A prospective study was performed on 50 patients who were thought or suspected to have cruciate ligament tears. Each patient had a clinical examination preoperatively and under anesthesia and instrumented examination using the MedMetric KT-1000 arthrometer, Stryker knee laxity tester, and Genucom knee analysis. The diagnosis was confirmed by arthrotomy or arthroscopy. Thirty had acute knee injuries (within 2 weeks), 9 had subacute (2 weeks to 3 months), and 11 had chronic injuries. The surgical findings demonstrated that five had other maladies, but no cruciate ligament tears. Two had a partial ACL tear, and the remaining 43 patients had at least one cruciate tear. The preoperative clinical examination for cruciate ligament integrity was completely correct in 92%, correct but incomplete in 6%, and incorrect in 2%. Examination under anesthesia was correct in 98%, the KT-1000 was correct in 75% (involved knee minus noninvolved knee was greater than or equal to 3 mm laxity equal to ACL tear), the Stryker was correct in 75%, and Genucom in 70%. The average laxity of those with ACL tears was 4.4 mm with the KT-1000, 4.6 mm with the Stryker, and 2.0 mm with the Genucom. The methods of testing with the Stryker and the KT-1000 are similar and the laxity recorded in patients with a torn ACL were almost identical. However, the KT-1000 can be used to identify the quadriceps neutral position and therefore more accurately determine PCL instability. The Genucom has the most versatility, but the laxity recorded in patients with a torn ACL was significantly lower than the other devices.(ABSTRACT TRUNCATED AT 250 WORDS)