Uncontrolled Diabetes Mellitus in the Year Prior to Total Hip or Knee Arthroplasty Is Associated With Similar Perioperative Glucose Control, Early Complication Rates, and Five-Year Reoperation Rates After Preoperative Diabetes Mellitus Optimization.

Abstract

Perioperative glycemic control is considered a modifiable risk factor prior to primary total hip (THA) and total knee (TKA) arthroplasty; however, it is still unknown whether preoperatively optimizing hemoglobin A1C values affects perioperative glucose control or postoperative complications. The purpose of this study was to compare patients who required preoperative diabetes mellitus (DM) optimization in the year prior to surgery (defined as hemoglobin A1C ≥ 8%) versus patients who had DM who did not require preoperative DM optimization to compare 1) postoperative glycemic control and 2) 90-day complications and incidence of revision arthroplasty. This was a retrospective, single-center analysis. Patients undergoing primary total joint arthroplasty (TJA) who had a diagnosis of DM (hemoglobin A1C > 6.5%) were eligible for inclusion (N = 585). Patients who had previously uncontrolled DM requiring optimization (hemoglobin A1C ≥ 8.0%) in the year prior to surgery (N = 164) were compared to patients who had controlled DM (N = 421). All patients had hemoglobin A1C < 8% preoperatively. Outcomes included median and peak perioperative glycemic control (postoperative day [POD] zero to POD two), 90-day complications, and incidence of revision arthroplasty at final follow-up (mean 64 months). Multivariable logistic regressions evaluated associations of underlying preoperative demographic and treatment characteristics with outcomes. Early postoperative median serum glucose and peak serum glucose were similar between patients who had controlled versus optimized DM. Higher preoperative hemoglobin A1C and peak hemoglobin A1C in the year prior to surgery were associated with worse median and peak serum glucose. The 90-day complication rates were similar in patients needing DM optimization versus controlled DM patients (OR [odds ratio] 1.67, 95% CI [confidence interval] 0.91 to 2.88; P = 0.10). Metformin use was independently associated with lower complication rates. Reoperations were similar between patients requiring preoperative DM optimization versus controlled DM (OR 1.60, 95% CI 0.59 to 4.36, P = 0.35). This study suggests that the need for DM optimization does not affect perioperative glucose control, postoperative complications, or reoperations, suggesting that efforts to optimize hemoglobin A1C prior to THA and TKA are warranted.