It has been a long-held tenet of total knee replacement that the restoration of a neutral mechanical axis promotes improved durability and function following this surgery. Observational data from over ten series have led to this belief and have in fact been corroborated by in vitro testing, in which increased wear and premature failure of components can occur if the total knee replacements are mechanically aligned in a varus position1-14. This concept is further reinforced by the advances in computerized navigation and other imaging systems that aim for this "ideal" coronal plane alignment to enhance knee replacement survival rates.
The report by Parratte and associates entitled "Effect of Postoperative Mechanical Axis Alignment on the Fifteen-Year Survival of Modern, Cemented Total Knee Replacements" challenges these assertions by finding that total knee replacements placed outside of a few degrees from the mechanical axis in the coronal plane had equal survival rates when compared with knees placed within ±3° of the mechanical axis. The authors analyzed the fifteen-year survival for 398 primary total knee arthroplasties that had been performed with cement in 280 patients between 1985 and 1990. Patients were stratified into two groups: those with knees that were aligned well in the mechanical axis (292 knees with a mechanical axis of 0° ± 3°) and those with knees that were not as well aligned (106 knees with a mechanical axis beyond 0° ± 3°). Survival of the knee replacements was analyzed according to three end points: (1) revision for any reason, including revision for septic and aseptic complications, (2) revision because of mechanical failure, aseptic loosening, radiographic wear, or patellar complications, and (3) revision because of mechanical failure, aseptic loosening, or radiographic wear, with exclusion of patellar complications. There was no difference in survival between the mechanically aligned group and the outlier group in terms of any of these end points. The authors concluded that this postoperative coronal plane mechanical axis dichotomous variable did not influence outcome. They did not assert that this alignment variable is completely irrelevant, but rather that it might not be as important a factor in terms of outcome as previously thought. The authors found that patient age and body-mass index had an effect on outcome and postulated that other factors such as gait may have an impact on total knee survival.
This is a well written evaluation of the observed effect of alignment on the durability of total knee replacements and is the only published series involving the use of modern cemented total knee designs with both preoperative and postoperative full-length radiographs available for 95% of the patients. Long-term follow-up (fifteen years) was available through chart review and telephone interviews conducted with patients without recent clinic visits. The statistical analysis was conducted in an appropriate fashion, and the conclusion is reasonable.
The importance of this work is that surgeons as well as companies that propagate the belief that simply achieving coronal plane alignment within a few degrees will lead to the best survival of total knee replacements may be incorrect. In addition, the use of computerized navigation systems to achieve ideal coronal plane alignment can be more expensive and time-consuming and may not necessarily improve the survival of total knee replacements.
We do, however, question the premise that both varus and valgus knees could be treated equally in this observational study. It is our opinion that patients whose knees are in varus alignment postoperatively are at higher risk for failure than patients whose knees have valgus alignment. This could be equated with an analysis of the health risks of obesity in which patients of "normal weight" are compared with patients of "abnormal weight," with underweight and overweight patients being combined in the "abnormal" group, thus diluting the well-recognized health risks seen in overweight individuals. In this case, the authors do not reveal what the failure rates were within the subgroups, potentially leading to a misinterpretation of the result. For example, in the most extreme case, there could have been fourteen revisions in the varus alignment group (32%) and zero revisions in the valgus alignment group (0%). Recognition of this limitation may have prompted the authors’ closing statement: "a neutral mechanical axis remains a reasonable target and should be considered as the standard for comparison if other alignment targets are introduced."
This report challenges previously held beliefs regarding the desirability of achieving a neutral mechanical axis in every patient undergoing total knee reconstruction and underscores the importance of exploring this and other total knee alignment or positioning issues. We do believe that to look at the multitude of other factors such as different degrees of outliers, varus or valgus outliers, rotational and translational positioning, and patient demographics would require a much larger study to be statistically valid and to draw major conclusions.
In summary, the authors found that total knee replacements in the outlier group had similar survival to those in the mechanically aligned group. This is an extremely important study that challenges previously accepted notions about the importance of coronal plane alignment, which may not be true. It also should foster further work on larger groups of patients in terms of the analysis of optimal total knee alignment for individual patients as well as other factors (i.e., gait) that might influence implant survival as well as functionality.
We did not receive any outside funding or grants in support of this research.