REMS for the diagnosis of osteoporosis
The multicenter clinical trial
Di Paola, Gatti, Gatti, Porcu, Conversano, Pisani & Casciaro. Osteoporosis International 30(2) (2019) 419–431. doi.org/10.1007/s00198-018-4686-3
1,914
Postmenopausal women across 7 Italian centers
91.7%
REMS sensitivity at the lumbar spine
92.0%
REMS specificity at the lumbar spine
88.8%
Diagnostic concordance with DXA, lumbar spine
Why this trial matters for your bone health
This is the foundational study behind REMS: the large, multi-center trial that established whether the technology could genuinely match DXA in diagnosing osteoporosis — not in a single lab with expert operators, but across 7 Italian centers, with many different operators scanning 1,914 postmenopausal women.
A large study at one site proves a technology can work under ideal conditions. A large study across many sites proves it works in ordinary clinical practice, where operator skill and equipment vary. That distinction is why multicenter trials carry more weight — and why this paper is the reference other researchers cite.
Each woman was scanned with both REMS and DXA at the lumbar spine and femoral neck. The two technologies were then compared on three things that matter clinically: whether REMS correctly identified osteoporosis (sensitivity), whether it correctly cleared patients who didn't have it (specificity), and whether it agreed with DXA's overall classification (concordance).
What it means in practice: REMS isn't merely correlated with DXA — it reaches diagnoses DXA would reach, with sensitivity and specificity both above 91%, in a study large and varied enough to reflect real-world clinical conditions.
The trial also tested precision — how consistent a scanner is when it remeasures the same patient. On this measure REMS outperformed the figures typically reported for DXA, which matters for anyone being monitored over time: a more repeatable measurement means real change is easier to distinguish from measurement noise.
What the trial found
Four findings, in plain language.
- 1
Sensitivity and specificity both cleared 91% at the spine
At the lumbar spine, REMS identified osteoporotic patients with 91.7% sensitivity and correctly cleared non-osteoporotic patients with 92.0% specificity. High sensitivity means few missed cases; high specificity means few false alarms — and REMS achieved both simultaneously.
- 2
The hip performed to the same standard
At the femoral neck, sensitivity was 91.5% with specificity of 91.8% — statistically comparable to the spine results. The technology held its accuracy across both clinically important measurement sites, not just one.
- 3
Diagnostic agreement with DXA was strong at both sites
Total diagnostic concordance — agreement on whether a patient was classified healthy, osteopenic, or osteoporotic — was 88.8% at the lumbar spine and 88.2% at the femoral neck. T-score correlation between the two technologies was r = 0.94 (spine) and r = 0.93 (femur).
- 4
Accuracy held across every age group, and precision beat DXA
Diagnostic accuracy stayed above 90% across all patient age brackets, and older age did not degrade REMS's performance. Separately, intra-operator precision for REMS was substantially better than the values reported in the literature for DXA — meaning more consistent repeat measurements over time.
Study details
- Full title
- Radiofrequency echographic multi-spectrometry for the diagnosis of osteoporosis: a multicenter clinical trial
- Authors
- Marco Di Paola, Annalisa Gatti, Domenico Gatti, Federica Porcu, Francesca Conversano, Paola Pisani, Sergio Casciaro
- Journal
- Osteoporosis International, volume 30, issue 2 (2019), pages 419–431
- Published online
- October 2018
- DOI
- 10.1007/s00198-018-4686-3
- Population
- 1,914 postmenopausal women enrolled across 7 Italian centers
- Comparison
- REMS vs. DXA bone density, osteoporosis classification, and measurement precision at the lumbar spine and femoral neck
- Key result
- Sensitivity 91.7% (spine) / 91.5% (femur); specificity 92.0% / 91.8%; diagnostic concordance with DXA 88.8% / 88.2%
Validated in nearly 2,000 women
A REMS bone density scan takes about 15 minutes, uses no radiation, and rests on a multicenter trial showing accuracy comparable to DXA across 7 centers. Serving patients across New York and New Jersey.
- No radiation at any point during the scan
- Accuracy validated against DXA in 1,914 women
- More repeatable than DXA for tracking changes over time
More published research on REMS technology is collected on the research library page.