Shifts in Diagnosing and Treating Osteoporosis
the gap between those at risk for fracture and those being treated, ultimately reducing the incidents of fracture. Using the WHO’s Fracture Risk Assessment Tool (FRAX)8
and the new NOF guidelines,2 only 26% of
US Caucasian women with a high risk for fracture receive treatment.7 These data suggest that diagnosis by BMD alone leaves a large population untreated and at risk for fracture.
Risk Factor Assessment Tools
FRAX was developed by the WHO to calculate the 10-year absolute risk for any major osteoporotic fracture and hip fracture using clinical risk factors. The risk factors for FRAX and other risk assessment tools are noted in Tables 1 and 2. Age, gender, BMD (or BMI) and a history of fracture are the common risk factors across the assessment tools. The models were developed from 12 international cohorts and have been validated in an additional 11.9
Probability of fracture is estimated using country-specific incidence rates of risk factors.
FRAX can be used to distinguish among individuals with low bone mass those who are at high risk and those at low risk for fracture. The probability of fracture can be used to establish a threshold for treatment. In the US, NOF guidelines suggest that individuals with low bone mass and a FRAX score >3% for risk for hip fracture or >20% risk for osteoporotic fracture are at high risk and should be considered for treatment.2 vary by age.4
In the UK, the threshold limits suggested by NOGG guidelines
The 20% US threshold is equalled at 65 years of age in the UK. In a retrospective cohort analysis comparing subjective judgment of risk factors by clinicians to the NOGG FRAX score threshold, determined without BMD, there was 72.9% conformity in treatment decisions, providing validity to the threshold.10
FRAX does have its limitations. Most importantly, its proper use depends on establishing interventional thresholds11
based on the analysis of risk
and economic factors of a country. The use of FRAX is limited to untreated patients, does not include traditional risk factor variables, does not account for dose–response relationships of risk factors (i.e. glucocorticoid use, smoking, alcohol intake and number of fractures), does not account for patient ascertainment of rheumatoid arthritis and has a poor definition of secondary osteoporosis.9
It is also a complex, Internet-based
algorithm and, therefore, its use is restricted, given the limited Internet access in patient exam rooms, even in industrialized countries.9
There might also be racial or ethnic limitations to its use. FRAX makes the assumption that the relationship between BMI and mortality for Asians, African-Americans and Hispanics is similar; however, there are no data to support this assumption. Furthermore, FRAX assumes that the variability in fracture rates across ethnic groups is similar. Data reports show that hip fracture rate variability in Hispanics is greater than that in Caucasians. When calculating fracture risk, if BMD is absent, fracture risk for African-Americans and Hispanics is underestimated. This suggests that there are ethnic differences that influence fracture risk that are not utilized in FRAX.12
Other widely used alternatives for fracture risk assessment tools have been developed. The Dutch algorithm developed by Pluijm et al.,13 guidelines by the COSC3
and the Nguyen nomogram14 use risk factors that are easily measurable and sufficiently prevalent (see Table 1). They are US MUSCULOSKELETAL REVIEW
also simpler to use compared with FRAX. The Pluijm algorithm and Nguyen nomogram both measure the 10-year risk for fracture, as does FRAX. However, these tools have not been tested and validated outside of their original cohorts, as FRAX has. Their intervention thresholds are limited to their country of origin: The Netherlands for the Pluijm algorithm and Australia for the Nguyen nomogram.
The algorithm developed by Pluijm et al. relies on age, previous fractures since 50 years of age, weight, use of a walking aid and smoking. The threshold limit for those considered at high risk is also dependent on age. Their data suggest that the secondary risk factors of family history, alcohol use and rheumatoid arthritis are not as important and can therefore be omitted.13
A comparison of FRAX scores with those of the
Pluijm model based on a sample of women in Guatemala showed that the threshold limits for treatment are similar. The Pluijm evaluation of the women showed conformity of 87.2 and 83.3% compared with the US Hispanic and Mexico FRAX thresholds for risk for fracture, respectively.15 The Pluijm algorithm has similar limitations to FRAX in terms of dose relationship and does not factor in height or BMI with weight.
The nomogram developed by Nguyen predicts the five- and 10-year absolute fracture risk in Australia using age, fracture history, fall history and BMD T-score or bodyweight.14
scores with those of FRAX in Polish patients showed a mean conformity 21
Table 2: COSC Factors that Identify Individuals Who Should be Assessed for Osteoporosis3
Major Risk Factors Age >65 years of age
Vertebral compression fracture
Fragility fracture after 40 years of age Family history of fracture Glucocorticoid therapy Malabsorption syndrome Primary hyperparathyroidism Propensity to fall
Osteopenia on X-ray
Hypogonadism Early menopause
COSC = Council of the Osteoporosis Society of Canada.
Minor Risk Factors Rheumatoid arthritis
History of clinical hyperthyroidism Chronic anticonvulsant therapy Low dietary calcium Smoker
Excessive alcohol intake Weight <57kg
Weight loss >10% of weight at 25 years of age
Chronic heparin therapy
Table 1: Clinical Risk Factors used in Assessment Tools FRAX8
Age Gender BMI or BMD
History of fracture Smoking
Parent with fractured hip
Glucocorticoids Rheumatoid arthritis Alcohol use
Secondary osteoporosis *Assessment tool developed for women only. BMI = body mass index; BMD = bone mineral density; FRAX = Fracture Risk Assessment Tool.
Pluijm13 Age
Gender* Weight <64kg
History of fracture Smoking
Use of walking aid
Nguyen14 Age
Gender Weight or BMD
History of fracture Number of falls
A comparison of the nomogram
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68