Cutoff point, upper and lower QUS parameters values in diagnosing demineralisation in cases with ESRD
Dana Bucuras, Dan Poenaru & Ovidiu Golea
Background: The British Society for Osteoporosis proposed a method for comparing different diagnostic measurements for BMD. It defines the upper and the lower limit for the investigated assay, that will identify osteoporosis with a sensitivity of 90% (the upper limit) and a specificity of 90% (the lower limit), compared with the defined golden standard method, mainly DXA for spine or hip. This rationale generate small number of false positive or negative results.
Materials and methods: One hundred and thirty-one cases with ESRD were evaluated by means of DXA and QUS with ISCD validated devices: Hologic Sahara/Hologic Hologic, aparat Delphi W (S/N 70489), lumbar L1L4, anteroposterior technique, nondominant hip.
Table 1 Upper limit, lower limit and threshold value for QUI compared with DXA results
|Upper limit (90% sensitivity)||Lower limit (90% specificity)||Threshold values|
|T score < −1 femural neck||93.90||65.50||74.40|
|T score < −2.5 femoral neck||76.90||61.40||69.60|
|T score < −1 total hip||93.90||67.20||76.10|
|T score < −1 total hip||76.50||59.20||69.60|
|T score < −1 spine||94.50||69.70||80.90|
|T score < −1 spine||85.70||62.20||67.50|
Results: We defined bone demineralisation in cases of T score below −1, confirmed by increased markers of bone turnover. We identified than the limits for each QUS parameter in diagnosis bine demineralisation. We also performed ROC diagnostic curves, comparing QUS parameters with DXA results, and we identified the best threshold value for QUI.
The best parameter seemed to be QUI. We selected different values for each measured sites. Table 1.
The threshold values were identified by finding the best specificity and sensitivity from the series of number. This values are listed in Table 2.
Table 2 Diagnostic quality and power for each identified QUI threshold value
The most precise interval is that one compared with the femoral neck compartment. When a patient has QUS result between the to limits, there is a very good sensitivity and specificity in diagnosing bone demineraşlisatoin. QUS can be used as a screening tool, for decreasing the unnecessary DXA measurements, also in these patients. The 9090 approach seems to be better then the threshold value method, because of the higher NPV and PPV.
Conclusion: QUS can be used as a screening method to identify the target population with decreased bone mass, also in patients with ESRD.