S292 Low Prognostic Nutritional Index as a Predictor of Mortality and Severity in Acute Pancreatitis: A Systematic Review

Introduction: Malnutrition and systemic inflammation aggravate acute pancreatitis (AP). The Prognostic Nutritional Index (PNI) summarizes both processes; however, its ability to predict mortality and disease severity in AP has not been synthesized. Methods: A literature search was conducted across PubMed and Google Scholar for all articles through May 2025. Of 70 titles and abstracts screened, 7 full texts were assessed and 5 were selected for detailed review. Two reviewers independently performed all steps of screening and bias assessment, with any discrepancies resolved by a third reviewer. Meta-analysis was precluded by heterogeneity in PNI cut-offs and effect metrics; findings were therefore narratively synthesized. Results: We analyzed 5 observational studies with a total of 1691 patients with AP. PNI was measured on admission in 5 cohorts and at 48 hours in 1 and on day 3 and day 7 in 1 study. Three studies (n = 1059) reported mortality. Li et al. (2017) reported that mean admission PNI was 41.7 ± 7.5 in survivors vs 34.0 ± 6.4 in non-survivors (P < 0.001); PNI ≤33.1 yielded an AUC of 0.781 and retained significance after adjustment (odd ratio [OR] 2.64, 95% CI 1.25-5.59). Akkuzu et al. observed 8 deaths among 210 patients with PNI <45 vs none among 151 with PNI ≥45 (crude OR 21.3, 95% CI 4.8-93.5). Efgan et al., using PNI as a continuous covariate, reported AUC of 0.758 for cut-off ≤43.45 and adjusted HR of 1.055 (1.003-1.156) per unit decrease. For severity, Li et al. found PNI <41.1 independently associated with severe AP (OR 7.75, 3.40-17.68). Pian et al. showed decreasing median PNI throughout the initial 48 hours with minimal improvement by the end of the week in severe cases and a day-3 AUC of 0.814 for the cut-off 40.63. Akkuzu et al. showed that severe AP occurred in 31% of low-PNI vs 1% of high-PNI patients (P < 0.001); and Shi et al. (2025) demonstrated day-2 PNI ≤38.53 predicted severity (AUC 0.871) with adjusted OR 0.82 (0.75-0.90). Across all studies, PNI at 48 hours was most predictive of the mortality/severity of AP. Conclusion: Across 1691 patients, lower PNI consistently predicted higher in-hospital mortality and greater disease severity. Given its derivation from routinely available labs, PNI offers a low-cost, accessible tool for early risk stratification. Although consistent, multicenter prospective studies using standardized PNI measurement are warranted to validate its bedside prognostic utility.