Randomized multicenter trial comparing synchronized and conventional intermittent mandatory ventilation in neonates.

Abstract

OBJECTIVE: To compare synchronized intermittent mandatory ventilation (SIMV) and conventional intermittent mandatory ventilation (IMV) in neonates. STUDY DESIGN: Prospective, multicenter, randomized clinical trial. SETTING: Level III neonatal intensive care units at six university or children's hospitals. PATIENTS: Three hundred twenty-seven infants receiving conventional IMV for respiratory distress syndrome, pneumonia, or meconium aspiration pneumonitis were randomly assigned a 7.5 +/- 6 hours of age to either continue with IMV or change to SIMV. Infants assigned to each mode of ventilation had similar birth weight (BW), gestational age, and Apgar scores at birth, and similar oxygenation indexes at randomization. They received similar surfactant therapy and had similar incidence of sepsis, seizures, secondary pneumonia, and necrotizing enterocolitis. In the infants with BW less than 1000 gm, more infants receiving IMV had surgical ligation of their patent ductus arteriosus than did those receiving SIMV (27 vs. 7 %; p = 0.02). ANALYSIS: Data was analyzed overall for all infants and also separately within three BW groups: less than 1000 gm, 1000 to 2000 gm, and more than 2000 gm. The 1000 to 2000 gm BW group was further analyzed in subgroups weighing 1000 to 1499 gm and 1500 to 2000 gm. RESULTS: In all infants, at 1 hour after randomization, the infants receiving SIMV had a lower mean airway pressure than those receiving IMV (8.08 +/- 2.15 vs. 8.63 +/- 2.59; p<0.05), with similar fractions of inspired oxygen and oxygenation indexes. Infants whose BW was 1000 to 2000 gm at 0.5 hour required a lower fraction of inspired oxygen with SIMV than with IMV (0.52 +/- 0.20 vs. 0.62 +/- 0.27; p<0.05) and had better oxygenation at 1 hour, as shown by lower oxygenation indexes with SIMV than with IMV (6.14 +/- 4.17 vs. 9.42 +/- 8.41; p = 0.01). Infants whose BW was 1000 to 2000 gm received a lower number of unit doses of sedative/analgesic drugs per infant during the first 4 days of SIMV than did infants receiving IMV (3.8 +/- 3.4 vs 6.3 +/- 5.5 unit doses; p = 0.02). Infants whose BW was more than 2000 gm had a shorter duration of mechanical ventilation with SIMV than with IMV (median, 72 vs 93 hours; p = 0.02). Three of the forty-six infants receiving IMV but none of the 47 infants receiving SIMV required extracorporeal membrane oxygenation. In the infants with BW less than 1000 gm, fewer infants treated with SIMV required supplemental oxygen at 36 weeks of postconceptional age than did those treated with IMV (47 vs 72%; p<0.05). In 83 infants whose lungs were mechanically ventilated for 14 days or longer, all with BW less than 2000 gm, those treated with SIMV regained their BW earlier than those treated with IMV (median, 21.5 vs 29 days; p<0.01). There were no differences in the rates of death, intraventricular hemorrhage (grades III and IV), air leak, need for pharmacologic paralysis, or need for supplemental oxygen at 28 days. CONCLUSIONS: We found that SIMV was at least as efficacious as conventional IMV, and may have improved certain outcomes in BW-specific groups.