OBJECTIVE. The aim was to determine whether inhaled nitric oxide therapy improves neurodevelopmental outcomes for infants with preterm persistent pulmonary hypertension of the newborn.
METHODS. We conducted a historical cohort study to compare the 3-year incidence of cerebral palsy in preterm singleton infants (<34 gestational weeks) with hypoxemic respiratory failure caused by persistent pulmonary hypertension of the newborn who received inhaled nitric oxide (16 patients) or 100% oxygen (15 patients) therapy. All neonates had clinical and echocardiographic evidence of pulmonary hypertension without structural heart disease.
RESULTS. The incidence of cerebral palsy among patients treated with inhaled nitric oxide was 12.5%, whereas that among patients treated with 100% oxygen was 46.7%. After adjustment for maternal fever (≥38°C) during delivery, birth weight, Apgar score at 5 minutes, high-frequency oscillatory ventilation, and surfactant therapy, inhaled nitric oxide therapy, compared with 100% oxygen therapy, was associated with a decreased risk of cerebral palsy in preterm infants with persistent pulmonary hypertension of the newborn.
CONCLUSION. Inhaled nitric oxide therapy decreases the risk of cerebral palsy in preterm infants with persistent pulmonary hypertension of the newborn.
Persistent pulmonary hypertension of the newborn (PPHN) is a disease in which the pulmonary vascular resistance remains elevated during the neonatal period. Preterm PPHN, which is associated with a high risk of adverse health and neurodevelopmental outcomes, continues to be one of the most challenging conditions encountered in the NICU.
Nitric oxide (NO) is produced by vascular endothelial cells and plays an important role in increasing the blood flow to the lungs after birth.1–5 Inhaled NO (iNO) causes selective pulmonary vasodilation in newborn lambs1 and has been shown to have a short-term benefit by improving oxygenation in infants with hypoxemic respiratory failure caused by PPHN.6 However, it is not clear whether, in preterm infants with PPHN, iNO therapy decreases the risk of cerebral palsy (CP), which is one of the most serious neurodevelopmental complications of preterm infants. We conducted a historical cohort study to compare the 3-year incidence of CP in preterm singleton infants with hypoxemic respiratory failure caused by PPHN who received either iNO or 100% oxygen therapy.
A historical cohort study was performed that involved 61 consecutive patients without congenital anomalies who were admitted to the ICU at the Osaka Medical Center and Research Institute for Maternal and Child Health between January 1988 and December 1999 and who were singleton infants of <34 gestational weeks (median: 25.0 weeks; interquartile range [IQR]: 24.0–28.0 weeks) with hypoxemic respiratory failure caused by PPHN. Thirty of the 61 patients in the original cohort were excluded; 26 patients died within 3 years after birth (median: 1.0 day; IQR: 0.6–7.8 days) and 4 patients were monitored at other hospitals after discharge. The gestational age was estimated by means of menstrual dates and ultrasound scans obtained before 20 weeks of gestation. The scan date was preferred if the menstrual date was uncertain or if there was a discrepancy of >14 days between the 2 dates. All neonates had clinical and echocardiographic signs of pulmonary hypertension, without structural heart disease. Clinical evidence of pulmonary hypertension was defined as >5% difference between preductal and postductal oxygenation saturation or recurrent decreases (<85%) in arterial oxygen saturation over a period of 12 hours despite optimal treatment of the patient's lung disease. Echocardiographic evidence of pulmonary hypertension was defined as estimated peak systolic pulmonary-artery pressure that was >35 mm Hg or more than two thirds of the systemic systolic pressure, as indicated by the presence of a tricuspid regurgitate jet, a right-to-left patent ductus arteriosus shunt, or a right-to-left artery-level shunt. The study protocol was in accordance with the institutional guidelines for human research, and the patients' parents provided written informed consent for the diagnostic and therapeutic procedures that were required, which allowed the results of the examinations to be used in this study.
Dr Tanaka reviewed the infant and maternal records. Between January 1988 and September 1993, all preterm infants with PPHN received 100% oxygen therapy to treat their respiratory failure. In 1992, iNO therapy was reported to improve respiratory failure in patients with PPHN6; therefore, after institutional ethics committee approval was obtained, iNO therapy was given to all preterm infants with PPHN between October 1993 and December 1999. During the 2 time periods, the patients received similar treatments except for the iNO therapy or 100% oxygen therapy.
NO gas (Taiyo Toyo Sanso, Osaka, Japan) was delivered from an 800-ppm cylinder and was introduced into the afferent limb of the ventilator circuit near the endotracheal tube, which mixed the fixed flow of gas in the ventilator circuit. The flow was adjusted to yield the predetermined NO concentration. The iNO concentration was increased by 10 ppm at 30-minute intervals, with an upper limit of 30 ppm. The response to iNO therapy was evaluated as an increase in Pao2 to >10 mm Hg. Infants who did not show a significant acute response continued to receive iNO therapy at 5 ppm for 12 hours; if there was still no satisfactory response, then the infants were weaned off iNO therapy. Infants who exhibited improvement continued to receive iNO at the minimal level found to be effective (attempts were made to decrease the concentration by reducing it by 5 ppm at 30-minute intervals). At that time, the fraction of inspired oxygen (Fio2) and ventilation were reduced to prevent additional lung injury. When the Fio2 could be decreased to ≤0.4, iNO therapy was terminated by slow weaning over several hours. Methemoglobin levels were measured before iNO therapy, 1 hour later, and then every 8 hours. If the methemoglobin level increased to >2%, then iNO therapy was discontinued. Infants were monitored for signs of increased bleeding (eg, pulmonary hemorrhage, gastrointestinal bleeding, or oozing from venipuncture sites). In addition, cranial ultrasonography was performed before the initiation, within 24 hours whenever possible and then every 24 hours after the initiation, and 24 hours after the final discontinuation of iNO therapy. The median duration of iNO therapy was 19.8 hours (IQR: 29.5–56.0 hours). The oxygenation index, calculated as [100 × Fio2 × mean airway pressure (in centimeters of water)]/Pao2 (in millimeters of mercury), was obtained within 1 hour before and at 1 hour after the initiation of inhalation therapy (iNO or 100% oxygen). The mothers' records were reviewed to determine the presence of maternal fever (≥38°C) during delivery, premature rupture of the membranes (≥24 hours), maternal bleeding, reason for delivery, and prenatal corticosteroid use. Neonatal data, which were obtained from the medical charts, included the number of gestational weeks at birth, gender, birth weight, Apgar scores at 1 and 5 minutes, use of iNO, type of ventilation (high-frequency oscillatory or intermittent mechanical ventilation), and surfactant therapy. All of the subjects' parents were Japanese. All surviving infants were scheduled to be seen by pediatricians at 3 years of age for a complete physical and neurologic examination. Necrotizing enterocolitis was diagnosed during surgery, at autopsy, or on the basis of a finding of pneumatosis intestinalis, hepatobiliary gas, or free intraperitoneal air on radiographs. Pulmonary hemorrhage was diagnosed if a blood-tinged tracheal aspirate was obtained. Intraventricular hemorrhage was graded 0 through 4, according to the highest grade on cranial ultrasonograms, by using the method described by Papile et al.7 CP was defined as abnormal muscle tone in ≥1 extremity and abnormal control of movement and posture.
Categorical variables were compared by using the χ2 test or Fisher's exact test. Differences in the median values between the 2 groups were compared by using the Mann-Whitney U test. Univariate and multivariate logistic regression analyses were used to estimate the odds ratio for the incidence of CP. We calculated the 95% confidence interval for each odds ratio. We limited the number of independent variables in each model to avoid overfitting the data. P values were 2-tailed. A P value of <.05 was considered significant. Statistical analyses were performed by using the SPSS 10.0 software package (SPSS, Chicago, IL).
Of the 61 preterm infants with PPHN, 26 infants died within 3 years after birth. Mortality rates at 3 years after birth were similar for infants treated with iNO and those treated with 100% oxygen (44.1% vs 40.7%; P = .791). The incidences of outcomes in the iNO-treated and 100% oxygen-treated groups were 8.8% vs 29.6% for patent ductus arteriosus, 17.6% vs 25.9% for intraventricular hemorrhage (grade 3 or 4), 8.8% vs 0% for necrotizing enterocolitis, and 11.8% vs 16.0% for pulmonary hemorrhage.
During the 3-year follow-up period, the incidence of CP among patients treated with iNO therapy was lower than that among patients treated with 100% oxygen therapy (12.5% vs 46.7%; P = .054). The baseline clinical characteristics of the study patients according to type of inhalation therapy are summarized in Table 1. Patients who received iNO therapy had lower Apgar scores at 5 minutes and were given surfactant therapy more often, compared with those who received 100% oxygen therapy (Table 1). In univariate logistic analysis, use of iNO therapy was associated with a decreased incidence of CP (Table 2). Although the oxygenation index values before the start of inhalation therapy were similar for the 100% oxygen-treated group (median: 24.0; IQR: 16.0–35.4) and the iNO-treated group (median: 23.3; IQR: 16.0–45.0; P = .695), the oxygenation index 1 hour after the start of inhalation therapy was lower for patients treated with iNO (median: 4.7; IQR: 3.8–7.7) than for patients treated with 100% oxygen (median: 12.5; IQR: 7.6–18.5; P = .013).
We tested several regression models to assess the effect of iNO therapy on the incidence of CP in preterm infants with PPHN. After adjustments for maternal fever during delivery, birth weight, Apgar score at 5 minutes, high-frequency oscillatory ventilation, and surfactant therapy, iNO therapy was associated with a decreased risk of CP, compared with 100% oxygen therapy (Table 3).
Our data demonstrate that, for preterm infants with PPHN, the incidence of CP, during a 3-year follow-up period, among patients treated with iNO therapy showed a trend toward being lower than that among patients treated with 100% oxygen therapy. After adjustments for maternal fever during delivery, birth weight, Apgar score at 5 minutes, high-frequency oscillatory ventilation, and surfactant therapy, iNO therapy was associated with a decreased risk of CP, compared with 100% oxygen therapy.
Previous research had not shown that iNO therapy reduces the risk of CP in preterm infants with hypoxemic respiratory failure. Bennett et al8 reported that, in a randomized, controlled study of 42 preterm infants who were thought to be at high risk of developing chronic lung disease, the rates of CP at 30 months of age were similar in the iNO-treated and control groups. In addition, Mestan et al9 conducted a double-blind, randomized, controlled trial of 138 preterm infants with respiratory failure and found that, although iNO therapy improved cognitive neurodevelopmental outcomes, the rates of CP at 2 years of age were similar in the iNO-treated and control groups. Those results are not consistent with our findings. The study groups in the previous studies were not limited to preterm infants with PPHN, whereas we included only preterm infants with PPHN in the present study. Therefore, the reduced risk of CP with iNO therapy may be limited to such cases. However, because the incidence of CP in the control group was low in the previous studies (14% in the study by Bennett et al8 and 10% in the study by Mestan et al9), compared with our study (47%), the beneficial neurodevelopmental outcomes associated with iNO treatment might have been underestimated in the previous studies; the incidence of CP in the iNO-treated group was 0% in the study by Bennett et al8 and 9% in the study by Mestan et al.9
We did not identify the reasons why iNO therapy decreased the risk of CP in preterm infants with PPHN. However, an in vitro study using rat brain slices showed that hypoxia induced white matter damage mainly through oxidation.10 In comparison with gray matter, white matter contains larger amounts of fat and iron, which are involved in free radical production, and a smaller amount of glutathione, which is an antioxidant; this suggests that white matter has greater susceptibility to oxidative stress. In addition, blood flow to cerebral white matter is extremely low in premature newborns,11 which indicates that cerebral white matter is particularly vulnerable to hypoxia in preterm infants. These results suggest that hypoxia easily can induce white matter damage in preterm infants. We demonstrated that, although the oxygenation index values before the start of inhalation therapy were similar in the iNO-treated and 100% oxygen-treated groups, the oxygenation index 1 hour after the start of inhalation therapy was lower for preterm infants with PPHN treated with iNO, compared with infants treated with 100% oxygen. Therefore, iNO therapy may decrease the risk of CP in these infants through the resolution of hypoxia during a critical phase of neurodevelopment. Alternatively, iNO therapy may affect the brain directly by stimulating neuronal maturation.12–14 However, there is no clear evidence that iNO affects brain development directly.
Our study has some potential limitations. First, we did not conduct a randomized, placebo-controlled trial, for ethical reasons; we conducted a historical cohort study to compare the 3-year incidence of CP in preterm infants with PPHN who received either iNO or 100% oxygen therapy. After adjustment for multiple potential confounding variables, iNO therapy was associated with a decreased risk of CP. In addition, the type of inhalation therapy was determined on the basis of the time of each subject's admission. Therefore, the selection of the type of inhalation can not introduce bias. Second, in our study, we enrolled only preterm infants with PPHN. Therefore, it is not clear whether iNO therapy, compared with 100% oxygen therapy, would decrease the risk of CP in preterm infants with hypoxemic respiratory failure not caused by PPHN. Third, we analyzed a limited number of patients. Therefore, other variables, such as early gestational age, low birth weight, presence of maternal fever during delivery,15,16 premature rupture of membranes of long duration,16–20 maternal bleeding, reason for preterm delivery,21,22 and low Apgar scores at birth,16,17,22 which are thought be risk factors for CP, might not have been identified as being useful for predicting CP in this study. In addition, use of surfactant therapy and use of high-frequency oscillatory ventilation were not associated with a decreased incidence of CP in preterm infants with PPHN. These results might be attributable to a limited number of study subjects. In particular, the use of high-frequency oscillatory ventilation had a high odds ratio, compared with intermittent mechanical ventilation. Therefore, to generalize the results of this study, studies involving a large number of patients are essential. Finally, because we conducted a historical cohort study, we could not explore unknown risk factors for CP.
Our results provide evidence that iNO therapy, compared with 100% oxygen therapy, decreases the risk for CP in preterm infants with PPHN. iNO therapy may protect brain white matter during a critical phase of neurodevelopment and thus reduce the risk of CP in these infants. Additional studies are needed to clarify the mechanism through which iNO therapy decreases the risk of CP in preterm infants with PPHN.
We thank the staff members, patients, and parents who took part in this study. We thank Dr Luba Wolchuk for correcting our manuscript.
- Accepted January 18, 2007.
- Address correspondence to Yuko Tanaka, MD, Department of Neonatal Medicine, Osaka Medical Center and Research Institute for Maternal and Child Health, 840 Murodo-cho, Izumi, Osaka 594-1101, Japan. E-mail:
The authors have indicated they have no financial relationships relevant to this article to disclose.
- Kinsella JP, McQueston JA, Rosenberg AA, Abman SH. Hemodynamic effects of exogenous nitric oxide in ovine transitional pulmonary circulation. Am J Physiol.1992;263 :875– 880
- ↵O'Shea TM, Klinepeter KL, Dillard RG. Prenatal events and the risk of cerebral palsy in very low birth weight infants. Am J Epidemiol.1998;147 :362– 369
- Copyright © 2007 by the American Academy of Pediatrics