The examination is performed with the patient in supine position with the hip at 45° flexion and the knee at 90° flexion. The lower leg is held in neutral rotation with the examiner using either their thigh to stabilize the patient’s forefoot or forearm to stabilize the patient’s shin. The examiner then places the fingers of both of their hands in the popliteal fossa with the thumbs over the tibial plateau and anterior joint line. A slow, but firm, anteriorly directed force is then applied to the proximal tibia. A positive test result is an increased anterior tibial translation of the involved leg in comparison to the uninvolved leg, and this is indicative of an ACL tear.

The anterior drawer test has the benefit of being the least challenging test to perform; however, there are several limitations to the test that may lead to inaccurate diagnoses. The first flaw arises from the occasional difficultly placing the knee at 90° flexion due to muscle guarding and effusion that often accompany ACL injuries. Furthermore, when at 90° flexion the hamstring muscles may act to stabilize the tibia, preventing anterior translation. Lastly, the anatomy of the knee joint when flexed at 90° is such that the convex surface of the posterior femoral condyle and the concave surface of the medial tibia plateau and posterior horn of the medial meniscus may interact in a manner similar to a “door stopper” which impedes anterior translation. Each of these potential flaws, due to the nature of the examination placing the knee at 90° flexion, could increase the likelihood of a false-negative result when performing the anterior drawer test [14]. False-positive results with the anterior drawer test may arise from PCL injuries where the sagging tibia is taken as the normal position of the proximal tibia causing a movement to normal position to appear as an anterior translation [9]. In order to avoid such errors, it is important to consider the “step-off” which is the shortest distance from the femur to a hypothetical line extending tangentially from the tibial tuberosity. A “step-off” less than 5 mm is indicative of a PCL injury, and any anterior movement from this position is not considered a positive anterior drawer test [8].

One meta-analysis of eight studies (n = 1,061) reported a pooled sensitivity of 62 % (95 % confidence interval [CI], 42–78 %) [13]. A second meta-analysis of 20 studies published before the year 2000 (including 7 of the 8 studies from the previous meta-analysis) with a total sample size of 1,809 patients reported a pooled sensitivity of 55 % (95 % CI, 52–58 %) [1] (Table 5.1). The former review also calculated a pooled specificity of 88 % (95 % CI, 83–92 %) based on a meta-analysis of seven studies (n = 929) [13] (Table 5.2). The latter review performed a meta-analysis of 12 studies (half of which were included in the previous meta-analysis analysis) with a total sample size of 1,420 patients and reported a pooled specificity of 92 % (95 % CI, 90–94 %), for a pooled positive likelihood ratio (LR+) of 7.3 (95 % CI, 3.5–15.2) and a pooled negative likelihood ratio (LR−) of 0.5 (95 % CI, 0.4–0.6) [1]. Another systematic review reported variable LR+ and LR− with ranges of 2.0–87.9 and 0.23–0.74, respectively; however, the authors of the review did not calculate pooled values because of the substantial heterogeneity of the data, wide confidence intervals in the reported values, and concerns of high bias risk in five of the six studies included in the review [15].