Predictive Value of Preoperative Prognostic Nutritional Index and Modified Surgical Apgar Score for Severe Postoperative Complications in Malignant Obstructive Jaundice

Objective: This study aimed to investigate the predictive value of preoperative prognostic nutritional index (PNI) and intraoperative modified surgical Apgar score (mSAS) for severe complications (Clavien–Dindo grade ≥ III) after curative-intent surgery in patients with malignant obstructive jaundice (MOJ) and to construct a nomogram prediction model. Methods: The clinical data of 308 patients who underwent curative-intent surgery for MOJ at the First Affiliated Hospital of Soochow University from January 2019 to January 2024 were retrospectively collected. Independent predictive factors were screened using univariate and multivariate logistic regression; the dose–response relationship between PNI and complication risk was analyzed using restricted cubic spline (RCS); subgroup analysis was conducted to evaluate the consistency of predictive effects; procedure-specific sensitivity analyses were performed separately for patients undergoing pancreaticoduodenectomy (PD) and hepatectomy with extrahepatic bile duct resection (HCAE); a nomogram was constructed based on the independent predictive factors, and the performance of the model was evaluated using receiver operating characteristic (ROC), calibration curve, and decision curve analysis (DCA). Results: Among the 308 patients, 65 (21.1%) had severe complications. The PNI (40.30 ± 4.19 vs. 45.91 ± 4.75) and mSAS (5.28 ± 1.28 vs. 7.28 ± 1.34) in the severe complication group were significantly lower than those in the non-severe complication group (all p < 0.001). After multivariable adjustment including surgical approach, higher PNI (OR = 0.73, 95% CI 0.66–0.82, p < 0.001) and higher mSAS (OR = 0.30, 95% CI 0.21–0.44, p < 0.001) were independently associated with lower odds of severe complications. The quartile analysis of PNI showed a graded inverse association (p for trend < 0.001). Multivariable-adjusted RCS analysis demonstrated a significant overall inverse association between PNI and severe complication risk (p for overall < 0.001), with no significant evidence of nonlinearity (p for nonlinearity = 0.425). In procedure-specific sensitivity analyses, the PNI–mSAS model achieved an AUC of 0.959 (95% CI 0.921–0.997) in the PD subgroup and 0.901 (95% CI 0.837–0.965) in the HCAE subgroup. The AUC of the nomogram prediction model was 0.924; the calibration curve indicated good agreement, and DCA suggested potential clinical utility. Conclusion: Preoperative PNI and intraoperative mSAS are independently associated with severe complications after curative-intent surgery in patients with MOJ. The PNI–mSAS nomogram showed good internal discrimination and calibration and may assist perioperative risk stratification; however, external validation is required before routine clinical application.