Evaluating the impact of surgery sequence on infection rates in hip or knee arthroplasty: does sequence matter?

Abstract

The potential influence of surgical sequence of elective hip-and-knee reconstructive surgery in relation to an infection-related procedure on postoperative infection rates is not clearly understood. Therefore, we aimed to examine the impact of surgical sequence on the incidence of postoperative infections within one-year and the longest available follow-up period in patients undergoing hip-and-knee reconstructive surgery. A case-control study with propensity matching was utilized to examine elective surgeries conducted at Sunnybrook Holland Orthopaedic & Arthritic centre, Toronto, Canada between 2015 and 2018. We determined and categorized them based on their operating room (OR) sequence in relation to an infected case; the cases were those performed right after (post-infection cohort), and the controls were those performed before an infection-related procedure in the same OR (pre-infection cohort). We employed survival analysis to compare the infection incidence within one year and at the longest available follow-up among the propensity-matched cohort. A total of 13,651 cases were identified during the four year period. We successfully matched 153 cases (21 post-infection and 132 pre-infection) using propensity scores. Demographic and clinical characteristics were balanced through matching. Kaplan-Meier survival analysis showed no significant difference in infection-free survival within one year and at a median follow-up of 2.2 years [interquartile range 0.9-5.0] between surgeries conducted before and after infected cases (both log-rank p-values = 0.4). The hazard ratios for infection within one year and the longest follow-up period were both 0.37 [95%Confidence Interval 0.03-4.09, p = 0.418], as no more events occurred after one year. The sequence of surgical procedures, whether or not an elective arthroplasty or lower limb reconstructive procedure occurs before or after an infection-related case in the same OR, does not significantly affect postoperative infection rates. This finding supports the efficacy of the current infection control measures and suggests a reconsideration of surgical scheduling standards.