Impact of early fluid intake and urine output on mortality in acute pancreatitis: A retrospective cohort study

Objective To evaluate the associations between early fluid-related parameters, including fluid intake, fluid balance, and urine output, within the first 24 h after intensive care unit admission and mortality among critically ill patients with acute pancreatitis. Methods We conducted a retrospective cohort study using the publicly available Medical Information Mart for Intensive Care III database. Adult patients with acute pancreatitis admitted to the intensive care unit between 2001 and 2012 were included. Fluid intake, fluid balance, and urine output during the first 24 h after intensive care unit admission were extracted and categorized into four predefined levels. Multivariable logistic regression models were used to assess the associations between early fluid-related parameters and in-hospital mortality, adjusting for baseline demographic characteristics and indicators of disease severity. Generalized propensity score analyses were additionally performed to evaluate the distribution of fluid-related exposures and the potential influence of measured severity-related confounding. Sensitivity analyses using E-values were conducted to assess the potential impact of unmeasured confounding. Results Among critically ill patients with acute pancreatitis, higher early fluid intake levels were associated with increased in-hospital and 28-day mortality after adjustment for baseline characteristics and disease severity. Conversely, higher urine output levels were associated with lower in-hospital and 28-day mortality. Early fluid balance showed weaker and less consistent associations with mortality. Generalized propensity score analyses demonstrated nonrandom distributions of fluid intake, fluid balance, and urine output according to baseline severity-related characteristics, indicating the presence of confounding by indication in early fluid management. After adjustment for measured confounding factors, the observed associations between fluid intake, urine output, and mortality remained consistent. However, residual confounding could not be completely excluded. Conclusion In critically ill patients with acute pancreatitis, early fluid intake and urine output were associated with mortality outcomes. These parameters may represent potential markers of disease severity, physiological reserve, and response to initial resuscitation rather than direct determinants of outcome. Further prospective studies are required to clarify their clinical significance.