Development of a nomogram for predicting pancreatic portal hypertension in patients with acute pancreatitis: a retrospective study.

Introduction: Pancreatic portal hypertension (PPH) is a rare complication of acute pancreatitis (AP) that can lead to severe gastrointestinal bleeding. The risk factors associated with PPH, as well as the overall prognosis, warrant further investigation. This study aims to develop and validate a nomogram to predict PPH in patients with AP.

Methods: Consecutive patients with AP from 2015 to 2023 were retrospectively included in the study. Demographic data, clinical manifestations within the first week of AP onset, and initial contrast-enhanced CT findings were used to develop the predictive model. Univariate and multivariate Cox regression analyses were performed to identify risk factors for PPH. Based on the results of the multivariate analysis, a nomogram was developed. The patients were randomly divided into training and validation sets at a 7:3 ratio. The accuracy and discriminative power of the predictive model were assessed using the area under the curve (AUC) from the receiver operating characteristic curve and the calibration curve.

Results: Of the 1473 patients with AP, 107 (7.3%) developed PPH within 6 months (range: 2-22 months) during follow-up. Multivariate regression analysis showed that body mass index (BMI) (HR, 1.10; 95% CI 1.04 to 1.16; p=0.001), moderately severe grade (HR, 9.36; 95% CI 4.58 to 19.13; p<0.001), severe grade (HR, 12.95; 95% CI 6.22 to 26.94; p<0.001), diabetes (HR, 2.26; 95% CI 1.47 to 3.47; p<0.001), acute fluid accumulation (HR, 2.13; 95% CI 1.31 to 3.47; p=0.002), and necrosis (HR, 3.64; 95% CI 2.30 to 5.78; p<0.001) were independent risk factors for PPH. A nomogram for predicting PPH was developed, with the predictive curves showing an AUC of 0.859 at 6 months and 0.846 at 9 months. In the validation set, the AUC at both time points was 0.812.

Conclusion: In summary, we identified BMI, moderately severe or severe AP, diabetes, acute fluid accumulation, and necrosis as risk factors for AP-related PPH. Using the largest cohort of patients with AP to date, we developed a highly accurate nomogram with strong discriminative ability for predicting PPH. Future studies with larger sample sizes are necessary to confirm our findings and conduct external validation.