Low accuracy of the KT-1000 arthrometer versus Telos radiographic measurements to assess knee anterior laxity after ACL graft. Intra and interobserver reproducibility of KT-1000 measurements

Purpose of the study The goal of this study was to compare KT-1000 and Telo s measurements after anterior cruciate ligament reconstruction (ACLR). Material and method Forty eight patients with asymptomatic ACLR (4 failures with positive pivot shift and 12 knees with positive (+) Lachman test) wer e assessed (mean 2.5 years after surgery) by the same examiner by means of: 1) Lachman radiographic with Telos at 150 N, 2) NIEDmetrie(TM) KT-1000 at 69 N, 89 N and maxi-manuel (MM). The examiner tested more than 200 patients each year. Measurements were per formed for KT-1000 according to the manufacturers' recommendations and for Telos according to Staubli. Only side to side differences in millimeters ar e reported. Reproducibility of KT-1000 measurements were also evaluated: in terobserver reproducibility was assessed by 16 examiners on a healthy patie nt, and the experienced examiner tested 20 times a healthy patient. Results An interobserver error of 4 mm range (+/- 2 mm related to 0) was ob served by 12 to 44 p. 100 of the examiners, respectively at 69 N to MM. An intraobserver error of 4 mm range (+/- 2 mm related to 0) was observed in 1 0 p. 100 at MM and in 20 p. 100 at 89 N. Mean side to side laxity with KT-l 000 was 0,93 mm +/- 1.1 [-1 to 5] at 69 N, 1.3 mm +/- 1.6 [- 2 to 6] at 89 N, and 1.41 +/- 1.8 [- 2 to 6] at MM. With Telos the mean side to side laxi ty was 3.95 mm +/- 3.84 [0 to 15]. Significant differences (p = 0,0001) wer e found between measurements obtained by the two methods. No statistical co rrelation could be detected between Values observed by Telos and KT-1000 (R < 0.1), if we consider a 3 mm side to side difference 23 knees (48 p. 100) had abnormal anterior laxity with Telos and with KT-1000 only 3 (6.2 p. 10 0) at 89 N and 6 (12.5 p. 100) at MM (1 (2 p. 100) at 69 N). With a 5 mm si de to side difference, 12 knees (25 p. 100) had abnormal anterior laxity wi th Telos and with KT-1000 only 1 (2 p. 100) at 89 N and 1 (2 p. 100) at MM (0 at 69 N). Only Telos measurements were correlated to positive pivot shif t (p = 0,007) and positive Lachman test (p = 0,01). Conclusion Interobserver reproducibility of KT-1000 measurements was low, b ut improved for intraobserver agreement. However, even for a unique KT-1000 experienced examiner, reliability of KT-1000 was poor when comparing Telos and KT-1000 predictive value to diagnose ACLR failure. Telos results were much more pejorative but the only ones correlated with ACLR failures. We re commend Telos instead of KT-1000 to assess rarity after ACLR.