Kinematic alignment did not result in statistically significant differences in six-degree-of-freedom knee kinematics compared with mechanical alignment one year following total knee arthroplasty, although patients with certain preoperative characteristics showed findings that supported a more patient-specific approach to alignment.
The single-center, double-blinded trial included 39 patients assigned to mechanical alignment (MA) and 33 assigned to kinematic alignment (KA) in the final analysis, along with nine healthy controls. Patients underwent three-dimensional treadmill gait analysis prior to surgery and one year postoperatively during level walking and walking on a 12% downhill slope. The investigators assessed all six degrees of freedom of knee motion, walking speed, and the Forgotten Joint Score (FJS), and examined outcomes according to Coronal Plane Alignment of the Knee (CPAK) classification and preoperative range of motion (ROM).
At one year, the researchers found no statistically significant differences in knee kinematics between MA and KA during level or downhill walking. Overall gait speed also did not differ between the alignment groups. Both groups continued to show deviations from healthy controls, including reduced knee joint distraction throughout the gait cycle.
Patient-reported outcomes differed between the groups. Mean FJS at one year was 65 with KA compared with 52 with MA. Among patients with CPAK type 1 alignment, mean FJS was 65 with KA and 44 with MA. Preoperative ROM was positively correlated with postoperative FJS among patients receiving KA (r = 0.5), whereas the correlation was negative among those receiving MA (r = −0.3).
Among patients with CPAK type 1 alignment, who had preoperative varus alignment, the researchers found no statistically significant direct kinematic differences between MA and KA. However, patients receiving KA showed smaller kinematic deviations from healthy controls than those receiving MA. Following surgery, gait speed among patients receiving KA was no longer statistically significantly different from healthy controls during level walking. The researchers reported that their initial hypothesis that KA would result in smaller deviations from healthy knee kinematics was supported only in the CPAK type 1 subgroup.
The authors also examined whether preoperative ROM could help distinguish outcomes between alignment strategies. Patients with a composite ROM score greater than 27.8 tended to have better postoperative outcomes with KA than with MA. When stratified by sex, preoperative ROM was positively correlated with postoperative FJS among male (r = 0.6) and female (r = 0.4) patients receiving KA. Among patients receiving MA, the correlation was weakly positive among males (r = 0.1) and strongly negative among females (r = −0.7). Female patients with lower preoperative ROM appeared to have better outcomes with MA, according to the researchers.
Another difference emerged in knee joint distraction. KA showed slightly greater joint distraction than MA, particularly among patients with CPAK type 1 alignment, although the difference was not statistically significant. In this subgroup, mean joint distraction was 43 mm with KA and 37 mm with MA. The researchers noted that the trend toward reduced joint space with MA may have resulted from subtle differences in joint-line restoration or postoperative soft-tissue tension.
The investigators cautioned that the study did not reach its prespecified sample size because of COVID-19–related losses to follow-up. The original calculation called for at least 46 patients per treatment group, compared with the final 39 patients receiving MA and 33 receiving KA. The CPAK type 1 subgroup was smaller, with 13 patients receiving MA and 14 receiving KA. Despite randomization, patients in the MA group had a higher preoperative body mass index, and the healthy controls were younger than both treatment groups.
Other limitations included the use of treadmill walking, which may not reproduce real-world overground walking. The ANOVA post hoc analysis implemented in statistical parametric mapping has also not been formally validated. In addition, eligibility was restricted to patients with a maximum varus deformity of 5°, limiting the generalizability of the findings to patients with more pronounced preoperative varus alignment.
Overall, the trial did not demonstrate statistically significant differences in knee kinematics between MA and KA during level or downhill walking one year following total knee arthroplasty (TKA). However, the findings among patients with CPAK type 1 alignment and the associations between preoperative ROM and postoperative FJS suggested that preoperative characteristics may help identify which patients are more likely to benefit from each alignment strategy.
“The present study represents an initial step toward shifting the focus from a general debate over the 'best' alignment philosophy to a more patient-specific approach,” wrote lead study author Ann-Kathrin Einfeldt, of the Laboratory for Biomechanics and Biomaterials, Department of Orthopaedic Surgery, Hannover Medical School, and colleagues.
Disclosures: The authors declared no competing interests. The study received research funding from Medacta International SA.
Source: Scientific Reports
