The Predictive Accuracy of PRE-BMD FRAX Scoring in Identifying the Need for Treating Osteoporosis in Patients with Coeliac Disease

Abstract

Purpose: Patients with coeliac disease are at increased risk of osteoporosis. Current guidance recommend the use of Dual-Emission X-ray absorptiometry (DEXA) scans at diagnosis of coeliac disease in order to assess for osteoporosis and guide the need for preventative therapy. The WHO fracture risk assessment web tool can be used with or without bone mineral density (BMD) to calculate a FRAX scores (10-year probability of major osteoporotic fracture and hip fracture). Patients are assigned, based on their pre-BMD FRAX outcome into low, intermediate (requiring BMD measurement) and high (requiring preventative therapy) risk groups in accordance with guidelines from the National Osteoporosis Guidelines Group (NOGG). Our study aimed to assess the accuracy of Pre-BMD FRAX scores in predicting need for preventative therapy in a coeliac cohort. Methods: We conducted a retrospective review of 70 biopsy proven coeliac disease patients who had undergone an index DEXA scan whilst attending follow up at our hospital between 2007 and 2011.Clinical data including demographics, DEXA scans was obtained from case note and electronic patient record review. Information regarding previous history of fractures and history of hip fracture in patient parents was obtained by direct interview. Pre and post-BMD FRAX assessment scores were calculated retrospectively. All coeliac patients were considered to have secondary osteoporosis as a risk factor for FRAX calculation. Pre-BMD FRAX risk status was compared with Post-BMD FRAX outcome to identify patients requiring and not requiring therapy. The study was approved by our Institutional Review Board. Results: Seventy biopsy proven coeliac patients (female 51) with a median age of 54 were analysed. Pre-BMD FRAX risk status (low and intermediate/high risk) had a sensitivity of 82 % (95 % CI 73-91) in correctly identifying 9 out of 11 patients as needing treatment. Specificity was 76% (95 % CI 66-86) identifying 44 out of 58 patients as not needing treatment. The positive predictive value (PPV) was 39% (95 % CI 28-50) with 9 out of 23 patients correctly identified as needing treatment but the negative predictive value was high 96% (95% CI 91-100) as 44 patients out 46 were correctly identified as not needing treatment. Specificity and PPV results were influenced by the intermediate risk Pre-BMD FRAX sub-group of which 12 (75%) required lifestyle advice and 4 (25%) required preventative therapy following post-BMD FRAX assessment. Conclusion: In patients with coeliac disease, pre-BMD FRAX risk status is sensitive as well as specific in predicting need for osteoporosis treatment and can help guide clinical decisions potentially avoiding DEXA scanning if appropriate.