Abstract
The purpose of this meta-analysis was to investigate patient and diagnostic parameters influencing the reported rates of recurrent rotator cuff defects after ARCR.
PubMed, EMBASE, Cochrane Library and Scopus databases were searched for clinical studies on tendon defects after ARCR. Imaging modalities, definitions, detection time points, and other known patient risk factors (patient age, tear severity, grade of fatty infiltration, repair technique) as well as reported defect rates were extracted. A meta-analysis of proportion and meta-regression analysis were used to investigate independent variables influencing reported defect rates.
Of 109 articles reviewed, the diagnostic studies used magnetic resonance imaging (MRI) only (n = 56), ultrasound (US) only (n = 28), MRI or computed tomography (CT) arthrography (CTA, n = 14) or a combination of US, MRI and CTA (n = 11) up to 57 months after ARCR. Definitions of tendon defects were highly variable, including those of partial tendon healing with insufficient thickness defined as either an acceptable outcome (n = 72) or a recurrent defect (n = 22). Reported defect rates demonstrated highly significant heterogeneity between studies and groups. Follow-up time and the evaluation of partial tendon healing were independent factors of the defect rate alongside age, tear severity and repair technique. The type of imaging did not significantly alter defect rates.
A number of specific factors significantly alter the rates of rotator cuff defects reported after ARCR. Standardized protocols in clinical practice are required for consistent diagnosis of recurrent defects after ARCR.
IV.