Accuracy of FRAX® in people with multiple sclerosis

Bisson EJ1, Finlayson ML1, Ekuma O2, Marrie RA3, Leslie WD4.

J Bone Miner Res. 2019 Jan 28. doi: 10.1002/jbmr.3682. [Epub ahead of print]

Abstract
People with multiple sclerosis (MS) have a higher risk of low bone mineral density (BMD), osteoporosis, and osteoporotic fractures than healthy adults. The Fracture Risk Assessment tool (FRAX®) has been reported to underestimate fracture risk in people with MS when BMD is unknown. We tested FRAX performance for people with MS when BMD is known, and determined if MS is a risk factor for fracture independent of FRAX score. Using population-based databases in Manitoba, Canada, we identified people with MS who underwent BMD screening after MS diagnosis (n = 744) and controls matched on age, sex and first BMD screening date (n = 3721). We calculated FRAX 10-year probabilities at the BMD screening date, and ascertained incident major osteoporotic fractures (MOF). Using Cox proportional hazards modeling we assessed the effect of MS on the hazard of MOF, adjusting for FRAX 10-year probabilities. MS cases had a higher mean FRAX 10-year probability of MOF calculated with BMD (8.32 ± 7.53) than controls (6.98 ± 5.18; p < 0.01). MS increased the risk for MOF after controlling for FRAX 10-year probability without BMD (HR 1.67, 95% confidence interval [CI] 1.29-2.16), and after controlling for FRAX individual risk factors (HR 1.45, 95%CI 1.12-1.89). MS remained a risk factor for MOF even when controlling for FRAX 10-year probability of MOF with BMD (HR 1.48, 95%CI 1.14-1.92). The FRAX 10-year probability with and without BMD underestimated the observed 10-year MOF risk in MS cases by 3-5%. Calibration improved when secondary osteoporosis was used to calculate FRAX without BMD. Calibration was best when the rheumatoid arthritis input was used to calculate FRAX probability along with BMD. Using secondary osteoporosis or rheumatoid arthritis as proxies for MS improves performance of FRAX and accurately predicts MOF outcomes in those with MS. This provides clinicians with a readily available approach to improve the accuracy of fracture prediction in MS.

Login

Registro | Contraseña perdida?