DIAGNOSIS OF OSTEOPOROSIS BY DUAL X-RAY AND LASER (DXL) DENSITOMETRY – A HEALTH TECHNOLOGY ASSESSMENT
Author(s)
Joo Sin Beh, BEng, Student, Dagmar Luehmann, MD, Senior ResearcherUniversity of Luebeck, Luebeck, Germany
Presentation Documents
OBJECTIVES: Although cDXA is considered the gold standard for diagnosis of osteoporosis, it requires massive and radiation-protection-intensive radiological settings. Peripheral DXA measurements would be more practically convenient but their results do not correlate very well with that of cDXA. Against this background, DXL was developed. It combines peripheral DXA measurement with a laser-defined region of interest to correct for artefacts. In order to clarify the diagnostic accuracy and hereby the clinical utility, a literature overview was undertaken. METHODS: The systematic literature search covered Medline, Embase (including Embase Alert) and the Cochrane Library. Diagnostic studies comparing the diagnostic accuracy of DXL against cDXA were identified and their methodological qualities assessed using the QUADAS instrument. Furthermore abstract publications were included if they contained sufficient data. It was planned to calculate pooled estimates of diagnostic accuracy parameters on the condition that the studies yielded sufficiently high quality data. RESULTS: Six published trials and two meeting abstracts were appraised. Owing to diversity in the study subjects’ origins, variable risk factor constellations and the inhomogeneity of the reference method (i.e. cDXA model, measurement site, execution procedures and reference population) employed in the studies, we decided on not performing meta-analysis. One abstract publication was invalid for calculating diagnostic accuracy statistics due to data inconsistency. The ranges of sensitivities (54.6%-100.0%), specificities (64.3%-86.0%), positive predictive values (33%-82%), negative predictive values (73%-100%), positive likelihood ratios (2.20-6.25), negative likelihood ratios (0.00-0.54) and diagnostic odds ratios (6.47-infinity) were calculated or confirmed and compared among the studies. Variability was explained by the use of different cut-offs, different reference methods (e.g. cDXA’s measurement sites) and different subject’s populations. CONCLUSIONS: Although the results of some studies appear promising; further refinement of indications, population-specific reference databases and studies to determine valid cut-off are needed before the technology can be successfully introduced into routine care.
Conference/Value in Health Info
2008-11, ISPOR Europe 2008, Athens, Greece
Value in Health, Vol. 11, No. 6 (November 2008)
Code
PMS2
Topic
Clinical Outcomes, Epidemiology & Public Health
Topic Subcategory
Comparative Effectiveness or Efficacy, Disease Classification & Coding
Disease
Musculoskeletal Disorders