Imaging studies done to evaluate chronic ankle instability (CAI) often fail to accurately detail injuries to the anterior talo-fibular ligament (ATFL) and may, therefore, also fail to provide guidance for selecting the most appropriate surgical procedure. Arthroscopy is now an indispensable tool for accurately diagnosing ATFL injuries. This study looked at agreement between arthroscopy and imaging study assessments of ATFL injuries. The primary objective was to adapt an arthroscopic classification of chronic ATFL lesions to the pre-operative imaging study findings in order to estimate the performance of computed tomography (CT)-arthrography, ultrasonography, and magnetic resonance imaging (MRI) in diagnosing ATFL lesions, using arthroscopy as the reference standard.

Agreement between arthroscopic and imaging findings of chronic ATFL injuries can be assessed by using a shared classification developed from the arthroscopic evaluation, used as the reference standard.

A prospective multicentre study was conducted in 286 patients with arthroscopically-treated CAI. In each patient, the arthroscopic assessment of the ATFL was compared to the pre-operative findings by CT-arthrography, ultrasonography, and MRI. A classification of ATFL lesions based on the arthroscopic assessment was used to analyse the imaging studies. Using this classification, two independent observers compared the findings and evaluated the agreement between arthroscopy and imaging studies.

Of the 286 patients, 157 had complete information on the arthroscopic assessment and on pre-operative imaging studies and were included in the analysis. Imaging studies were CT-arthrography in 49 patients, ultrasonography in 63 patients, and MRI in 45 patients; both ultrasonography and MRI were performed in 3 patients. Agreement with arthroscopy was 82% and 88.5% for CT-arthrography, 66.7% and 76.2% for ultrasonography, 70.5% and 79.5% for MRI, and 73.4% and 81.2% for all imaging studies pooled.

Arthroscopy plays a crucial role in the definitive assessment of ligament lesions in patients with CAI, as it supplies far more accurate information than any of the current imaging studies and, in addition, provides a dynamic evaluation of the ligaments and assesses mechanical strength. In the study population, a simple arthroscopic evaluation consistently visualised ATFL lesions, thereby either correcting or confirming the pre-operative imaging study findings. Importantly, the arthroscopic assessment provided more accurate information on the lesions and quality of the ATFL compared to the imaging studies. Arthroscopy has improved our knowledge of chronic ATFL lesions and allowed the development of a simple but accurate four-grade classification of direct relevance to choosing the optimal therapeutic procedure. This classification must be disseminated among radiologists to improve the diagnostic performance of pre-operative imaging studies, assist surgeons in selecting the most appropriate ankle-stabilising procedure, and improve patient information. Further studies are needed to confirm the promising results reported here and the usefulness of this common arthroscopy and imaging-study classification for chronic ATFL lesions.

II.