Reducing the need for central dual-energy X-ray absorptiometry in postmenopausal women: efficacy of a clinical...

We evaluated the efficacy of a triage approach based on a combination of osteoporosis risk-assessment tools plus peripheral densitometry to identify low bone density accurately enough to be useful for clinical decision making in postmenopausal women. We conducted a cross-sectional diagnostic study in postmenopausal Caucasian women from primary and tertiary care. All women underwent dual-energy X-ray absorptiometric (DXA) measurement at the hip and lumbar spine and were categorized as osteoporotic or not. Additionally, patients had a nondominant heel densitometry performed with a PIXI densitometer. Four osteoporosis risk scores were tested: SCORE, ORAI, OST, and OSIRIS. All measurements were cross-blinded. We estimated the area under the curve (AUC) to predict the DXA results of 16 combinations of PIXI plus risk scores. A formula including the best combination was derived from a regression model and its predictability estimated. We included 505 women, in whom the prevalence of osteoporosis was 20 %, similar in both settings. The best algorithm was a combination of PIXI + OST + SCORE with an AUC of 0.826 (95 % CI 0.782-0.869). The proposed formula is Risk = (-12) × [PIXI + (-5)] × [OST + (-2)] × SCORE and showed little bias in the estimation (0.0016). If the formula had been implemented and the intermediate risk cutoff set at -5 to 20, the system would have saved 4,606.34 in the study year. The formula proposed, derived from previously validated risk scores plus a peripheral bone density measurement, can be used reliably in primary care to avoid unnecessary central DXA measurements in postmenopausal women.