To determine the effect of lateral retinacular release on patellofemoral alignment in primary total knee arthroplasty (TKA).

From June 1990 to December 1998, 351 cases of TKA, which were followed up more than 3 years, were divided into two groups, 147 cases (group I) with lateral retinacular release and 204 cases (group II) were without release. We compared the two groups in terms of the range of motion, patellofemoral alignment and clinical results using a Hospital for Special Surgery score.

Average range of motion in group I (112 ° preoperatively and 113 ° on final follow-up) and group II (114 ° preoperatively and 113 ° on final follow-up) showed no statistical difference. The average functional knee score in group I (58 preoperatively and 88 on final follow-up) and group II (59 and 85 respectively) also showed no statistical difference. Patellar maltracking occurred in 4 cases (2.7%) in group I and 31 cases (15.2%) in group II. Patellar tilting (4.5 ° in group I and 7.7 ° in group II) and patellar translation (3.7 mm in group I and 7.8 mm in group II) showed significant statistical difference between the two groups at the final follow-up.

Patellar tracking was better in the lateral retinacular release group, although there was no clinically significant difference at the short term follow-up period.