Valeria Perez, RPh, PharmD,
Texas Health Presbyterian Hospital Dalla
Texas
Disclosure information not submitted.
Andrew Faust, BCPS, PharmD
Clinical Pharmacist
Texas Health Presbyterian Hospital Dallas, United States
Disclosure information not submitted.
Title: Efficacy of an Insulin Protocol for Treatment of Hypertriglyceridemia-Induced Acute Pancreatitis
Introduction: The risk of acute pancreatitis rises with triglyceride (TG) levels above 500 mg/dL and is markedly higher when >1000 mg/dL. Historically, treatment for hypertriglyceridemia-induced acute pancreatitis (HTG-AP) is plasmapheresis; however, this carries risks, more limited accessibility, and higher costs. Insulin, a low-cost alternative to plasmapheresis, is supported by literature but does not have a standardized, protocol-based strategy. The purpose of this study is to determine efficacy of an insulin protocol for treatment of patients with HTG-AP.
Methods: This was a retrospective analysis of pre- and post-implementation of an insulin protocol for hospitalized patients for the treatment of HTG-AP. The target insulin infusion rate is 15 units/hour while maintaining a blood glucose of 120-180 mg/dL by use of intravenous dextrose. Inclusion criteria were age ≥ 18 years, baseline TG level >1000 mg/dL, and a diagnosis of acute pancreatitis. Exclusion criteria were any causes of pancreatitis other than HTG and patients with no follow up TG reading. Pre-protocol patients were admitted January 2019 to May 2020, and insulin-protocol patients were admitted June 2020 to April 2021. Primary outcome was the time to a serum TG level ≤ 1000 mg/dL. Secondary outcomes included time to TG ≤ 500 mg/dL in the subgroup of patients with TG levels ≤ 500 mg/dL. Incidence of hypoglycemia requiring D50W was collected as a safety outcome.
Results: Forty-seven patients were included in the study (26 pre-protocol and 21 insulin-protocol). Time to serum TG ≤ 1000 mg/dL level was 50.9 ± 31.2 hours in the pre-protocol group vs. 34.9 ± 25.2 hours in the insulin-protocol group (difference 16 hours; 95% CI, -0.98 to 32.9 hours; p = 0.06). For the 31 patients treated to TG levels ≤ 500 mg/dL, there were no differences in time to resolution (69.7 ± 37.3 hours vs. 50.9 ± 32.3 hours; p = 0.15). Hypoglycemia requiring D50W administration was no different in the groups (5 patients vs. 2 patients).
Conclusions: Use of an insulin-based protocol for HTG-AP demonstrated a strong trend towards faster resolution to TG ≤ 1000 mg/dL with no increase in the risk for hypoglycemia. While additional, larger, better powered studies are needed for treatment of HTG-AP, an insulin infusion should be considered a viable alternative to plasmapheresis.