Abstract / Summary
Abstract Purpose To determine whether abnormal baseline magnetic resonance imaging (MRI) or diagnostic ultrasound in athletes asymptomatic in the imaged region predicts subsequent same-region symptomatic injury. Methods A multi-database literature search was screened against prespecified criteria. Eligible studies imaged asymptomatic athletes before longitudinal clinical follow-up and reported an abnormal-imaging denominator. Risk of bias used the Quality In Prognosis Studies framework. Achilles and patellar tendon risk ratios (RRs) were pooled separately using exploratory DerSimonian–Laird random-effects models. Results Of 979 screened records, 78 reports were sought, 61 were assessed, and 16 reports representing 11 cohorts were included; four reports contributed six structure-specific estimates. In collegiate athletes, abnormal imaging predicted time-loss injury of the patellar tendon (adjusted RR 10.7, 95% confidence interval [CI] 4.1–27.8), Achilles tendon (adjusted RR 19.2, 95% CI 6.6–55.7), and plantar fascia (adjusted RR 13.3, 95% CI 1.4–104.8). Pooled RRs were 5.43 (95% CI 1.28–23.13; I²=81.1%) for Achilles outcomes and 3.15 (95% CI 1.11–8.94; I²=83.4%) for patellar outcomes. Youth medial-elbow abnormality also predicted injury (17/59 vs 18/151; crude RR 2.42, 95% CI 1.36–4.31). MRI findings were less consistent. Conclusion Baseline imaging abnormalities identify higher-risk groups, most consistently on Achilles and patellar tendon ultrasound, but prediction is heterogeneous and many abnormal structures remain asymptomatic. Imaging should inform monitoring and contextual load management, not serve alone as an indication for restriction or invasive treatment.