Objective: To determine the occurrence of narcotic withdrawal in critically ill children who receive continuous infusions of fentanyl.
Design: Prospective case series.
Setting: A university hospital pediatric intensive care unit.
Patients: Twenty-three children, aged 1 wk to 22 months (mean 6 months), who required assisted mechanical ventilation and who received continuous infusions of fentanyl for > 24 hrs.
Measurements and main results: Total fentanyl dose received, length of infusion, and peak infusion rate were recorded. Patients were evaluated for narcotic withdrawal by the Neonatal Abstinence Scoring System of Finnegan. Children with scores of > or = 8 were considered to have narcotic withdrawal. Withdrawal was observed in 13 (57%) of 23 infants. Total fentanyl dose (2.96 +/- 4.10 vs. 0.53 +/- 0.37 mg/kg, p < .005) and length of fentanyl infusion (13.1 +/- 11.3 vs. 3.8 +/- 1.5 days, p < .0001) were significantly greater in those infants with narcotic withdrawal than in those infants with no withdrawal, respectively. Peak fentanyl infusion rate (9.9 +/- 7.8 vs. 9.2 +/- 4.4 micrograms/kg/hr) did not differ significantly between the two groups. A total fentanyl dose of > 2.5 mg/kg or a duration of infusion of > 9 days was 100% predictive of withdrawal.
Conclusions: Continuous infusions of fentanyl produce a high occurrence rate of narcotic withdrawal when administered to critically ill children. This effect is both dose- and duration-dependent.