other studies.14 Extension of the lesion was classified according to
over the guidewire by use of a stent-delivery system. The size and involvement of the common carotid artery, the medial or lateral wall
type of stent selected were determined by both baseline and of the carotid bulb, the external carotid artery, and/or the distal ICA.
predilatation angiographic evaluations. After stent placement, a Similarly, it was recorded whether the stent covered the common
second dilatation was performed to anchor the stent in normal vessel carotid artery, the bulb, and/or the ICA. Each of these angiographic
by full expansion with a larger balloon catheter. A final angiogram features was further analyzed as a dichotomized variable.
was obtained to confirm that the final dilation of the stenotic segment
was adequate and no local complications such as dissection had Statistical Analysis
occurred. Intracranial vessels were also imaged to avoid undetected
compromise of intracranial circulation by thromboembolic events. The effects of 43 variables collected from patient charts and 11
variables obtained from angiographic images were evaluated for
The patient was transferred to the neurointensive care unit for each of the following outcomes: postprocedural hypotension, post-
overnight observation. Heart rate and respiratory rate were moni- procedural hypertension, and postprocedural bradycardia. Stepwise
tored continuously. Blood pressure was monitored every 15 minutes logistic regression was used to analyze these relationships with an
by use of an automated-cuff-inflation sphygmomanometer (model entry criteria of PϽ0.1. A P value of Ͻ0.05 was considered
56, Hewlett Packard). Neurological evaluations were performed significant. The effects of continuous variables, such as age or the
every 2 hours and more frequently if the patient’s condition was degree of stenosis, were entered as linear factors after they were
deteriorating. Postprocedural hypotension was treated with intrave- tested for nonlinearity by use of the Box-Tidwell transformation.15
nous infusion of dopamine titrated to maintain systolic blood Interactions among the effects variables were tested for significant
pressure greater than 100 mm Hg. Postprocedural hypertension was contribution to the logistic-regression model (PϽ0.1). All possible
treated with intravenous labetalol or hydralazine to reduce systolic interaction terms were nonsignificant and were therefore excluded
blood pressure to preprocedural values. Postprocedural sinus brady- from the model.
cardia was treated with intravenous atropine (0.5 to 1.0 mg), and
bradycardia due to second- or third-degree atrioventricular block was Results
treated with transvenous pacemaker.
Fifty-four consecutive patients were initially included in the

study. Three patients were excluded because angiographic

Downloaded from http://stroke.ahajournals.org/ by guest on April 11, 2016

2088 Stroke October 1999

TABLE 1. Demographic and Clinical Characteristics

Postprocedural

Variable All Hypertension Hypotension Bradycardia

Number 51 19 of 49 (38.8%) 11 of 49 (22.4%) 14 of 51 (27.5%)

Age, mean (SD), y 68.31 (8.99) 65.84 (10.67) 74.09 (6.53) 71.07 (5.99)

Male sex 33 (64.7%) 8 (42.1%) 7 (63.6%) 10 (71.4%)

History, n (%)

Coronary artery disease 31 (60.8) 13 (68.4) 6 (54.5) 8 (57.1)

Diabetes mellitus 25 (49.0) 11 (57.9) 3 (27.3) 6 (42.9)

Hyperlipidemia 28 (54.9) 11 (57.9) 6 (54.5) 9 (64.3)

Hypertension 44 (86.3) 19 (100.0) 9 (81.8) 12 (85.7)

MI 14 (27.5) 4 (21.1) 6 (54.5) 5 (35.7)

Neurological symptoms 6 (11.8) 3 (15.8) 1 (9.1) 1 (7.1)

Valvular heart disease 8 (15.7) 4 (21.1) 0 (0.0) 1 (7.1)

Dysrhythmia 9 (17.6) 3 (15.8) 2 (18.2) 2 (14.3)

Symptoms of carotid stenosis, ipsilateral 29 (56.9) 11 (57.9) 3 (27.3) 7 (50.0)

Symptoms of carotid stenosis, contralateral 7 (13.7) 4 (21.1) 3 (27.3) 4 (28.6)

␤-Blocker medication 18 (35.3) 8 (42.1) 4 (36.4) 6 (42.9)

Previous CEA, ipsilateral 14 (27.5) 9 (47.4) 3 (27.3) 3 (21.4)

Previous CEA, contralateral 8 (15.7) 3 (15.8) 3 (27.3) 2 (14.3)

Race, history of smoking, alcohol abuse, angina pectoris, chronic obstructive pulmonary disease, chronic heart failure, coronary
artery bypass, claudication, and renal insufficiency were not significantly associated with either of the hemodynamic events.

material was not available. For the remaining 51 patients (18 104.3 mm Hg (mean 85.4 mm Hg) in the nonhypotensive
women, 33 men; mean age 68.3 years), descriptive charac- group. These differences in minimal (PϽ0.001), maximal
teristics are given in Tables 1, 2, and 3. The mean degree of (PϽ0.05), and mean (PϽ0.01) MAP values were statistically
stenosis on the treated side was 69Ϯ19% (Table 3). The significant. Accordingly, significant differences also were
severity of stenosis ranged from 50% to 97.5% in all but 1 found for time-averaged systolic and diastolic blood pres-
patient. One patient had a measured stenosis of 42.5% despite sures (PϽ0.01). No differences were found in heart or
high-grade stenosis because the stenosis extended up to the respiration rates between hypotensive and nonhypotensive
petrous portion, therefore affecting the distal reference point. patients.
In 45 patients, the contralateral ICA showed stenoses of
varying degrees (complete occlusion in 6 cases). Stenoses in In 6 (54.5%) of 11 patients experiencing postprocedural
42 patients were treated with Wall stents (Schneider), in 1 hypotensive episodes, hypotension was also present intraop-
patient with a GFX stent (Arterial Vascular Engineering), and eratively. In contrast, only 2 (5.3%) of 38 nonhypotensive
in 4 patients with Integra stents (SCIMED Boston Scientific). patients had intraprocedural hypotension (Pearson ␹2
In procedures in 4 patients treated early in our series, Palmaz PϽ0.001). Postprocedural bradycardia was significantly
stents (Johnson & Johnson) were used. The mean intensive more frequent in hypotensive than in nonhypotensive patients
care unit and hospital stays after angioplasty and stent (64% versus 18.4%; Pearson ␹2 PϽ0.01). However, the
placement were 2.2Ϯ2.5 and 4.2Ϯ3.5 days, respectively. changes in blood pressures and heart rate were not associated
in time. Patient age was a predictor for postprocedural
Postprocedural Hypotension hypotension: the mean age of hypotensive patients was 74.1
Postprocedural hypotension was observed in 13 cases. Two of years versus 66.7 years in the nonhypotensive group (2-sided
these patients were excluded from additional analysis because t test with unequal variances PϽ0.005). The incidence of
the hypotensive event followed postoperative antihyperten- hypotension was highest in patients between 70 and 75 years
sive treatment. Of the 49 patients included, the frequency of of age (38.5%). In patients older than 75 years, the incidence
postoperative hypotension was 22.4% (nϭ11). The period of decreased slightly (30.0%). A history of MI was often found
blood pressure instability (end point defined by the latest in the hypotensive group (54.5% versus 21.1% in the remain-
reading Ͻ90 mm Hg with intermittent higher readings) lasted ing patients; Pearson ␹2 PϽ0.05). Carotid stenosis was less
for 4 to 27 hours (mean 14.8 hours). Eight of 11 (72.7%) frequently symptomatic in hypotensive patients than in non-
hypotensive patients required dopamine treatment (Table 2). hypotensive patients (27.3% versus 63.2%; Pearson ␹2
In the remainder, hypotension resolved spontaneously. PϽ0.05). An additional variable associated with postopera-
tive hypotension was the change in vessel diameter after
When all mean arterial pressure (MAP) readings for each stenting, as described by the ratio of the diameter of the
patient were averaged over the entire monitoring period, these unstented stenotic segment divided by that of the stented
values ranged from 55.7 to 95.6 mm Hg (mean 72.5 mm Hg) stenotic segment. Greater diameter changes were protective
in the hypotensive group. This compares with 70.5 to

Downloaded from http://stroke.ahajournals.org/ by guest on April 11, 2016

Qureshi et al Hemodynamic Instability After Carotid Angioplasty and Stenting 2089

TABLE 2. Procedure and Treatment-Related Characteristics

Postprocedural

Procedure All Hypertension Hypotension Bradycardia
(nϭ51) (nϭ19 of 49) (nϭ11 of 49) (nϭ14 of 51)

Side, n (%)

Left 31 (60.8) 10 (52.6) 5 (45.5) 8 (57.1)

Right 20 (39.2) 9 (47.4) 6 (54.5) 6 (42.9)

PrePRO heart rate, mean (SD), bpm 80.41 (16.63) 83.94 (19.59) 82.82 (17.99) 77.36 (16.93)

PrePRO systolic blood pressure, mean (SD), mm Hg 146.31 (24.41) 159.56 (26.78) 147.36 (31.92) 146.36 (28.00)

PrePRO diastolic blood pressure, mean (SD), mm Hg 79.24 (12.48) 85.47 (13.10) 78.09 (14.77) 76.71 (13.06)

Diameter of largest balloon,* mean (SD), mm 4.94 (0.79) 4.94 (1.09) 4.68 (0.40) 4.82 (0.50)

Diameter of stent,† mean (SD), mm 8.22 (1.41) 7.78 (1.22) 8.22 (2.05) 7.92 (1.98)

Ratio of largest balloon diameter to nondiseased vessel diameter, mean (SD) 1.38 (0.48) 1.46 (0.50) 1.28 (0.38) 1.44 (0.63)

Ratio of prestent to poststent diameter, mean (SD) 0.30 (0.18) 0.36 (0.17) 0.25 (0.15) 0.29 (0.16)

IntraPRO hypertension, n (%) 28 (54.9) 16 (84.2) 3 (27.3) 8 (57.1)

IntraPRO hypotension, n (%) 8 (15.7) 0 (0.0) 6 (54.5) 7 (50.0)

IntraPRO bradycardia, n (%) 13 (25.5) 4 (21.1) 1 (9.1) 7 (50.0)

IntraPRO tachycardia, n (%) 1 (2.0) 1 (5.3) 1 (9.1) 1 (7.1)

PostPRO vasopressors, n (%) 9 (17.6) 0 (0.0) 8 (72.7) 4 (28.6)

PostPRO antihypertensives, n (%) 10 (19.6) 9 (47.4) 0 (0.0) 2 (14.3)

PrePRO indicates preprocedural; IntraPRO, intraprocedural; and PostPRO, postprocedural.
*Diameter of balloon represented inflated diameter provided by manufacturer.
†Diameter of stent was provided by manufacturer for each stent.

for hypotension, ie, the greater the change in diameter, the CEA (PϽ0.01), and preprocedural diastolic blood pressures
lower the incidence of hypotension (2-sided t test with (each PϽ0.05). The preprocedural diastolic blood pressure
unequal variances PϽ0.05). was 85.5Ϯ13.1 and 76.0Ϯ10.9 mm Hg in patients with and
without postprocedural hypertension, respectively. In the
After the influence of each variable was tested separately, multivariate model, the variables ipsilateral CEA, intraproce-
a logistic regression model with hypotension as the dependent dural hypertension, and preprocedural diastolic blood pres-
variable was constructed. The final model consisted of the sure remained in the final model (Table 5). The strongest
following independent variables: age, previous MI, symptom- predictor was ipsilateral CEA, followed by intraprocedural
atic carotid stenosis (ipsilateral), and intraprocedural hypo- hypertension. The overall predictability of the model was
tension (Table 4). The strongest predictor was intraprocedural 80.9%.
hypotension, followed by MI history. The overall predictabil-
ity of the model was 91.8%. Postprocedural Bradycardia
Postprocedural bradycardia occurred in 14 (27.5%) of 51
Postprocedural Hypertension patients in our series. Bradycardia was noted in 7 patients
Compared with postprocedural hypotension, postprocedural with postprocedural hypotension and in 4 with postprocedural
hypertension was observed more frequently. Of 21 cases of hypertension. One patient who had postprocedural hypoten-
hypertension during the monitoring period, 2 were excluded sion and hypertension also experienced postprocedural bra-
from additional analysis because the respective patients had dycardia. During the procedure, 7 of the 14 patients were
hypertension after vasopressive treatment for hypotension. given atropine, and 1 patient required transvenous pacing
Therefore, the frequency of hypertension in our series was after bradycardia developed.
38.8%. All 19 patients had a history of hypertension at
baseline. Four of the 19 patients had systolic blood pressure In univariate analyses, significant predictors of bradycardia
Ͼ160 mm Hg before the procedure. Nine of 19 hypertensive were claudication (protective, PϽ0.05), intraprocedural hy-
patients required intravenous antihypertensive therapy in the potension (PϽ0.01), and intraprocedural bradycardia
postprocedural period. Two patients had both hypertension (PϽ0.05). The final model included intraprocedural hypoten-
and hypotension during the monitoring period. Because these sion and intraoperative bradycardia as independent variables
patients received neither vasopressive nor antihypertensive (Table 6). Both variables had a high predictive power. The
medication, it is assumed that both events were procedure overall predictability of the model was 83.7%. When prophy-
related. lactic use of atropine was added to the model, atropine use
was independently associated with higher risk of postproce-
Univariate testing revealed the following significant pre- dural bradycardia (OR 32.7, 95% CI 3.2 to 340.5). Both
dictors of postprocedural hypertension: sex (female intraprocedural hypotension and intraoperative bradycardia
riskϾmale risk, PϽ0.05), history of hypertension (PϽ0.05), remained significant in this model as well.
intraprocedural hypertension (PϽ0.01), previous ipsilateral

Downloaded from http://stroke.ahajournals.org/ by guest on April 11, 2016

2090 Stroke October 1999

TABLE 3. Angiographic Characteristics

Angiographic Data All Hypertension Hypotension Bradycardia
(nϭ51) (nϭ19 of 49) (nϭ11 of 49) (nϭ14 of 51)

Degree of treated stenosis, mean (SD), % 69 (19) 71 (19) 68 (16) 72 (11)
62 (35) 48 (40)
Degree of contralateral stenosis, mean (SD),* % 47 (36) 58 (37)
5 (45.5) 6 (42.9)
Treated lesion involved, n (%) 7 (63.6) 9 (64.3)
8 (72.7) 11 (78.6)
ECA 18 (35.3) 8 (42.1) 4 (36.4) 7 (50.0)
3 (27.3) 4 (28.6)
Medial wall of bulb 32 (62.7) 11 (57.9) 1 (9.1) 2 (14.3)

Lateral wall of bulb 41 (80.4) 14 (73.7) 8 (72.7) 9 (64.3)
9 (81.8) 13 (92.9)
ICA 26 (51.0) 10 (52.6) 6 (54.5) 9 (64.3)

CCA 15 (29.4) 3 (15.8)

Calcification 6 (11.8) 3 (15.8)

Stent covered, n (%)

CCA 37 (72.5) 13 (68.4)

Bulb 46 (90.2) 18 (94.7)

ICA 34 (66.7) 13 (68.4)

ECA indicates external carotid artery; CCA, common carotid artery.
*45 patients had contralateral stenosis.

Temporal Relationship Between Postprocedural were transient and had resolved at the conclusion of the
and Intraprocedural Hemodynamic Instability intensive care unit monitoring period (mean 25.7 hours,
The majority of postprocedural events were temporally sep- range 18 to 43 hours). Previous studies have reported on
arate from the intraprocedural hemodynamic events. Both hemodynamic complications after angioplasty and stent
postprocedural hypotension and bradycardia represented a placement.16–21 The frequency of these complications has
continuum of intraprocedural events in 2 patients and 1 varied depending on patient selection and procedural
patient, respectively. Postprocedural hypertension was tem- technique. Most studies have focused on intraprocedural
porally separate and continuous with intraprocedural hyper- hemodynamic events. Mendelsohn et al16 observed that
tension in 11 and 7 patients, respectively. Exclusion of bradycardia or hypotension occurred during 13 (68.4%) of
patients in whom intraprocedural and postprocedural events 19 CAS procedures studied. Vasopressive medication was
were continuous did not affect the prediction model for either required in 7 patients with sustained hemodynamic insta-
postprocedural hypotension or bradycardia. For postproce- bility. Wholey et al20 observed bradycardia requiring
dural hypertension, both intraprocedural hypertension and temporary pacemaker insertion in 11 (9.6%) of 114 pa-
preprocedural diastolic blood pressure remained significant tients who underwent CAS. Hypotension requiring vaso-
predictors. However, ipsilateral CEA lost significance in the pressor therapy was observed in 1 patient. Yadav et al21
model. Because our results were affected minimally by reported bradycardia in 71% of 107 patients who under-
exclusion of events that continued from the intraprocedural went elective CAS. The bradycardia persisted for a few
into the postprocedural period, such patients are not excluded minutes after balloon deflation. One patient required
in the reported analysis. permanent pacemaker placement 3 days after carotid
stenting. Waigand et al19 reported the development of a
Discussion sinoatrial block in 40 of 50 patients with severe coronary
artery disease during high-pressure inflation for carotid
This report suggests that sustained hemodynamic instabil- angioplasty followed by stenting. Teitelbaum et al18 per-
ity occurs in a significant proportion of patients in the formed CAS in 22 patients. All patients experienced
acute period after CAS. In our patient series, postproce- transient bradycardia with or without asystole (lasting Ͻ30
dural hypertension occurred in 38.8%, bradycardia in seconds) that resolved after balloon deflation and intrave-
27.5%, and hypotension in 22.4% of patients. All events

TABLE 4. Determinants of Postprocedural Hypotension Derived From
Logistic-Regression Analysis

95% CI

Variable Coefficient SE Wald Significance OR Lower Upper
Age 0.139 0.082 2.869 0.090 1.149 0.978 1.350
Previous MI 2.643 1.290 4.199 0.040 14.055 1.122 176.069
Symptoms, ipsilateral 1.406 3.548 0.060 0.071 0.005 1.113
Intraprocedural hypotension Ϫ2.648 1.185 5.107 0.024 14.567 1.427 148.717
Constant 2.679 5.782 3.995 0.046

Ϫ11.555

Downloaded from http://stroke.ahajournals.org/ by guest on April 11, 2016

Qureshi et al Hemodynamic Instability After Carotid Angioplasty and Stenting 2091

TABLE 5. Determinants of Postprocedural Hypertension Derived From
Logistic-Regression Analysis

95% CI

Variable Coefficient SE Wald Significance OR Lower Upper
Carotid endarterectomy, ipsilateral 2.024 0.868 5.441 0.020 7.567 1.382 41.442
Intraprocedural hypertension 2.022 0.822 6.054 0.014 7.556 1.509 37.843
Preprocedural blood pressure, diastolic 0.056 0.033 2.913 0.088 1.057 0.992 1.127
Constant 2.7330 6.103 0.014
Ϫ6.752

nous administration of glycopyrrolate or atropine. Al- symptomatic carotid stenosis (protective effect), and intrapro-
though the frequency of postprocedural hemodynamic cedural hypotension. The strongest predictor was intraproce-
instability was not mentioned, 1 patient had transient dural hypotension. Mendelsohn et al16 observed that postpro-
neurological deficits associated with postprocedural cedural hypotension only occurred in patients with
hypotension. intraprocedural hypotension. In 54.5% of our patients who
had hypotensive episodes after the procedure, hypotension
Hemodynamic complications that occur during and after was already present during the procedure. However, we
CAS are probably mediated through dysfunction of adventi- observed that patients without intraprocedural hypotension
tial baroreceptors in arterial segments that are dilated and can develop hypotension after the procedure. History of MI
covered with intravascular stents.17 The baroreceptors are was another determinant of postprocedural hypotension. Pre-
stretch receptors located in the carotid sinus (dilated segment vious studies have demonstrated increased sensitivity of
of the ICA at its origin from the common carotid artery).22 carotid artery baroreceptors in patients with coronary artery
Impulses arising in the carotid sinus travel through the sinus disease.24,25 Increased baroreceptor sensitivity in conjunction
and glossopharyngeal nerves to the nucleus tractus solitarius with a history of MI may have predisposed patients in the
(NTS) in the caudal medulla. Stimulation of the carotid sinus present study to postprocedural hypotension. The mechanism
inhibits sympathetic neurons in the NTS and reduces sympa- underlying increased responsiveness to carotid sinus stimu-
thetic tone to peripheral blood vessels, leading to a reduction lation is not known. Activation or sensitization of the vagal
in systemic blood pressure. In conjunction with aortic barore- nerve by receptors located in the atrial or ventricular region or
ceptors, the carotid sinus plays a key role in short-term the atrioventricular node due to chronic or acute coronary
adjustments of blood pressure when relatively abrupt changes ischemia has been implicated in the exaggerated carotid sinus
in blood volume, cardiac output, or peripheral resistance response.25 Mendelsohn et al16 additionally observed that the
occur. Impulses from the carotid sinus also initiate excitatory occurrence of postprocedural hypotension was also related to
impulses from the NTS to the nucleus ambiguus and dorsal the placement of large-sized stents. We found that postpro-
vagal nucleus.23 The increase in vagal activity results in a cedural hypotension was associated with the change in vessel
decrease in heart rate. Carotid sinus activity is dependent on diameter after stenting, as described by the ratio of the
arterial pressure. Below a MAP of 60 mm Hg, there are no diameter of the unstented stenosis divided by the diameter of
sinus nerve impulses. Progressively increasing sinus activity the stented stenosis. Greater diameter changes were protec-
is seen above 60 mm Hg, which plateaus at 200 mm Hg.22 tive for hypotension, ie, the greater the change in diameter,
Carotid artery baroreceptors modulate blood pressure by the lower the incidence of hypotension. However, this asso-
reciprocal changes in vagal and sympathetic neural activity. ciation was not significant in multivariate analysis after
At low arterial pressures, carotid sinus impulses only invoke adjustment for other variables. Nimodipine, which can exert
a sympathetic response, with no vagal response. In contrast, a hypotensive effect, was administered hours before the
at high levels of MAP, the predominant response is vagal procedure. However, the hypotensive effect is seen soon after
activity, with minimal sympathetic response. administration and is not consistent with the late hypotension
observed in the present study.
We evaluated the predictors for each hemodynamic com-
plication, ie, hypotension, hypertension, and bradycardia, to Postprocedural hypertension was associated with ipsilat-
gain a better understanding of the underlying pathophysiol- eral CEA, intraprocedural hypertension, and preprocedural
ogy and to identify at-risk individuals. Postprocedural hypo- diastolic blood pressure. Ipsilateral CEA was the strongest
tension was associated with age, previous MI, ipsilateral

TABLE 6. Determinants of Postprocedural Bradycardia Derived From
Logistic-Regression Analysis

95% CI

Variable Coefficient SE Wald Significance OR Lower Upper
Intraprocedural hypotension 4.303 1.297 11.004 0.001 73.951 0.817 940.181
Intraprocedural bradycardia 2.485 0.941 6.972 0.008 12.000 1.898 75.895
Constant 0.731 13.185 0.0003
Ϫ2.656

Downloaded from http://stroke.ahajournals.org/ by guest on April 11, 2016

2092 Stroke October 1999

predictor. Because preprocedural diastolic blood pressure was vessel diameter, or magnitude of change in vessel lumen
entered as a continuous variable, no cutoff value could be diameter on postprocedural hemodynamic instability. Indi-
identified. The preprocedural diastolic blood pressure was vidual variations in sensitivity of the baroreceptors and
85.5Ϯ13.1 and 76.0Ϯ10.9 mm Hg in patients with and adaptive responses may have confounded the association
without postprocedural hypertension, respectively. Carotid between hemodynamic instability and these mechanical fac-
baroreceptor modifications were reported by Angell-James tors. Furthermore, desensitization of baroreceptors due to
and Lumley26 and Tyden et al27 in the acute period after CEA longstanding hypertension may also diminish response to
in humans. However, these alterations are transient. Long- baroreceptor stimulation. The stent covered the sinus region
term changes in carotid baroreceptor function after restenosis in most patients (Table 3), and therefore, an effective analysis
subsequent to CEA are not well described. Dehn and Angell- of the association between sparing of sinus and hemodynamic
James28 assessed the long-term effect of CEA on carotid sinus instability could not be performed.
baroreceptor function and observed decreased baroreceptor
function in a minority of patients. They attributed this Previous investigators have suggested that baroreceptors
diminished activity to periarterial fibrosis (reduced compli- have a short-lasting effect (minutes) on the regulation of
ance) and operative trauma to the carotid sinus nerve and systemic blood pressure.30 However, we observed that acute
baroreceptors. Our results suggest that postendarterectomy intraprocedural dysfunction was a strong predictor of pro-
restenosis predisposes patients to postprocedural hyperten- tracted hemodynamic instability in the postprocedural period.
sion, presumably by diminution in baroreceptor function. Fadali and Walstad31 observed that enlargement of the canine
Theron et al11 similarly observed that bradycardia during carotid sinus diameter by angioplasty or vein-patch grafting
balloon inflation in patients undergoing CAS was not ob- produced decreases in blood pressure in normal and hyper-
served in patients who had previously undergone CEA. tensive dogs that persisted for days in the presence of
Potentially confounding these observations is the unquantifi- normally active arterial baroreceptors at other sites. They
able effect of alterations in the daily antihypertensive regi- postulated that adaptation of carotid sinus receptors to
mens of hypertensive patients. changes in mechanical properties is slow and incomplete.
Bagshaw and Barrer29 supported this hypothesis by showing
Intraprocedural hypotension and intraprocedural bradycar- that carotid angioplasty resulted in steady-state increases in
dia predicted postprocedural bradycardia. Bradycardia was sinus baroreceptor stimulation and activity. However, the
noted in 7 patients who experienced postprocedural hypoten- NTS is not capable of determining the source of accurate
sion. Four patients with postprocedural hypertension also had blood pressure information, ie, the altered carotid sinus or
bradycardia. Although postprocedural bradycardia was sig- intact aortic and contralateral carotid baroreceptors.29
nificantly more frequent in hypotensive (64%) than in non- Burystyn et al32 suggested that brain-stem centers will not
hypotensive (18.4%) patients, there was no temporal associ- ignore abnormal afferent signals from the reset baroreceptors
ation between blood pressure and heart rate in either group. if their pulsatile nature is intact. Furthermore, aortic barore-
The lack of a temporal association may be attributed to ceptors may have a higher threshold for stimulation than
concomitant administration of inotropic agents, such as do- carotid receptors.33 Our observations suggest that modifica-
pamine, that alter the heart rate or to mediation of heart rate tion of baroreceptor sensitivity after angioplasty does not
or blood pressure responses by separate pathways. Paradox- persist indefinitely, because all hemodynamic complications
ically, prophylactic use of atropine during the procedure was had resolved within 43 hours.
associated with a higher risk of postprocedural bradycardia.
The exact reason underlying this association is unclear. The mean intensive care unit and hospital stays after
Because prophylactic atropine use was based on the discre- angioplasty and stent placement were 2.2Ϯ2.5 and 4.2Ϯ3.5
tion of the physician performing the procedure, there may be days, respectively. The intensive care unit stay was Յ2 days
bias toward use of atropine in patients considered at risk for in 76% of the patients. The hospital stay was Յ4 days in 66%
developing bradycardia. of the patients. The stay was prolonged in a small proportion
of patients because CAS was followed by coronary bypass
Our observations suggest that intraprocedural hemodynam- surgery. Other factors contributing to prolonged stay included
ic instability was an important determinant of postprocedural the high frequency of hemodynamic instability and concur-
hemodynamic complications. Modification of the elasticity rent medical problems in these patients.
(compliance) of the arterial wall, eg, during angioplasty and
stent placement, may alter the sensitivity of carotid barore- The phenomenon of hemodynamic instability after CAS
ceptors. Angioplasty stretches the vessel wall, resulting in has clinical implications. At present, this procedure has been
superficial splitting of the intima and atherosclerotic plaque.8 reserved mainly for patients who are at high risk for CEA
Retraction of the intima and distention of the media results in owing to preexisting coronary artery disease.8,9 Postoperative
a permanent increase in vessel diameter. Bagshaw and hemodynamic instability places additional stress on patients
Barrer29 demonstrated in dogs that angioplasty of nondis- with coronary artery disease. Furthermore, preexistent car-
eased carotid arteries increases the sensitivity of carotid sinus diac rhythm abnormalities may be exacerbated in the post-
baroreceptors. They attributed this increased sensitivity to procedural period. Hemodynamic instability may also in-
changes in the mechanical properties of the carotid sinus, crease the risk of altered perfusion and stroke in the
such as greater compliance and increased diameter for a given postprocedural period. None of our patients with hypotension
MAP. We were unable to observe any effect of diameter of suffered any permanent cardiac or neurological conse-
largest balloon used, ratio of largest balloon diameter to quences. Because of the early and aggressive use of inotropic
and antihypertensive medications, the full severity and neu-

Downloaded from http://stroke.ahajournals.org/ by guest on April 11, 2016

Qureshi et al Hemodynamic Instability After Carotid Angioplasty and Stenting 2093

rological consequences of these hemodynamic events are 14. Rothwell PM, Gibson RJ, Slattery J, Warlow CP. Prognostic value and
undermined in our analysis. Therefore, the potential detri- reproducibility of measurements of carotid stenosis: a comparison of
mental effects of untreated hypotension or hypertension three methods on 1001 angiograms: European Carotid Surgery Trialists’
should not be underestimated despite the lack of clinical Collaborative Group. Stroke. 1994;25:2440 –2444.
consequences in the present study. Our observation also
argues against postprocedural monitoring of patients in set- 15. Hosmer DW, Lemeshow S. Applied Logistic Regression. New York, NY:
tings not suited for intensive cardiovascular monitoring. John Wiley & Sons; 1989.
Furthermore, same-day discharge for patients after CAS may
not be adequate given the risk of hemodynamic instability. 16. Mendelsohn FO, Weissman NJ, Lederman RJ, Crowley JJ, Gray JL,
Phillips HR, Alberts MJ, McCann RL, Smith TP, Stack RS. Acute
Conclusions hemodynamic changes during carotid artery stenting. Am J Cardiol.
The risk of hemodynamic instability after CAS should be 1998;82:1077–1081.
recognized. Intensive care unit management may facilitate the
identification of hemodynamic complications and potentially 17. Qureshi AI, Luft AR, Lopes DK, Lanzino G, Fessler RD, Sharma M,
reduce their severity. Therefore, we recommend postproce- Guterman LR, Hopkins LN. Postoperative hypotension after carotid
dural monitoring of blood pressure and electrocardiographic angioplasty and stenting: report of three cases. Neurosurgery. 1999;44:
changes in patients undergoing CAS, particularly in those 1320 –1323; discussion 1324.
patients who have intraoperative hemodynamic instability.
18. Teitelbaum GP, Lefkowitz MA, Giannotta SL. Carotid angioplasty and
References stenting in high-risk patients. Surg Neurol. 1998;50:300 –312.

1. Bove EL, Fry WJ, Gross WS, Stanley JC. Hypotension and hypertension 19. Waigand J, Gross CM, Uhlich F, Kramer J, Tamaschke C, Vogel P, Luft
as consequences of baroreceptor dysfunction following carotid endarter- FC, Dietz R. Elective stenting of carotid artery stenosis in patients with
ectomy. Surgery. 1979;85:633– 637. severe coronary artery disease. Eur Heart J. 1998;19:1365–1370.

2. Hans SS, Glover JL. The relationship of cardiac and neurological com- 20. Wholey MH, Wholey MH, Jarmolowski CR, Eles G, Levy D, Buecthel J.
plications to blood pressure changes following carotid endarterectomy. Endovascular stents for carotid artery occlusive disease. J Endovasc Surg.
Am Surg. 1995;61:356 –359. 1997;4:326 –338.

3. Satiani B, Vasko JS, Evans WE. Hypertension following carotid endar- 21. Yadav JS, Roubin GS, Iyer S, Vitek J, King P, Jordan WD, Fisher WS.
terectomy. Surg Neurol. 1979;11:357–359. Elective stenting of the extracranial carotid arteries. Circulation. 1997;
95:376 –381.
4. Skudlarick JL, Mooring SL. Systolic hypertension and complications of
carotid endarterectomy. South Med J. 1982;75:1563–1565. 22. Berne RM, Levy MN. The peripheral circulation and its control. In: Berne
RM, Levy MN, eds. Physiology. St. Louis, Mo: Mosby Year Book;
5. Smith BL. Hypertension following carotid endarterectomy: the role of 1993:478 – 493.
cerebral renin production. J Vasc Surg. 1984;1:623– 627.
23. Benarroch EE. Arterial pressure. In: Benarroch EE, ed. Central
6. Tarlov E, Schmidek H, Scott RM, Wepsic JG, Ojemann RG. Reflex Autonomic Network: Functional Organization and Clinical Correlations.
hypotension following carotid endarterectomy: mechanism and man- New York, NY: Futura Publishing Co; 1997:197–229.
agement. J Neurosurg. 1973;39:323–327.
24. Alpman A, Oral D, Guldal M, Erol C, Omurlu K, Berkalp B, Dagal Z,
7. Wong JH, Findlay JM, Suarez-Almazor ME. Hemodynamic instability Akyol T. Cardioinhibitory response to carotid sinus massage in patients
after carotid endarterectomy: risk factors and associations with operative with coronary artery disease. Int J Cardiol. 1993;42:277–283.
complications. Neurosurgery. 1997;41:35– 43.
25. Brown KA, Maloney JD, Smith CH, Haritzler GO, Ilstrup DM. Carotid
8. Brown MM. Balloon angioplasty for cerebrovascular disease. Neurol Res. sinus reflex in patients undergoing coronary angiography: relationship of
1992;14(suppl 2):159 –163. degree and location of coronary artery disease to response to carotid sinus
massage. Circulation. 1980;62:697–703.
9. Criado FJ, Wellons E, Clark N. Evolving indications for and early results
of carotid artery stenting. Am J Surg. 1997;174:111–114. 26. Angell-James JE, Lumley JS. The effects of carotid endarterectomy on
the mechanical properties of the carotid sinus and carotid sinus nerve
10. Roubin GS, Yadav S, Iyer SS, Vitek J. Carotid stent-supported angio- activity in atherosclerotic patients. Br J Surg. 1974;61:805– 810.
plasty: a neurovascular intervention to prevent stroke. Am J Cardiol.
1996;78(suppl 3A):8 –12. 27. Tyden G, Samnegard H, Melcher A, Thulin L. Effect of carotid endar-
terectomy on the antihypertensive properties of the carotid sinus reflex.
11. Theron JG, Ayelle GG, Coskun O, Huet HF, Guimaraens L. Carotid Acta Chir Scand. 1981;147:15–17.
artery stenosis: treatment with protected balloon angioplasty and stent
placement. Radiology. 1996;201:627– 636. 28. Dehn TC, Angell-James JE. Long-term effect of carotid endarterectomy
on carotid sinus baroreceptor function and blood pressure control. Br J
12. Guterman LR, Budny JL, Gibbons KJ, Hopkins LN. Thrombolysis of the Surg. 1987;74:997–1000.
cervical internal carotid artery before balloon angioplasty and stent
placement: report of two cases. Neurosurgery. 1996;38:620 – 624. 29. Bagshaw RJ, Barrer SJ. Effects of angioplasty upon carotid sinus
mechanical properties and blood pressure control in the dog. Neuro-
13. North American Symptomatic Carotid Endarterectomy Trial Collabo- surgery. 1987;21:324 –330.
rators. Beneficial effect of carotid endarterectomy in symptomatic
patients with high-grade carotid stenosis. N Engl J Med. 1991;325: 30. Cowley AW Jr, Liard JF, Guyton AC. Role of baroreceptor reflex in daily
445– 453. control of blood pressure and other variables in dogs. Circ Res. 1973;32:
564 –576.

31. Fadali AM, Walstad PM. Effect of carotid sinus enlargement on systemic
arterial hypertension. Surgery. 1972;71:452– 458.

32. Burystyn PG, Horrobin DF, Lloyd IJ. Chronic hypertension in rabbits
induced by bilateral placement of rigid casts around the carotid sinus
regions. Cardiovasc Res. 1972;6:54 –56.

33. Bagshaw RJ. Evolution of cardiovascular baroreceptor control. Biol Rev
Camb Philos Soc. 1985;60:121–162.

Downloaded from http://stroke.ahajournals.org/ by guest on April 11, 2016

Frequency and Determinants of Postprocedural Hemodynamic Instability After Carotid
Angioplasty and Stenting

Adnan I. Qureshi, Andreas R. Luft, Mudit Sharma, Vallabh Janardhan, Demetrius K. Lopes,
Jehanzeb Khan, Lee R. Guterman and L. Nelson Hopkins

Stroke. 1999;30:2086-2093
doi: 10.1161/01.STR.30.10.2086

Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1999 American Heart Association, Inc. All rights reserved.
Print ISSN: 0039-2499. Online ISSN: 1524-4628

The online version of this article, along with updated information and services, is located on the
World Wide Web at:

http://stroke.ahajournals.org/content/30/10/2086

Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Stroke can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.
Once the online version of the published article for which permission is being requested is located, click
Request Permissions in the middle column of the Web page under Services. Further information about this
process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Stroke is online at:
http://stroke.ahajournals.org//subscriptions/

Downloaded from http://stroke.ahajournals.org/ by guest on April 11, 2016