Effect of hemoglobin A1c levels on 30-day complications after elective anterior cervical discectomy and fusion: a retrospective cohort study.

Background: Anterior cervical discectomy and fusion (ACDF) is one of the most common spinal procedures due to its strong efficacy and consistency. Patients with type 2 diabetes mellitus (DM) constitute a large portion of patients undergoing ACDF, and DM status has been associated with worse post-operative complication rates. Hemoglobin A1c (HbA1c) levels have been utilized as a surrogate marker to assess diabetes control perioperatively. However, whether surgeons should risk delaying ACDF to better optimize HbA1c levels pre-operatively is a common dilemma faced in clinical practice. Further clarifying the perioperative risks associated with increasing HbA1c levels is essential to resolving this ongoing clinical ambiguity. Therefore, we sought to evaluate the impact of rising HbA1c on post-operative complications following ACDF in order to better inform evidence-based thresholds that may help guide surgical timing and preoperative optimization. Methods: A total of 10,508 adult patients (≥18 years) undergoing elective ACDF from 2021-2023 with recorded HbA1c values in a national surgical registry were included in the study. Patients were stratified into five HbA1c categories: <5.7%, 5.7-6.49%, 6.5-7.49%, 7.5-8.49%, and ≥8.5%. Multivariable regression analyses were performed to evaluate postoperative outcomes. To assess non-linear dose-response relationships, HbA1c was additionally modeled as a continuous variable using restricted cubic splines (RCSs). Time-to-event outcomes were analyzed using Cox proportional hazards regression, accounting for days from operation to complication within 30 days postoperatively. Inverse probability of treatment weighting (IPTW) and trend testing were used to further evaluate associations between HbA1c and postoperative outcomes. Results: Unadjusted rates of unplanned readmission increased across HbA1c categories (P<0.001). On multivariable analysis, HbA1c ≥8.5% was independently associated with increased odds of unplanned readmission [odds ratio (OR) =1.72; 95% confidence interval (CI): 1.10-2.59; P=0.01], and HbA1c 7.5-8.49% was associated with higher odds of sepsis (OR =2.53; 95% CI: 0.95-6.15; P=0.048). Cochran-Armitage trend testing revealed significant increases in any complication, myocardial infarction (MI), and minor complications with rising HbA1c. ROC analysis showed poor predictive performance [area under the curve (AUC) range, 0.519-0.564] with no clinically meaningful HbA1c cutoff. Conclusions: Elevated HbA1c is associated with increased risk of readmission and sepsis following elective ACDF. However, HbA1c alone demonstrates poor discriminative ability for predicting complications and should not be used solely to risk stratify. Therefore, while HbA1c levels are helpful in discerning comprehensive preoperative risk, we suggest that HbA1c values alone should not be used to assess a diabetic patient's risk of complications post-ACDF.