Initial Dosing of Inhaled Nitric Oxide in Infants with Hypoxic Respiratory Failure

Abstract

BACKGROUND: Inhaled nitric oxide (iNO) is a potent and selective pulmonary vasodilator that decreases pulmonary resistance, and improves ventilation-perfusion matching, thereby improving oxygenation and reducing the need for more invasive therapies. Despite the efficacy of iNO at reducing the use of extracorporeal membrane oxygenation, significant concern remains over the potential toxicity from oxidative derivatives and methemoglobinemia. At present, there is no universal agreement on the lowest effective starting dose. Reported initial doses in the neonatal literature have ranged from 1 to 80 ppm.

PURPOSE: To determine if the initial dose of iNO altered the incidence of adverse outcome.

METHODS: A cohort of neonates who received iNO for treatment of hypoxic respiratory failure and were entered into the Duke Neonatal Nitric Oxide Registry were evaluated. Neonates with congenital anomalies were excluded. This registry collects data from 36 centers that voluntarily report their experiences with iNO. From this database, the starting dose was recorded and the clinical course was followed. Adverse outcomes were prospectively defined and monitored in the database and included: methemoglobinemia, chronic lung disease, treatment with extracorporeal membrane oxygenation, or death.

RESULTS: Data on 476 patients were analyzed. Based on starting doses, records were sorted into three groups: a low-dose group (LDG; <18 ppm, n=57), a mid-dose group (MDG; 18 to 22 ppm, n=320), and a high-dose group (HDG; >22 ppm, n=99). ANOVA showed no statistically significant differences among the groups except for PaO₂/FiO₂ (p<0.05). Neonates in the high starting dose group were more often classified as treatment failures (21% in the LDG, 27% in the MDG, and 38% in the HDG, p=0.04) and treated with extracorporeal membrane oxygenation (19% in the LDG, 23% in the MDG, and 34% in the HDG, p=0.05) compared to the lower dose groups. In addition, survival without the need for oxygen at 30 days or at discharge was higher in the lower dose groups (93% in the LDG, 84% in the MDG, and 76% in the HDG, p=0.03). Logistic regression, however, showed that the starting dose of iNO did not significantly influence these outcomes when corrected for the degree of hypoxemia (PaO₂/FiO₂) at the start of therapy (p>0.1). High initial doses of iNO (>22 ppm) were associated with higher levels of methemoglobin (p< 0.05). There were no differences in mortality or length of hospital stay between the groups.

CONCLUSIONS: There is significant variation in the starting dose of iNO between centers. Our retrospective study shows no evidence that higher doses improve outcome. A low concentration of iNO (<18 ppm) should be considered to minimize the potential toxicity of methemoglobin. Furthermore, a well-designed, prospective trial should be undertaken to further define the optimal starting dose.