Abstract
It is unclear whether activated partial thromboplastin time (APTT) is predictive of survival in patients with acute pancreatitis (AP). Our study aimed to investigate the relationship between APTT and short-term prognosis in AP. From the Medical Information Mart for Intensive Care (MIMIC)-IV database, a total of 844 patients with AP were randomly divided into the training cohort (n = 591) and the validation cohort (n = 253) at a ratio of 7:3. Based on their APTT values, the patients were divided into the normal and high groups. The primary outcome of this study was 30- and 60-day survival. Kaplan-Meier survival analysis and Cox regression models were used to analyze associations between groups and outcomes. The training and validation cohort matched well on all parameters (p > 0.05). In terms of 30- and 60-day survival, Kaplan-Meier survival curves from both training and validation cohorts demonstrated a lower survival probability for patients in the high APTT group than the normal group (log-rank p < 0.05). In the training cohort, patients in the high APTT group had a statistically significantly higher risk of death than those in the normal group after controlling for possible confounders in Cox regression (p < 0.05). For the high APTT group, the hazard ratios (95% confidence interval [CI]) were 1.63 (95% CI 1.10, 2.61, p = 0.035) and 1.49 (95% CI 1.01, 2.38, p = 0.041), respectively. APTT performed as well as BISAP, Ranson, and APACHE II models in predicting 30- and 60-day survival in patients with AP. The results above have been verified in the validation cohort. Prolonged APTT in patients with AP may increase the risk of short-term death.