Parameters Predictive of Propofol-Associated Acute Pancreatitis in Critically Ill Patients with COVID-19 Pneumonia: A Retrospective Cohort Study

Abstract

Background Propofol, a commonly used agent for short- and long-term sedation, is associated with acute pancreatitis. The main indirect mechanism of propofol-associated acute pancreatitis is by inducing hypertriglyceridemia. Patients with severe coronavirus disease 2019 (COVID-19) pneumonia often require prolonged mechanical ventilation and sedation. We examined the incidence rate of acute pancreatitis among critically ill adults with COVID-19 pneumonia on mechanical ventilation receiving propofol. In addition, we attempted to determine cutoff levels of serum triglycerides and doses of propofol that are predictive of propofol-associated acute pancreatitis. Methods This was a multicenter retrospective cohort study using a large dataset of hospitalized patients with COVID-19. The collected data included the number of days on propofol, cumulative doses of propofol, peak levels of serum triglycerides, serum lipase levels, and abdominal imaging findings. We used receiver-operating characteristic analysis in conjunction with Youden's index to identify the optimal thresholds for propofol administration parameters and levels of triglycerides that would provide maximal sensitivity and specificity for predicting acute pancreatitis. Results Out of 499 critically ill patients with COVID-19 pneumonia, 154 met the inclusion criteria. Six (4%) patients had suspected acute pancreatitis based on elevated serum lipase levels. Cutoff values greater than 688 mg/dL for peak level of triglycerides, 4.5 days on propofol, 3007 mg/day for average daily propofol dose, and 24 113 mg for cumulative propofol dose were associated with high risk of suspected acute pancreatitis. The negative predictive values for suspected acute pancreatitis using these cutoffs ranged from 98% to 100%. Conclusions Propofol use in critically ill COVID-19 patients is associated with a low incidence rate of acute pancreatitis. We identified cutoff values for serum triglycerides and cumulative propofol dose that are linked to higher risk of propofol-associated pancreatitis. More research is needed to examine the true incidence of propofol-associated pancreatitis and help develop optimal cutoff values for certain parameters to help guide safe propofol administration.