Accuracy of diagnostic ultrasound in patients with suspected subacromial disorders: a systematic review and meta-analysis.
Ottenheijm RP., Jansen MJ., Staal JB., van den Bruel A., Weijers RE., de Bie RA., Dinant G-J.
OBJECTIVE: To determine the diagnostic accuracy of ultrasound for detecting subacromial disorders in patients presenting in primary and secondary care settings. DATA SOURCES: Medline and Embase were searched on June 9, 2010. In addition, the reference list of 1 systematic review and all included articles were searched to identify relevant studies. STUDY SELECTION: Two reviewers independently selected the articles evaluating the accuracy of ultrasound for detecting subacromial disorders from the title and abstracts retrieved by the literature search. Selection criteria were ultrasound frequency greater than or equal to 7.5MHz as index test, surgery, magnetic resonance imaging and/or radiography as reference standards, and subacromial disorders as target conditions. DATA EXTRACTION: Two reviewers independently extracted the data on study characteristics and results to construct 2 by 2 tables and performed a methodologic quality assessment. DATA SYNTHESIS: Twenty-three studies were included: 22 reported on full-thickness rotator cuff tears, 15 on partial-thickness tears, 3 on subacromial bursitis, 2 on tendinopathy, and 2 on calcifying tendonitis, respectively. For full-thickness tears, pooled sensitivity of ultrasound was .95 (95% confidence interval, .90-.97), and specificity .96 (.93-.98). For partial-thickness tears, pooled sensitivity was .72 (.58-.83), and specificity .93 (.89-.96). Statistical pooling was not possible for the other disorders. For subacromial bursitis, sensitivity ranged from .79 to .81, and specificity from .94 to .98. For tendinopathy, sensitivity ranged from .67 to .93, specificity from .88 to 1.00. Sensitivity for calcifying tendonitis was 1.00 in both studies, with specificity ranging from .85 to .98. CONCLUSIONS: We strongly recommend ultrasound in patients for whom conservative treatment fails, to rule in or out full-thickness tears, to rule in partial-thickness tears, and to a lesser extent to diagnose tendinopathy, subacromial bursitis, and calcifying tendonitis. These results can help physicians tailor treatment.