Introduction: Clinical diagnosis of anterior cruciate ligament (ACL) tears (Lachman test and Pivot shift test in valgus and internal rotation) is reliable in case of complete ACL tear but reveals elusive in case of partial tears. Quantitative assessment of anterior tibial translation proves to be imprecise, subjective and poorly reproducible especially with the KT-1000 arthrometer. We developed the GNRB, an alternative original anterior knee laxity measurement device. The lower limb is placed in a rigid support with the knee at 0 degrees of rotation, the restraining power being recorded. A 0-250 N thrust force is transmitted by a jack to the upper segment of the calf. This force is only applied in the absence of hamstring muscles contraction. Displacement of the anterior tibial tubercle is recorded using a sensor with a 0.1mm precision.
Hypothesis: We hypothesize that this knee laxity measurement device is more reliable and reproducible than other currently available arthrometers.
Material and methods: During a first validation study, the GNRB was compared to the KT-1000 arthrometer, in 20 pairs of healthy knees, measurements being performed by two investigators. Variance analyses were carried out at 134 N. In a second clinical study, 21 complete ACL tears (the notch is devoid of ACL) and 24 partial ACL tears (anterior or posterior bundle tear and cicatricial ACL remnant in continuity) were tested with these arthrometers to exact a differential laxity threshold value between both knees at 250 N. Statistical analysis was subsequently performed using variance and ROC curves analysis.
Results: The GNRB arthrometer reveals to be significantly more reproducible than the KT-1000, irrespectively of the tester's experience level. Moreover, unlike the KT-1000, the achieved measurement is independent from the uninvolved side. Reproducibility of laximetry proves to be significantly better with the GNRB than with the KT-1000, wherever the examiner's experience stands and whatever the evaluated side condition could be. When differential laxity threshold value was 3mm in complete ACL tears, sensitivity was 70% and specificity 99% at 134 N. Using a 1.5mm threshold value in ACL partial tears, the arthrometer sensitivity was 80% and specificity was 87% at 134 N.
Discussion: Reproducibility of laximetry was significantly better with the GNRB than the KT-1000 device, wherever the examiner's experience stands and whatever the evaluated side-condition could be. The GNRB reports various supplementary advantages compared with other available laximeters. Good control of the investigated limb position in rotation, recording of translation in the absence of hamstring muscles contraction and in direct comparison with the KT-1000: reproducibility, constant pressure, arthrometry improved accuracy and automated measurements recording. The GNRB might be used for diagnosis of partial and complete ACL tears and during follow-up of reconstructed or not ACL tears.
Level of evidence: type II. Prospective comparative study.