Prevalence of shoulder labral abnormalities on MRI in a non-athletic asymptomatic young adult population.

Abstract

To determine the prevalence of shoulder (specifically labral) abnormalities on MRI in a young non-athletic asymptomatic cohort. We hypothesize that this population will have fewer labral abnormalities than an athletic population. In this cross-sectional study, non-athletic young adults age 18-29 with no history of shoulder pain received bilateral shoulder MRIs. A total of 58 total shoulder MRIs were completed on a 3-T MRI scanner (PRISMA-Fit Siemens Medical). MRIs were read by two board-certified fellowship-trained musculoskeletal radiologists at two time points 3 months apart to determine prevalence of labral and other shoulder anatomy abnormalities. Kappa statistics and the associated 95% confidence intervals were computed for inter/intra-reader reliability. Fisher's exact test was used to compare rates of abnormalities in our study with a similarly designed study involving ice hockey athletes. Prevalence of labral abnormalities was 9% (5/58). Kappa coefficient was 1.0 for both readers for intra-reader reliability and 0.57 for inter-reader reliability of labral abnormalities. We further compared our results in asymptomatic athletes with previously published work using the same protocol at our institution. The prevalence of labral abnormalities on MRI in asymptomatic professional and collegiate ice hockey players (49 imaged shoulders) was 24%, which demonstrated a statistically significant (p value < 0.05) difference compared with our data with a p value of 0.03. Non-athletic young adults with no history of shoulder pain/injury had an overall prevalence of shoulder MRI abnormalities less than asymptomatic professional and collegiate ice hockey players in a similarly designed study.