A Comparative Analysis of Preemptive Versus Targeted Sedation on Cardiovascular Stability After High-Risk Cardiac Surgery in Infants.

OBJECTIVE: To compare the effect of two sedation practices on cardiovascular stability during the early postoperative period in young infants following cardiac surgery: the routine early use of midazolam infusion (preemptive sedation) and the discretionary use of sedatives tailored to the patient's clinical condition (targeted sedation). DESIGN: Retrospective cohort study with matched controls. SETTING: A 15-bedded pediatric cardiac ICU. PATIENTS: Sedation strategies were compared by matching patients before and after the introduction of a targeted sedation guideline, replacing the existing practice of preemptive sedation. Inclusion criteria were age less than 6 months and cardiopulmonary bypass time greater than 150 minutes. Matching criteria were surgical procedure, age, and duration of cardiopulmonary bypass and cross-clamp. The main outcome was cardiovascular instability, defined by the presence of one of the following criteria in the first 12 hours after PICU admission: 1) simultaneous administration of greater than or equal to two inotropic or vasopressor drugs; 2) administration of greater than 60 mL/kg fluid boluses. Secondary outcomes were: 1) markers of cardiac output adequacy (heart rate, blood pressure, vasoactive inotropic score, urine output, volume of fluid boluses, central venous oxygen saturation, lactate); 2) occurrence of adverse events (cardiac arrest, extracorporeal membrane oxygenation, death); 3) sedatives administered and depth of sedation. INTERVENTIONS: Introduction of a guideline of targeted sedation. MEASUREMENTS AND MAIN RESULTS: Thirty-three patients with preemptive sedation were matched to 33 patients with targeted sedation. Targeted sedation resulted in less frequent oversedation, without compromising cardiovascular stability, as indicated by similar occurrence of cardiovascular instability (68.8% with preemptive sedation vs 62.5% with targeted sedation; p = 0.53) and adverse events, and similar markers of cardiac output adequacy. Although all preemptively sedated patients received an infusion of midazolam in the first 12 hours after surgery, only 19.4% of patients in the targeted sedation group received a sedative infusion (p < 0.001). CONCLUSIONS: Our data suggest that after high-risk cardiac surgery in young infants, routine sedation with midazolam may not prevent low cardiac output syndrome. When accompanied by a careful assessment of level of sedation, routine sedation of infants after high-risk cardiac surgery can be avoided without compromising hemodynamic stability or patient safety. The potential benefit of this approach is reduced exposure to sedative.