MRI-Based Classification of Achilles Tendon Ruptures: Reliability and Predictors of Tear Location

Background: The absence of a comprehensive classification system for Achilles tendon (AT) injuries contributes to the lack of consensus regarding optimal treatment strategies, as it hinders accurate differentiation of injury types. Purpose: To evaluate epidemiology of AT ruptures to develop a magnetic resonance imaging–based classification system for rupture location. Study Design: Cross-sectional study; Level of evidence, 3. Methods: A search was conducted for all patients with acute AT injuries between 2016 and 2023. Tear locations were classified as follows: type 1 (musculotendinous junction; MTJ), type 1A (2-cm zone distal to MTJ), type 2 (midsubstance), type 3 (2-cm zone proximal to insertion), type 3A (soft tissue “peel-off”), type 3B (calcaneal bony avulsion). Clinicodemographic data were collected, and rupture gap size, tendon thickness, tibiotalar, and plantarflexion angle were measured. Results: A total of 428 patients were included (age 45 ± 15 years; 17% female). Mean rupture gap size was 30 ± 16 mm in 73°± 8° tibiotalar flexion. AT tear distribution: type 1, 62.7%; type 1A, 16.3%; type 2, 13.8%; type 3, 0.7%; type 3A, 5.8%; and type 3B, 0.7%. Female patients demonstrated lower tendon thickness ( P < .001) and a significantly higher prevalence of type 3, 3A, and 3B ( P < .001). Multivariate analysis revealed that proximal AT tears were associated with presence of tendinopathy (odds ratio [OR], 2.19; 95% CI, 1.24-3.83), while younger age and larger gap size were significant but with a weak effect size. Midsubstance tears were predicted by older age and increased tendon thickness (OR, 1.02-1.17). Distal tears were associated with absence of tendinopathy (OR, 0.17; 95% CI, 0.07-0.39), while age only had a small effect. Tear location measurements demonstrated substantial to almost perfect interobserver (3 observers; n = 35; Cohen's kappa = 0.93; 95% CI, 0.74-0.99) and intraobserver (1 observer; n = 35; Cohen's kappa = 0.93; 95% CI, 0.86-0.99) reliability. Conclusion: This study introduces an AT tear classification system that demonstrated substantial to almost perfect reliability. AT ruptures at the MTJ were most common (63%). Presence of tendinopathy, age, gap size, and tendon thickness were significant predictors for AT tear type. Female patients exhibited lower tendon thickness and a significantly higher prevalence of distal tear location. Studies with larger cohort sizes are necessary to confirm the current findings.