199. HbA1c and BMI are modifiable risk factors for reducing wound complications but do not influence risk of pseudarthrosis

Abstract

<h3>BACKGROUND CONTEXT</h3> Patients with diabetes mellitus (DM) undergoing posterior spinal fusion (PSF) with elevated HbA1cs are at higher risk of postoperative complications than those with lower HbA1cs. DM patients with an HbA1c >7.05% are prone to postoperative complications such as wound complications (WC). Pseudarthrosis (PA) is another complication for PSF and requires further insight in determining risk factors for DM. <h3>PURPOSE</h3> To determine if DM patients with higher preoperative HbA1c levels will have higher rates of WC and PA. <h3>STUDY DESIGN/SETTING</h3> Single-center retrospective cohort study. <h3>PATIENT SAMPLE</h3> This study included 229 DM patients. <h3>OUTCOME MEASURES</h3> Patient demographics, osteopenia, preoperative HbA1c, WC, PA, and levels-fused (LF). <h3>METHODS</h3> Included: DM patients undergoing thoracolumbar PSF from 2014-2019 with preoperative HbA1C levels, and a minimum 1-year postoperative follow-up. PA was defined by the treating surgeon via clinical presentation, radiographs and postoperative CT scan with minimum 1-year postoperative interval. WC included dehiscence, seroma requiring aspiration or infection of the surgical site. Excluded: Patients undergoing LLIFs and ALIFs or diagnosed with prior pars fracture, spinal tumor, ankylosing spondylitis, fracture and infection. Relationships between complications and HbA1c were determined using t-test, chi-squared analyses and multivariate regressions (α=0.05). In risk factors, the cutoff HbA1C value as a predictor for multiple postoperative complications was identified via receiver operating characteristic (ROC) analyses using Youden's index. <h3>RESULTS</h3> A total of 229 patients met these criteria. Age (Normal: 65.4±11.0, PA: 63.0±8.2, p=0.249; Normal: 65.2±10.6, WC: 64.5±9.7, p=0.753), CCI (Normal: 4.6±2.0, PA: 4.0±1.2, p=0.06; Normal: 4.6±1.9., WC: 4.6±1.7, p=0.761), gender, percent female, (Normal: 57%, PA: 47%, p=0.286; Normal: 55%, WC: 58%, p=0.829), osteopenia (Normal: 3%, PA:3%, p=0.968; Normal: 3%, WC: 0.0%, p=0.377), and smoker status (Normal: 11%, PA: 16%, p=0.44; Normal: 11%, WC: 12%, p=0.915) did not have a significant impact in the tested postoperative complications. Mean HbA1c (7.6±1.5%, p=0.001), LF (3.9±3.2, p=0.024) and BMI (35.8±7.4, p=0.004) in patients with WC were significantly higher than those without WC (HbA1C: 6.8±1.0%; LF: 2.5±2.9 BMI: 31.7±6.7). HbA1c (OR: 1.7 [1.2-2.5], p=0.005), BMI (OR: 1.1 [1.0-1.2], p=0.007) and LF (OR: 1.1[1.1-1.3], p=0.031) was an independent risk factor WC. In those with PA, mean HbA1c (7.0±1.3, p=0.289) was not significantly higher than those without PA (HbA1C:7.0±1.1). HbA1c (OR: 1.039 [0.73-1.478], p=0.831) was not an independent risk factor for PA.ROC analyses demonstrated HbA1c>7.05% (AUC: 0.693[0.587-0.800], p=0.001], BMI>32.92 (AUC: 0.664[0.555-0.772], p=0.007) and >1.5 LF (AUC: 0.626[0.502-0.750], p=0.036) as fair predictive models for WC. <h3>CONCLUSIONS</h3> Preoperative HbA1C>7.05%, BMI>32.92 and LF>1.5 are independent risk factors of WC; however, neither of these are predictors for PA in diabetics who had PSF. Regardless, DM patients undergoing PSF should strive to lower their HbA1c<7.05% and BMI<32.92 before surgery. <h3>FDA DEVICE/DRUG STATUS</h3> This abstract does not discuss or include any applicable devices or drugs.