Development and validation of a nomogram for early assessment the severity of acute pancreatitis.

ABSTRACT

BACKGROUND: Acute pancreatitis is an acute inflammatory disorder of the pancreas, and severe acute pancreatitis is associated with high mortality. Early assessment the severity of AP has an important significance for improving clinical outcomes. Our object aimed to develop a nomogram with high simplicity and rapidity for predicting the severity of acute pancreatitis. METHODS: Patients admitted to the Hunan Provincial People's Hospital within 72 h from onset of AP from January 2010 and December 2020 were enrolled to establish a nomogram. Independent predictors were determined using univariate and multivariate analysis and then assembled to construct a predicting nomogram. The performance of proposed nomogram was evaluated by Brier score and Harrell's concordance index (C-index). Meanwhile, clinical data of AP patients from January 2021 to January 2022 were collected for external validation. RESULTS: Album (OR 0.891, 95%CI 0.867-0.917), calcium (OR 0.151, 95%CI 0.084-0.273), neutrophil to lymphocyte ratio (OR 1.055, 95%CI 1.023-1.088) and systemic inflammatory response syndrome (OR 6.292, 95%CI 4.459-8.879) were identified as independent factors of SAP after univariate and multivariate analysis (p < .05). A predictive nomogram was accordingly established using these four independent variables. The internally verified C-index was 0.796 (95% CI 0.773-0.818), Brier score was 0.138. The externally verified C index was 0.820 (95% CI 0.754-0.887). CONCLUSION: A nomogram for predicting the severity of AP was well developed, it may be of great significance for clinicians to quickly assess the progress of AP and choose more-targeted strategies.