Effect of Morphine on Respiratory Drive in Infants With Acute Viral Bronchiolitis Receiving Neurally Adjusted Ventilatory Assist

BACKGROUND: Neurally Adjusted Ventilatory Assist (NAVA) is a patient-triggered ventilatory mode requiring intact respiratory drive. We compared 2 intravenous morphine infusion doses in infants with acute viral bronchiolitis receiving NAVA. The objectives were, to evaluate (1) impact on neural respiratory drive, measured by electrical activity of the diaphragm (EAdi) and estimated airway occlusion pressure (P0.1) and (2) tolerability, via the COMFORT behavioral scale and physiological variables. METHODS: Randomized, nonblinded, crossover (morphine dose), feasibility trial. Infants likely requiring mechanical ventilation for >48 h at enrollment received intravenous morphine infusion rates of 5 μg/kg/h (low dose) and 20 μg/kg/h (standard dose) on alternate days, via a randomized crossover design. On each day, a combination of 16 ventilatory settings were used, incorporating differing NAVA, PEEP and pressure support levels. RESULTS: Thirteen participants received the intervention. The primary outcome, peak EAdi did not differ between morphine groups (P = .27), with an overall average of 12.2 ± 7.0 μV. The mean P0.1 was higher during low-dose morphine (1.14 vs 0.89 cm H2O, mean difference .25 cm H2O, 95% CI 0.12-0.39, P = .002). The low-dose group exhibited higher mean heart rates (123 vs 113 beats/min, mean difference 9.9 beats/min, 95% CI 5.2-14.6, P < .001), systolic blood pressures (90 vs 86 mm Hg, mean difference 3.9 mm Hg, 95% CI 0.4-7.5, P < .001), and COMFORT-B scores, (13.7 vs 13.2, mean difference 0.5, 95% CI 0.1-1.0, P = .03). Low-dose morphine was associated with lower transcutaneous CO2 levels (37 vs 43 mm Hg, mean difference 6 mm Hg, 95% CI 3.9-8.4 mm Hg, P < .001), despite no difference in between-group breathing frequency (P = .11), tidal volume (P = .16) or minute ventilation (P = .33). CONCLUSIONS: Intravenous morphine at 20 μg/kg/h did not blunt neural respiratory drive during NAVA and appeared more favorable than 5 μg/kg/h in terms of tolerability. This may inform a larger trial evaluating NAVA in bronchiolitis.