Inhaled nitric oxide does not increase lipid peroxidation in preterm infants.

Abstract

Sir: Oxygen-derived free radical injury may be a common pathogenic mechanism in several neonatal diseases including chronic lung disease (CLD), retinopathy of prematurity, intraventricular haemorrhage, and necrotising enterocolitis. Previous work suggests that there is a high background level of lipid peroxidation in very low birth weight preterm infants and this is associated with adverse neonatal outcome. Since the introduction of inhaled nitric oxide (INO) there have been fears that it could increase free radical damage to cells. This is a particular concern in ventilated preterm infants already under considerable oxidative stress and with compromised anti-oxidant defences. We performed a randomised controlled trial of treatment with INO therapy and/or intravenous dexamethasone in preterm infants at risk of developing CLD [4]. This pilot study was designed to look for evidence of increased lipid peroxidation after INO. Urine and breath samples were collected before treatment commenced and 6, 24, 48 and 72 h after beginning therapy in treated and control subjects. Urinary malondialdehyde (uMDA) was measured by HPLC [1], and breath pentane by GC with thermal desorption and cryofocussing [3]. Gestation and birth weight were similar in both groups. There were no statistical di€erences in outcome measures between treated and non-treated infants [4]. The pre-treatment pentane exhalation rate was similar in the two groups. After 6 and 24 h, infants who received INO exhaled less pentane than control infants, but this was not statistically signi®cant. Exhaled pentane was similar after 48 h of treatment in both groups (Table 1). Median pre-treatment levels of uMDA were higher in the control group, and remained so for the duration of the treatment. None of the di€erences achieved statistical signi®cance (Table 1). Infants in the INO group had a lower oxygenation index when compared with control infants. In conclusion, we have been unable to demonstrate a further increase in lipid peroxidation above baseline as evidenced by uMDA or breath pentane. However, the power of the study to detect such changes is low ± 50% power to detect an increase of 50% in pentane or uMDA ± the magnitude of increase typically observed in our population with adverse outcome such as signi®cant intraventricular haemorrhage, retinopathy of prematurity or death [3]. If anything, the trend is to lower levels of lipid peroxidation products in the group of infants receiving INO. The number of infants studied is small; the trial was terminated prematurely since interim analysis showed that neither INO nor dexamethasone, either alone or in combination, were likely to decrease the incidence of CLD or death. Despite this, our results are in agreement with limited data available showing lower lipid peroxidation in animals exposed to hyperoxia and INO [2], [5]. Although we were unable to detect changes in lipid peroxidation during the administration of INO to preterm infants, we cannot exclude the possibility of a type II error.