Diagnostic accuracy of clinical tests assessing ligamentous injury of the ankle syndesmosis: a systematic review with meta-analysis

Objective
To summarise and evaluate research on the diagnostic accuracy of clinical tests for ligamentous injury of the ankle syndesmosis.

Methods
CINAHL, Embase, and MEDLINE were searched from inception to February 12, 2021. Studies comparing clinical examination to arthroscopy, magnetic resonance imaging, or ultrasound were considered eligible. Meta-analysis was based on random effect modelling and limited to studies fulfilling all QUADAS-2 criteria. Sensitivity (SN), specificity (SP) and likelihood ratios determined diagnostic accuracy, all with 95% confidence intervals (CI).

Results
Six studies were included (512 participants; 13 clinical tests; 29% median prevalence). No individual test was associated with both high sensitivity and high specificity. Tests with the highest sensitivity were: palpation [SN 92% (95%CI 79–98)] and dorsiflexion lunge [SN 75% (95% CI 64–84%); n = 2 studies]. Tests with the highest specificity were: squeeze test [SP 85% (95% CI 81–89%); n = 4 studies] and external rotation [SP 78% (95% CI 73–82%); n = 4 studies].

Conclusions
Clinical examination should involve initial clustering of tests with high sensitivity (palpation; dorsiflexion lunge), followed by a test with high specificity (squeeze). However, as these tests cannot definitively stratify syndesmotic injuries into stable vs unstable, decisions on optimal management (conservative vs surgery) require additional imaging or arthroscopy.