Quantitative evaluation of the pivot shift by image analysis using the iPad

Abstract

To enable comparison of test results, a widely available measurement system for the pivot shift test is needed. Simple image analysis of lateral knee joint translation is one such system that can be installed on a prevalent computer tablet (e.g. iPad). The purpose of this study was to test a novel iPad application to detect the pivot shift. It was hypothesized that the abnormal lateral translation in ACL deficient knees would be detected by the iPad application. Thirty-four consecutive ACL deficient patients were tested. Three skin markers were attached on the following bony landmarks: (1) Gerdy's tubercle, (2) fibular head and (3) lateral epicondyle. A standardized pivot shift test was performed under anaesthesia, while the lateral side of the knee joint was monitored. The recorded movie was processed by the iPad application to measure the lateral translation of the knee joint. Lateral translation was compared between knees with different pivot shift grades. Valid data sets were obtained in 20 (59 %) ACL deficient knees. The remaining 14 data sets were invalid because of failure to detect translation or detection of excessive translation. ACL deficient knees had larger lateral translation than the contra-lateral knees (p < 0.01). In the 20 valid data sets, which were graded as either grade 1 (n = 10) or grade 2 (n = 10), lateral translation was significantly larger in the grade 2 pivot shift (3.6 ± 1.2 mm) than the grade 1 pivot shift (2.7 ± 0.6 mm, p < 0.05). Although some technical corrections, such as testing manoeuvre and recording procedure, are needed to improve the image data sampling using the iPad application, the potential of the iPad application to classify the pivot shift was demonstrated.