Fracture Risk Assessment in Canada: How We Got Here and Where We Might Go

Fracture Risk Assessment in
Canada: How We Got Here and

Where We Might Go

William D Leslie MD FRCPC MSc CCD

Departments of Medicine and Radiology

2010 Guidelines

Integrated Management Model Algorithm

10-year Fracture Risk Assessment

FRAX Canada CAROC 2010

2010 Guidelines

Integrated Management Model Algorithm

Initial BMD Testing
Assessment of fracture risk

Low risk Moderate risk High risk
(10-year fracture risk < 10%) (10-year fracture risk 10%-20%) (10-year fracture risk > 20% or
prior fragility fracture of hip or
Unlikely to benefit from Lateral thoracolumbar spine or > 1 fragility fracture)
pharmacotherapy radiography (T4-L4) or
Reassess in 5 yr vertebral fracture assessment Always
may aid in decision-making by consider
identifying vertebral fractures patient
preference
Factors warranting
consideration of pharmacologic Good evidence of
therapy… benefit from

pharmacotherapy

Moderate risk

(10-year fracture risk 10%-20%)

Lateral thoracolumbar radiography (T4-L4) or vertebral
fracture assessment may aid in decision-making by identifying

vertebral fractures

Repeat BMD in Factors warranting consideration of pharmacologic therapy: Good
1-3 yr and • Additional vertebral fracture(s) (by vertebral fracture assessment or evidence
reassess risk of benefit
lateral spine radiograph)
• Previous wrist fracture in individuals aged > 65 or those with from
pharmaco-
T-score < -2.5
• Lumbar spine T-score much lower than femoral neck T-score therapy
• Rapid bone loss
• Men undergoing androgen-deprivation therapy for prostate cancer
• Women undergoing aromatase inhibitor therapy for breast cancer
• Long-term or repeated use of systemic glucocorticoids (oral or

parenteral) not meeting conventional criteria for recent prolonged
use
• Recurrent falls (> 2 in the past 12 mo)
• Other disorders strongly associated with osteoporosis, rapid bone
loss or fractures

2002 Guidelines

Who Should Be Treated for Osteoporosis?

Long-term Personal history Non-traumatic Clinical risk Low
glucocorticoid of fragility fracture vertebral factors DXA BMD
(T-score <−2.5)
therapy after age 40 compression (1 major or 2 minor)
deformities

Start AND
bisphosphonate Low DXA BMD (T-score <−1.5)

therapy

Obtain Consider
DXA BMD therapy
for follow-up

Repeat DXA BMD
after 1or 2 years

Brown JP et al. CMAJ 2002.

10-year Risk Assessment: CAROC

• Semiquantitative method for estimating 10-year
absolute risk of a major osteoporotic fracture* in
postmenopausal women and men over age 50

– Three zones (low: < 10%, moderate, high: > 20%)

• Considers two additional risk factors

– Fragility fracture after age 40
– Recent prolonged systemic glucocorticoid use*

* Fractures of proximal femur, vertebra [clinical], forearm, and proximal humerus

Siminoski K, et al. Can Assoc Radiol J 2005; 56(3):178-188.

10-year Risk Assessment: CAROC

• “Since these are the first Canadian
recommendations integrating clinical risk factors
in a quantitative fracture risk assessment, these
should be considered a work in progress. It is
anticipated that they will be updated periodically
to accommodate advances in this field.”

• “One anticipated area of change is in the choice
of skeletal sites used for determining absolute
fracture risk status, and another will be the
probable addition of further clinical variables…”

Siminoski K, et al. Can Assoc Radiol J 2005; JCD 2007.

FRAX andFORCABXesint P2ra0c0ti8ces in 2008

OC Statement on FRAX --

September 30 2008

• Different FRAX tools are needed for each
country since fracture rates are very different,
even between the United States and Canada.

• Currently, no Canadian FRAX tool exists.
• Therefore, Osteoporosis Canada does not

recommend the use of FRAX in Canada at this
time and has initiated a plan to evaluate all
available fracture risk methods.
• Until assessment completion, it is recommended
that fracture risk continue to be determined using
the 2005 CAR/OC Recommendations for Bone
Mineral Density Reporting in Canada.

FRAX Recommendations for OC --
February 8 2009

1. That WHO generate a Canadian FRAX
model calibrated to the most recent year of
national hip fracture data (2005).

2. That the validity of the model and of the
imputation of non-hip fracture rates be
assessed in specific Canadian cohorts.

3. That a FRAX Educational Strategy be
developed by the ISCD Canadian Panel,
OC, CAR/OAR.

FRAX Updates for OC

1. June 12, 2008 (Toronto, Guidelines/Exec)
2. February 2, 2009 (Toronto, BP/Exec)
3. April 18, 2009 (Montreal, BP/Exec)
4. June 13, 2009 (Board AGM)
5. November 7, 2009 (BP Expert Panel)
6. January 16, 2010 (Toronto, BP)
7. September 9, 2010 (ASBMR, Exec)

Canadian FRAX History

• Canadian FRAX v1.0:

– 2005 CIHI data for hip fracture calibration (40-45, to 85+), 1997
mortality data, hip/4-fracture ratios from Malmo, plateau age 85

• Canadian FRAX v1.1:

– 2005 CIHI data for hip fracture calibration (40-45, to 85+), 2004
mortality data, hip/4-fracture ratios from Malmo, plateau age 85

• Canadian FRAX v2.0:

– 2005 CIHI data for hip fracture calibration (40-45, to 85+), 2004
mortality data, hip/4-fracture ratios from US, plateau age 85

• Canadian FRAX v2.1 (approved):

– 2005 CIHI data for hip fracture calibration (40-45, to 90+), 2004
mortality data, hip/4-fracture ratios from US, plateau age 92 (with
piecewise curve-fitting)

Percent fracture0510152025 30

Sweden Female Age 65 years, prior fracture with femoral neck T-score -2.5 10-Year Major Fracture Probability Canadian FRAX Went Live July 2010
Switzerland
US Caucasian Male

Austria
United Kingdom

CANADA
Belgium
Japan
Italy

Argentina
Hong Kong

Finland
Germany
US Hispanic
US Asian

France
New Zealand

US Black
Spain

Lebanon
China
Turkey

FRAX Limitations in 2010

• The software for the Canadian FRAX tool is not yet widely
available on BMD machines; therefore, for purposes of
reporting BMD, CAROC is the only system that can be
applied on a national basis at the present time. This
situation may change as FRAX software becomes more
widely available. Clinical practitioners need to be aware of
FRAX, given its international importance as the 10-year
risk assessment system developed and recommended by
the WHO and given that many patients will have access
to the FRAX website. Some clinical practitioners may also
prefer the versatility of FRAX, which allows assessment of
risk in the absence of a BMD measurement and is more
accurate for patients with one or more of the additional
risk factors listed above.

2005 CAROC Limitations

2005 CAROC overestimated fracture risk in Canadians

Reasons:
(a) Sweden has highest fracture rates in the world,
(b) Use of minimum T-score for femoral neck T-score
systematically produces higher risk estimates,
(c) Published Swedish fracture probabilities were for an
average individual and included some with the CAROC
clinical risk factors with “double counting”,
(d) The published Swedish fracture probabilities were
from almost twenty years ago and do not reflect secular
decreases in fracture rates that have been observed in
Canada.

Updating CAROC

CAROC is appealing since it is easily summarized
on a pocket card and does not require elaborate
calculations.

It was proposed at the Best Practices Expert Panel
meeting in November 2009 that it might be
possible to update the CAROC risk tables based
upon Canadian FRAX thereby providing a
simpler alternative for individuals that do not
have access to computer-based FRAX
calculations.

Updating the CAROC System

2005 2010
• Simple table/graph • Simple table/graph
• 2 CRFs • 2 CRFs
• Overestimated risk • Accurate risk

– ~1990 Swedish data – 2005 Canadian data
– Double counting – No double counting
– Minimum T-score – Fem neck T-score

• Not validated • Well validated

FRAX vs Updated 2010 CAROC

Same risk category 90% of cases

Low (0-9) CaMos High (20+)

Moderate (10-19)

30%

10-year Fractures 20%

10%

0% 2010CCaAnFRROACX
Full FRAX

Leslie WD and the FRAX Working Group. Osteoporos Int 2010.



METHODS:
DXA registry for Manitoba, 34,060 individuals ≥50 years not receiving
RESULTS:
Mean 9.8 years of follow-up, 3905 individuals sustained fractures.
10 (of 35 total) situations where observed fracture risk fell outside of the
predicted range, and all 10 discordances favoured FRAX. Significant
improvement in risk categorization (NRI overall +0.028, P < 0.001). Within nine
pre-specified subgroups, no case of significantly worse NRI. Number needed to
FRAX (instead of CAROC) to yield an improvement in prediction = 36 (8 with
prior fracture and 4 with prolonged glucocorticoid use).
CONCLUSIONS:
FRAX provides improvement in fracture risk prediction compared with the
simplified CAROC tool in individuals referred for osteoporosis screening,
supporting the use of FRAX as the international reference tool for fracture risk
assessment.

Participants
1 054 815 members aged 50 to 90 years (Israel).
Outcome measure
First MOF (QFracture and FRAX tools) and hip fractures (for all three tools).
Results
AUCs for hip fracture prediction were 82.7% for QFracture, 81.5% for FRAX,
and 77.8% for Garvan. For MOF, AUCs were 71.2% for QFracture and 71.4%
for FRAX. All the tools underestimated the fracture risk, but average observed to
predicted ratios and the calibration slopes of FRAX were closest to 1.
Conclusions
Both QFracture and FRAX had high discriminatory power for hip fracture
prediction, with QFracture performing slightly better. The simpler FRAX
performed almost as well as QFracture for hip fracture prediction.

Advances with FRAX

• Adjustments for:
–Glucocorticoid dose
–Spine T-score
–Trabecular bone score (TBS)
–Hip axis length

Problems with the Current System

• Confusion: CAROC vs FRAX
• Chaos: “Intermediate risk”
• Conflict: CAR, OAR, OC
• Clinical dilemma:

– age 50 and no other CRFs requires
femoral neck T-score of -4.4 or lower;

– lowering the intervention cutoff for the
young massively overtreats the elderly.

Alternatives to the Current System

• Fixed single cutoff (20% MOF)
• Age-specific cutoff (NOGG)
• Age-specific cutoff, fixed after age 70

(new NOGG “hybrid”)
• Age-specific cutoff with lower 10%

(Lebanon “reverse hybrid”)

NOGG System – FRAX Website
(Woman age 65, T-score -2.5)

New NOGG System

Lebanese System - FRAX Website
(Woman age 65, T-score -2.5)

Proposed FRAX System for Canada
• Age-specific cutoff with upper 20%

and lower 10% (“double hybrid”)

Proposed FRAX System for Canada

• Addresses under-treatment in the
young (fixed 10%) and the elderly
(fixed 20%) with a smooth transition
(age-specific NOGG, age 57 to 72)

• Eliminates “Intermediate Risk”
• Sunsets CAROC
• Can be used with and without BMD
• Can be activated on the FRAX website