P25. Association of preoperative hemoglobin A1c and body mass index with wound infection rate in spinal surgery

Abstract

BACKGROUND CONTEXT Diabetes mellitus and obesity are common comorbidities in patients undergoing surgery for degenerative spine conditions. The association of preoperative glycemic control with surgical outcomes and complications is well described for general and orthopedic surgery, but less clear in spinal surgery. PURPOSE To undertake a large retrospective database review to examine the relationship between documented pre-operative hemoglobin A1c as a measure of glycemic control, as well as BMI as a measure of obesity, and their influence of post-operative infectious complications in patients undergoing surgery for degenerative spinal conditions. STUDY DESIGN/SETTING Large retrospective database review via the Veterans Administration Corporate Data Warehouse, accessing the National VA Database using VA Informatics and Computing Infrastructure (VINCI). PATIENT SAMPLE All VA patients undergoing surgery for degenerative spine conditions from 2007 to 2016, queried from a national database using CPT codes. A total of 430,000 CPT codes were identified representing 191,320 unique procedures, of which 175,895 patients had sufficient records for including in our analysis. OUTCOME MEASURES Preoperative hemoglobin A1c, preoperative height and weight, age at the time of surgery; postoperative infectious complications as identified by ICD codes in the 6 months after surgery. METHODS Logistic regression models were constructed using age at surgery, preoperative hemoglobin A1c, and preoperative BMI as predictors for the odds of infectious complications following surgery. RESULTS Age, BMI, and preoperative hemoglobin A1C were all found to be significant predictors of the odds of infection following spinal surgery. Increasing age was associated with a small (0.9% per year of age) relative risk, whereas increasing BMI (per 1kg/m^2) was associated with a 2.7% increase in relative risk, and hemoglobin A1c was associated with a 4.2% increase in relative risk per 1 percent increase in the lab value. Twenty-eight percent of patients had their hemoglobin A1c checked prior to surgery, this group had a much higher infection rate (43.5% greater relative risk) than those patients who were not tested, regardless of the actual hemoglobin A1c level. CONCLUSIONS BMI and hemoglobin A1c were significant predictors of postoperative infection in our study sample. This is the largest study to date on the subject and the first to our knowledge to establish a dose-response relationship between these comorbid factors and the risk of postoperative complications. FDA DEVICE/DRUG STATUS This abstract does not discuss or include any applicable devices or drugs.