Beyond the Principal CPT Code: Secondary Procedural Burden and Risk Stratification in NSQIP Colectomy Data

This is the authors' abstract. We don't add key points for this paper.

Background: The American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) colectomy studies often define cohorts by principal Current Procedural Terminology (CPT) code, overlooking heterogeneity introduced by simultaneous secondary procedures. We evaluated the prevalence of related and unrelated secondary procedures (SPs) during colectomy and their association with postoperative complications. Methods: This was a retrospective cohort analysis of 2012–2022 ACS-NSQIP colectomy cases (N=353,550). Secondary CPT codes were classified as related (R) or unrelated (UR) to colectomy. Multivariable logistic regression compared colectomy with no SPs (NS) to R and UR cohorts and assessed cumulative procedural burden, adjusting for demographics, comorbidities, and operative factors. Primary outcome was any 30-day complication; secondary outcomes were serious complications, anastomotic leak, prolonged postoperative ileus, and mortality. Results: 51.0% underwent ≥1 SP (R 42.2%, UR 8.8%). Compared with NS, both R (aOR 1.17, 95%CI:1.15–1.20; p<0.001) and UR (aOR 1.19, 95%CI:1.15–1.23; p<0.001) were associated with increased complication risk. Risk increased stepwise with procedural burden; patients with ≥3 SPs had the highest adjusted risk (aOR 1.33, 95%CI:1.29–1.37; p<0.001). Similar associations were observed for serious complications, prolonged ileus, and mortality, but not anastomotic leak. Among UR cases, repair/rescue-type procedures were associated with increased odds of complications and mortality, whereas management of extra-colonic pathology did not significantly impact risk. Conclusions: SPs are common during colectomy and associated with adverse outcomes regardless of relatedness. Failure to account for SPs obscures operative complexity; risk-adjustment models and outcomes research should incorporate secondary procedural burden to improve benchmarking.