Diagnosing (severe) osteoporosis by Hologic vs. Lunar measurements: A single-center retrospective study

Abstract

Introduction: Although different dual-energy X-ray absorptiometry (DXA) scanners provide different bone mineral density (BMD) values, there is not a gold standard DXA scanner. T-score is used to facilitate the interpretation of BMD, and osteoporosis (OP) is diagnosed based on T-scores. In this retrospective study, we aimed to evaluate the BMD and T-score differences between Lunar Prodigy and Hologic Horizon DXA scanners.Methodology: Data were collected for patients with previous BMD measurement on Lunar Prodigy and Hologic Horizon DXA scanners within one year in the same medical center.Results: In a total of 55 patients, BMD values of femoral neck/total, and lumbar vertebrae were all lower at Hologic than Lunar (all p < 0.01). The mean T-score difference at the lumbar spine was 0.74 ± 0.42 (p < 0.001). Of the 49 patients diagnosed as OP (T-score ≤−2.5) with the Hologic, the diagnoses were changed for 25 individuals (51.0 %) with Lunar (p < 0.001). Herewith, although the diagnoses of OP did not change by the repeat technique in other 24 patients (49 %), 13 of them (26.5 %) were categorized as having “high fracture risk” instead of “very high fracture risk” group (i.e., T-score <−3.0). We observed moderate-to-good reliabilities (with an intraclass correlation coefficient [ICC] of 0.633–0.878 and 0.733–0.842 for BMD and T-scores, respectively) between measurements with the Lunar and Hologic scanners. Except for one measurement in L3, L4, L1–4 vertebrae, the Bland–Altman plot did not reveal any consistent bias between the measurements of the Lunar and Hologic scanners.Conclusions: The consistency between different DXA scanners (especially for Hologic vs. Lunar) is important for proper management, especially in patients with low T-scores and OP.