772 - The Effects of Low Dose Morphine Protocol on Post-Operative Congenital Heart Patients

Poster Number: 772
Publication Number: 772.306

Mackenzie R. Dreher, Akron Children's Hospital, Copley, OH, United States; Christopher Page-Goertz, Akron Children's Hospital, Hudson, OH, United States; James Besunder, Akron Children's Hospital, Akron, OH, United States; Neil L. McNinch, Akron Children's Hospital, Akron, OH, United States; Cassandra A.. Ruggles, Akron Children's Hospital, Uniontown, OH, United States; Jonathan H. Pelletier, Akron Children's Hospital, Akron, OH, United States; Patricia L.. Raimer, Akron Children’s Hospital, Akron, OH, United States; Robert D. Stewart, Akron Children's Hospital, Akron, OH, United States; Ryan Nofziger, Akron Children's Hospital, Akron, OH, United States

Background: The use of opioids in the Pediatric Intensive Care Unit (PICU) after congenital heart (CV) surgery is common, but not without adverse effects such as hypotension, increased length of mechanical ventilation, PICU length of stay, and drug dependence. Previous studies found that children < 3 years old who underwent CV surgery have a larger volume of distribution and lower clearance of morphine. This supports the use of a loading dose following the surgery and a lower continuous infusion rate in the PICU. In November 2019, a low dose morphine infusion protocol was implemented for post-op CV patients.

Objective: We hypothesized the protocol for analgosedation would decrease the cumulative dose of opioids following CV surgery while providing adequate analgesia.

Design/Methods: Patients admitted following CV surgery between 11/1/2017 and 12/31/2020 were analyzed. Those who met inclusion criteria were grouped based on date of surgery (pre or post implementation phase), with a washout of 2 months. Cumulative opioid doses were converted to total morphine equivalents per kilogram. The use of adjunct medications including acetaminophen, ketorolac, ibuprofen, dexmedetomidine, midazolam, lorazepam, and chloral hydrate were compared. Pain, state behavioral, and delirium scores were compared. Wilcoxon Rank Sum Test assessed differences by phase. Quantile regression assessed phase effect, adjusting for demographic and clinical characteristics, with similar results noted regardless of adjusted covariate. Data are expressed as median (IQR).

Results: There were 189 patients in phase 1 and 96 in phase 2. Baseline characteristics were comparable. There were no differences in age (months), 9.0 (3.0-56.0) vs 5.5 (3.0-53.5); [p=0.883]; PRISM3 5.0 (3.0-8.0) vs 4.5 (2.0-8.0); [p = 0.074]; PELOD 10.0 (1.0-12.0); [p = 0.102].
Opioids and acetaminophen were used less in the post-implementation phase; [median (IQR) was 0.5 mg/kg (0.2-1.3) vs 0.2 mg/kg (0.1-0.4); p< 0.001, and 168.6 mg/kg (124.9-237.0) vs 147.8 mg/kg (102.0-224.8); p-values < 0.001 and = 0.041 respectively].; NSAID use was lower in phase 1; [median (IQR) was 26.4 mg/kg (0.0-59.2) vs 60.3 mg/kg (30.2-83.4); p< 0.001]. LOS (days) and pain scores were not significantly different by phase [median (IQR) 2.4 (1.4-5.2) vs 2.4 (1.4-4.2); p = 0.561], and 3.0 (1.5-4.6) vs 2.8 (1.4-4.2); p = 0.431].

Conclusion(s): Implementation of a low dose morphine protocol in CV surgical patients resulted in significantly lower cumulative doses of opioids and acetaminophen compared to pre-protocol phase while maintaining adequate analgesia. LOS in the PICU was not affected.