Surgical Outcomes in Urogynecology—Assessment of Perioperative and Postoperative Complications Relative to Preoperative Hemoglobin A1c—A Fellows Pelvic Research Network Study

Abstract

Poor control of diabetes mellitus is a known predictor of perioperative and postoperative complications. No literature to date has established a hemoglobin A1c (HbA1c) cutoff for risk stratification in the urogynecology population. We sought to identify an HbA1c threshold predictive of increased risk for perioperative and postoperative complications after pelvic reconstructive surgery. This multicenter retrospective cohort study involving 10 geographically diverse U.S. female pelvic medicine and reconstructive surgery programs identified women with diabetes who underwent prolapse and/or stress urinary incontinence surgery from September 1, 2013, to August 31, 2018. We collected information on demographics, preoperative HbA1c levels, surgery type, complications, and outcomes. Sensitivity analyses identified thresholds of complications stratified by HbA1c. Multivariate logistic regression further evaluated the association between HbA1c and complications after adjustments. Eight hundred seven charts were identified. In this diabetic cohort, the rate of overall complications was 44.1%, and severe complications were 14.9%. Patients with an am HbA1c value of 8% or greater (reference HbA1c, <8%) had an increased rate of both severe (27.1% vs 12.8%, P < 0.001) and overall complications (57.6% vs 41.8%, P = 0.002) that persisted after multivariate logistic regression (odds ratio, 2.618; 95% confidence interval, 1.560-4.393 and odds ratio, 1.931; 95% confidence interval, 1.264-2.949, respectively). Mesh complications occurred in 4.6% of sacrocolpopexies and 1.7% of slings. The average HbA1c in those with mesh exposures was 7.5%. Preoperative HbA1c of 8% or higher was associated with a 2- to 3-fold increased risk of overall and severe complications in diabetic patients undergoing pelvic reconstructive surgery that persisted after adjustments.