Pilot study for the evaluation of tests used for diagnosis of canine cranial cruciate ligament (CCL) failure

Introduction: Failure of the CCL and associated secondary osteoarthritis is the most common cause of chronic hind limb lameness in the dog. With respect to tests used for diagnosis of CCL failure, sensitivity is defined as the percentage of patients known to have CCL failure which tested positive and specificity is the percentage of patients known to be free of CCL failure which tested negative. The purpose of this study was to estimate the sensitivity and specificity of a range of tests and signs used in the diagnosis of canine CCL failure. Materials and methods: Seventeen dogs were recruited. Eleven ‘affected’ dogs–patients with hind limb lameness subsequently confirmed to be due to CCL failure either at arthrotomy or by arthroscopy during subsequen surgical treatment and 6 ‘control’ dogs–patients that had a radiographically normal stifle and were free from signs of pelvic limb orthopedic disease at the time of examination and for at least six weeks subsequently. Only one stifle in each dog was evaluated. The following tests and assessments were made in the conscious patient: cranial drawer, tibial compression, patellar tendon palpation and palpation of the medial aspect of the stifle joint. Under general anesthesia, cranial drawer and tibial compression tests were repeated and a lateral stifle radiograph was collected and evaluated for evidence of change to the infrapatellar fat pad. For each test, a result of ‘Normal’, ‘Abnormal’ or ‘Don’t Know’ was recorded. The results of each investigation were analyzed twice using a 2 2 table method: the first time all ‘Don’t know’ results were counted as if they were ‘Normal’ and the second time the ‘Don’t knows’ counted as if ‘Abnormal’. Discussion: A ‘Don’t know’ group was included to reflect the essential subjectivity of these tests. The sensitivity and specificity of the CDT are not very high, especially in the conscious patient. However, in this pilot study, a definite normal CDT was strongly indicative of a normal CCL. The TCT appears even less sensitive and specific than the CDT but again, anaesthetizing the patient improves the test. Palpation of the medial aspect of the stifle joint cannot be considered a good test because it fails to detect 63.3% of affected dogs though it appears rather better at confirming normality. Our results suggest that patellar tendon palpation and the evaluation of a lateral stifle radiograph were very sensitive and specific tests for the diagnosis of CCL. Affected dogs were confirmed as such by direct inspection of the damaged CCL either at arthrotomy or by arthroscopy at the time of therapeutic surgery. Control dogs were considered normal on the basis of a single radiographic examination plus two normal clinical examinations at least 6 weeks apart. However, because dogs in the control group did not undergo CCL inspection there is a possibility of false negatives. Following on from this pilot study, data collection is ongoing with a target of analyzing data from 50 affected and 50 controls to estimate positive predictable value (PPV) and negative predictable value (NPV) for these test