There have been significant changes in the management of neonates and infants undergoing cardiac surgery in the past decade. We have evaluated in this prospective, randomized, double-blinded study the effect of large-dose fentanyl anesthesia, with or without midazolam, on stress responses and outcome. Forty-five patients < 6 mo of age received bolus fentanyl (Group 1), fentanyl by continuous infusion (Group 2), or fentanyl-midazolam infusion (Group 3). Epinephrine, norepinephrine, cortisol, adrenocortical hormone, glucose, and lactate were measured after the induction (T1), after sternotomy (T2), 15 min after initiating cardiopulmonary bypass (T3), at the end of surgery (T4), and after 24 h in the intensive care unit (T5). Plasma fentanyl concentrations were obtained at all time points except at T5. Within each group epinephrine, norepinephrine, cortisol, glucose and lactate levels were significantly larger at T4 (P values < 0.01), but there were no differences among groups. Within groups, fentanyl levels were significantly larger in Groups 2 and 3 (P < 0.001) at T4, and among groups, the fentanyl level was larger only at T2 in Group 1 compared with Groups 2 and 3 (P < 0.006). There were no deaths or postoperative complications, and no significant differences in duration of mechanical ventilation or intensive care unit or hospital stay. Fentanyl dosing strategies, with or without midazolam, do not prevent a hormonal or metabolic stress response in infants undergoing cardiac surgery.
Implications: We demonstrated a significant endocrine stress response in infants with well compensated congenital cardiac disease undergoing cardiac surgery, but without adverse postoperative outcome. The use of large-dose fentanyl, with or without midazolam, with the intention of providing "stress free" anesthesia, does not appear to be an important determinant of early postoperative outcome.