DOES MALALIGNMENT AFFECT OUTCOME FOLLOWING KNEE REPLACEMENT?: A SYSTEMATIC REVIEW

Background Total Knee Replacement (TKR) is an effective treatment for knee arthritis. One long held principle of TKRs is positioning the components in alignment with the mechanical axis to restore the overall limb alignment to 180 ± 3 degrees. However, this view has been challenged recently. Given the high number of replacements performed, clarity on this integral aspect is necessary. Our objective was to investigate the association between malalignment and outcome (both PROMs and revision) following primary TKR. Metod A systematic review of MEDLINE, CINHAL, and EMBASE was carried out to identify studies published from 2000 onwards. The study protocol including search strategy can be found on the PROSPERO database for systematic reviews. Results From a total of 2107 citations, 18 studies, comprising of 2,214 patients, investigated the relationship between malalignment and PROMs. Overall 41 comparisons were made between a malalignment parameter and a PROM. Eleven comparisons (27%) demonstrated an association between malalignment and worse PROMs, with 30 comparisons (73%) demonstrating no association. Eight studies investigated the relationship between malalignment and revision. Four studies found an association between a measure of coronal alignment and revision rate, with four not demonstrating an association. Only one study examined axial and sagittal alignment and found an association with revision rates. Conclusion When considering malalignment in an individual parameter, there is an inconsistent relationship with outcome. Malalignment may be related to worse outcome, but if that relationship exists it is weak and of dubious clinical significance. However, this evidence is subject to limitations mainly related to the methods of assessing alignment post operatively and by the possibility that the premise of traditional mechanical alignment is erroneous. Larger longitudinal studies with a standardised, timely, and robust method for assessing alignment outcomes are required. Level of evidence 2a