Background:​ The limits of passive motion of the tibiofemoral joint, a measure of the laxities in different degrees of freedom, are computed as the bidirectional limits of motion of the tibia relative to the femur about a neutral position. The limits of passive motion are commonly used as a benchmark both when creating and validating computational models and when evaluating laxity before, during, and after total knee arthroplasty (TKA). However, the variability in the limits of passive motion both between and within normal tibiofemoral joints is unknown;​ these are both important to know so that patient-to-patient differences are accounted for when using the limits of passive motion as a benchmark. Also, it is unknown whether the laxities of the normal tibiofemoral joint support a common goal of soft tissue balancing in TKA, gap balancing, to create equal laxities at 0° and 90° of flexion. Accordingly, the objectives of this thesis were to determine (1) how variable the limits are both between knees by determining the range of the internal-external (I-E), varus-valgus (V-V), anterior-posterior (A-P), and compression-distraction (C-D) limits and within a knee by determining whether these limits are predictive of one another, and (2) whether the VV, I-E, D, and A-P laxities (i.e. limits of passive motion) are equal at 0° and 90° of flexion.

Methods:​ The limits of passive motion in I-E (±3 N-m), V-V (±5 N-m), A-P (±45 N), and C-D (±100 N) were measured in ten normal human cadaveric knees at flexion angles from 0° to 120° in 15° increments using a six degree-of-freedom load application system.

Results:​ The largest ranges for the I-E, V-V, A-P, and C-D limits were 10.8° and 16.3°, 2.6° and 1.1°, 5.5 mm and 5.6 mm, and 1.2 mm and 2.9 mm respectively. Seventeen of the twenty-eight pair-wise comparisons between the limits had a correlation coefficient less than 0.5. At 90° of flexion, the mean laxity was fivefold greater in varus (p<0.0001), fourfold greater in distraction (p<0.0001), threefold greater in valgus (p=0.0003), internal (p=0.0002), and external rotation (p<0.0001), and twofold greater in anterior translation (p=0.003) than at 0° of flexion.

Conclusions:​ The wide variability in the limits of passive motion both between and within normal tibiofemoral joints demonstrates that a patient-specific approach, including all degrees of freedom of interest, is necessary when creating and validating computational models and when evaluating laxity of the tibiofemoral joint. The result that six of seven laxities were five to two times greater at 90° than at 0° of flexion in the normal knee calls into question the goal behind gap-balancing TKA.