Accuracy of Ultrasound and MRI in Preoperative and Postoperative Management of Flexor Tendon Injuries: A Systematic Review and Meta-Analysis.

BACKGROUND: Complete and partial flexor tendon lacerations are challenging injuries to diagnose and manage. Imaging modalities can determine grade of laceration, and location of tendon ends preoperatively while detecting presence of adhesions, repair failure, and gap formation postoperatively. Despite these clear advantages, imaging modalities are underutilized because of issues with availability and concerns about accuracy. METHODS: A systematic search of MEDLINE and Embase was conducted to identify papers examining the accuracy of ultrasonography (US) and MRI in preoperative and postoperative management of flexor tendon lacerations. COVIDENCE was used in blinded selection of papers for abstract and full-text review. R Studio was used for meta-analysis of pooled sensitivities and specificities, diagnostic odds ratios, and summary receiver operating curves of both US and MRI. RESULTS: A total of 1197 papers were returned, with 40 being selected after full-text review and 24 being sufficient for statistical analysis. Significant heterogeneity existed for preoperative sensitivity of US and MRI, as well as preoperative specificity of US. MRI was more specific than US in the postoperative period (P < 0.01). Diagnostic odds ratios were >1 for all imaging modalities. The area under the curve for summary receiver operating curves in US preoperative, US postoperative, MRI preoperative, and MRI postoperative were 0.92, 0.81, 0.83, and 0.91, respectively. CONCLUSION: MRI is likely more specific than US in postoperative detection of tendon adhesions, tendon rupture, and gap formation following tendon repair. Notable heterogeneities exist in the literature, highlighting the future need for standardized comparisons of imaging modalities in preoperative management.