PURPOSES:A fixed severe valgus knee is a surgical challenge. A safe post-operative Hip-Knee-Ankle angle (HKA) range of 180° ± 4 was recommended, but recent studies mentioned equal results from outliers of this range. Nevertheless, no distinction was made between varus and valgus knees, as well as over-corrected or under-corrected knees. Did post-operative nonaligned total knee replacements (TKR) from fixed severe valgus knees behave differently from the properly aligned population? Did over-corrected knees behave differently from under-corrected knees? METHODS:Through a multi-center retrospective cohort study, we provided 557 knees of at least 10° of minimal pre-operative valgus; in this population 75 presented a post-operative Hip-Knee-Ankle angle (HKA) outside of the 180° ± 4 range; 23 of them had at least 5° of varus; 52 of them had at least 5° of valgus. Median pre-operative HKA of the entire cohort was 194° (range 190-198). Median follow-up was 8 years (range 5-11); Knee Society Score (KSS) results, HKA, Femoral and Tibial Mechanical Angles (FMA, TMA) and complication rates were obtained. The outlier group (HKA ≤ 175 or ≥ 185) was compared to the control group (HKA 180 ± 4); over-corrected (HKA ≤ 175) and under-corrected (HKA ≥ 185) sub-groups were individually tested against the control group. RESULTS:The outlier group had a lower Final Knee Score than the aligned group (p = 0.023). In the over-corrected sub-group, median post-operative FMA was 88° (SD 4°) and median TMA was 87° (SD 4°). The complication rate was higher (p = 0.019). Knee (p = 0.018), Function (p = 0.034) and Final Knee Scores (p = 0.03) were statistically lower than in the control group. In the under-corrected sub-group, mean post-operative FMA was 93° (SD 2°) and mean TMA was 91° (SD 2°). The complication rate was lower (p = 0.019) and there was no difference with the control group concerning KSS. CONCLUSIONS:In case of pre-operative fixed severe valgus knee, one should avoid over-correcting HKA angle and especially the TMA. Over-correction of a severe preoperative valgus in a post-operative varus was prejudicial for TKA survival. Keeping a severe valgus knee in low valgus to avoid using a more constrained implant and/or ligament releases will not decrease the 5-10 year implant survival and functional scores.