NCPAP indicates nasal continuous positive airway pressure.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Pandita A, Murki S, Oleti TP, et al. Effect of Nasal Continuous Positive Airway Pressure on Infants With Meconium Aspiration Syndrome: A Randomized Clinical Trial. JAMA Pediatr. 2018;172(2):161–165. doi:10.1001/jamapediatrics.2017.3873
Can the use of nasal continuous positive airway pressure early in the course of meconium aspiration syndrome reduce the subsequent need for mechanical ventilation in newborns?
In this randomized clinical trial that included 135 infants with moderate or severe meconium aspiration syndrome, 2 infants (3%) supported with nasal continuous positive airway pressure required subsequent mechanical ventilation in the first 7 days of life vs 17 infants (25%) who were supported with hood oxygen, a significant difference.
There is a possibility for reduction in the need for subsequent mechanical ventilation if infants with moderate or severe meconium aspiration syndrome are initially supported with nasal continuous positive airway pressure in place of hood oxygen.
Nasal continuous positive airway pressure (NCPAP) as a primary respiratory therapy in meconium aspiration syndrome (MAS) has not been studied extensively. Nasal continuous positive airway pressure, when applied in newborns with MAS, may resolve atelectasis by sufficiently expanding partially obstructed small airways and stabilizing the collapsing terminal airways to enhance oxygen exchange.
To compare NCPAP vs standard care in neonates with moderate to severe respiratory failure due to MAS in reducing the need for invasive ventilation.
Design, Settings, and Participants
This multicenter open-label, parallel-group (1:1 ratio) randomized clinical trial was conducted from August 5, 2014, to May 26, 2016. Data were collected from 3 tertiary care neonatal intensive care units. All infants admitted with respiratory distress, defined as Downe score greater than 4 and peripheral capillary oxygen saturation less than 90%, were assessed for study eligibility if the chest radiograph was suggestive of MAS and they met the other inclusion criteria: gestation longer than 35 weeks, a birth weight greater than 2000 g, and born through meconium-stained amniotic fluid.
Infants were randomly assigned to either NCPAP or standard care (5-10 L/min hood oxygen).
Main Outcomes and Measures
The primary outcome was the need for mechanical ventilation in the first 7 days of life.
After excluding 14 infants, 67 infants were randomized to bubble NCPAP and 68 infants to standard care. Baseline characteristics were similar between the 2 groups. Infants randomized to the bubble NCPAP group needed mechanical ventilation less frequently in the first 7 days of life compared with standard care (2 [3.0%] vs 17 [25.0%]); odds ratio, 0.09; 95% CI, 0.02-0.43; P = .002). The need for surfactant (3 [4.5%] vs 11 [16.2%]; odds ratio, 0.24; 95% CI, 0.05-0.87) and culture-positive sepsis (4 [6.0%] vs 13 [19.0%]; odds ratio, 0.28; 95% CI, 0.09-0.93) were higher in the standard care group. There was an increased duration of oxygen therapy (median [interquartile range], 45.5 [28.0-78.3] vs 26 [20.0-48.0] hours; P = .001) in the standard care group. In the NCPAP group vs standard care group, incidence of persistent pulmonary hypertension (9 [13%] vs 19 [28%]; odds ratio, 0.42; 95% CI, 0.17-1.01) and duration of hospital stay (median [interquartile range], 5.0 [4.0-8.8] vs 4.0 [4.0-6.0] days; P = .14) were similar.
Conclusions and Relevance
Bubble NCPAP in comparison with standard care for infants with MAS reduces the need for mechanical ventilation in the first 7 days of life.
Clinical Trial Registry, India Identifier: CTRI/2015/03/005631
Meconium staining of amniotic fluid (MSAF) occurs in 10% to 15% of all deliveries. Infants born with MSAF are 100-fold more likely to develop substantial respiratory distress than infants born through clear amniotic fluid. About 1.5% to 8.0% of infants born through MSAF develop meconium aspiration syndrome (MAS).1,2 Approximately 30% to 50% of infants with MAS have severe MAS, defined as the need for mechanical ventilation (MV).3-5 The optimum methods of providing respiratory support to infants with MAS are not known.6,7 Though ventilation is a life-saving measure, ventilation in itself results in ventilation-induced lung injury, and the need for MV translates into prolonged hospital stays, increased burden on the health care system, and increased treatment costs.8,9 Nasal continuous positive airway pressure (NCPAP) as a primary respiratory therapy in MAS has not been studied extensively. When NCPAP is applied in newborns with MAS, it may resolve atelectasis by expanding partially obstructed small airways and stabilizing the collapsing terminal airways to enhance oxygen exchange. Nasal continuous positive airway pressure in MAS may be used as a substitute or as a bridge to MV. Therefore, we conducted this multicenter randomized clinical trial to evaluate the efficacy of NCPAP in comparison with standard care (5-10 L/min hood oxygen) in reducing the subsequent need for MV in newborns with MAS.
This open-label, parallel-group (1:1 ratio) randomized clinical trial was conducted from August 5, 2014, to May 26, 2016, in India at 3 tertiary care neonatal intensive care units. Each neonatal unit has an average of 1200 admissions per year. The study protocol was approved by the institutional review board from each of the 3 participating centers and was registered with clinical trial registry India. Written informed consent was obtained from the families of eligible infants within an hour of admission to the neonatal unit.
All infants with gestation more than 35 weeks and birth weight greater than 2000 g admitted to the neonatal intensive care unit in the first 24 hours of birth were assessed for study eligibility if they were born through MSAF, had respiratory distress (defined as Downe score >4 and peripheral capillary oxygen saturation [SpO2] <90% on room air), and if chest radiograph at admission was suggestive of MAS (hyperinflated lung fields with diffuse nonhomogenous opacity, reticulonodular pattern or low-volume lungs with reticulogranularity, and air bronchograms). All pediatricians and nurses in the participating hospitals were trained in the use of the Downe score10 (trial protocol in the Supplement) for assessment of respiratory distress. The exclusion criteria included intubation at admission, severe asphyxia (5-minute Apgar score <3 and cord potential of hydrogen level <7.00), pneumothorax and/or air leak (visible on the admission chest radiograph), and major malformations. Chest radiograph and arterial blood gas were performed in all infants prerandomization.
Eligible neonates were randomized (stratified for center) to either bubble NCPAP or standard care, using a 1 to 1 ratio in randomly permuted blocks of 2 or 4. Random numbers were generated using a web-based computer program (https://www.randomizer.org/). Individual group assignments were placed in a serially numbered opaque sealed envelope that was opened only after obtaining consent from the parents. The statistician analyzing outcomes was masked to the group allocation.
Infants randomized to the NCPAP group were started on the bubble NCPAP generator (Fisher and Paykel Healthcare, Inc) using short binasal prongs (Hudson Respiratory Care, Inc or Fisher and Paykel Healthcare, Inc). The starting NCPAP was 5 cm of water. Nasal continuous positive airway pressure and fraction of inspired oxygen (FiO2) were adjusted to maintain target saturations between 90% to 95%. The neonate was weaned from NCPAP when the SpO2 was consistently greater than 90%, the FiO2 was less than 0.25, and respiratory distress was passive (respiratory rate <60 breaths/min, no or mild retractions, and no grunting). After weaning from NCPAP, oxygen if needed, was administered either with a hood or with binasal oxygen prongs. Failure of NCPAP was defined as SpO2 levels less than 90% on maximum NCPAP pressure of 6 cm of water and FiO2 of 1.0. All infants with NCPAP failure were intubated and placed on MV.
Infants randomized to the standard care group were started on hood oxygen, administered at 5 to 10 L/min. Infants whose hood oxygen failed (SpO2 < 90% for more than 15 minutes on FiO2 of 1.0) were rescued either with NCPAP or MV but at the discretion of the treating team. Those rescued with NCPAP qualified for MV if SpO2 levels were less than 90% consistently on a maximum NCPAP of 6 cm of water and FiO2 of 1.0. After extubation or after weaning from NCPAP, oxygen if needed was administered either with a hood or with binasal oxygen prongs.
Apart from specified criteria for MV and irrespective of allocation group, any infant found to have persistent shock (inotrope >10 μg/kg/min), partial pressure of carbon dioxide greater than 60 mm Hg with a potential of hydrogen level less than 7.20, or a new-onset pneumothorax with hemodynamic instability was mechanically ventilated.
All enrolled infants were actively assessed for any cardiorespiratory dysfunction using vital signs (heart rate, blood pressure, oxygen saturations, and urine output), Downe score, and echocardiography. Management of comorbid conditions such as pulmonary hypertension, shock, seizures, renal dysfunction, fluid, electrolyte, acid and base imbalances, use of high-frequency oscillation, or sildenafil were as per the existing unit protocols. Surfactant was given in infants requiring MV (FiO2 ≥0.50 for >2 hours) and in those with low-volume lungs on the chest radiograph (<7 posterior intercostal spaces). All the relevant perinatal data and neonatal data until discharge or death were collected prospectively in special forms designed for this trial.
Diagnosis of pulmonary hypertension was based on clinical and echocardiographic criteria. Infants diagnosed with pulmonary hypertension were managed with inotropes, respiratory support (NCPAP or MV), and/or sildenafil. Inhaled nitric oxide was not used. Shock was defined as presence of clinical features and need for dopamine or dobutamine at a dose exceeding 10 µg/kg/min.
The primary outcome was the need for MV in the first 7 days of life. The secondary outcomes were death, pneumothorax, need for surfactant, pulmonary hypertension, culture-positive sepsis (onset of sepsis >72 hours of birth), duration of oxygen, and duration of hospital stay.
Based on the available evidence, we assumed 30% incidence of MV in the standard care group,3,4 and for an absolute reduction of 20% using NCPAP (based on a pilot study10), the sample needed to recruit was 66 infants in each group with an α error of .05 and a power of 80%. Comparison between the study groups for discrete variables was done using logistic regression adjusting for the center. For continuous variables (duration of oxygen and hospital stay) where distribution was skewed, normality was achieved by log transformation. For obtaining comparative estimates, linear regression was performed for the log-transformed outcomes to adjust for centers. The coefficient and its 95% confidence interval was exponentiated to obtain the percentage reduction in the dependent variable between the 2 groups adjusted for center.
During the study period, 149 infants were assessed for eligibility, and 14 were excluded. Among the exclusions, 9 infants had severe birth asphyxia, 3 were intubated at birth for poor respiratory efforts, and parents of 2 infants refused consent. Sixty-seven infants were randomized to the bubble NCPAP group and 68 infants to the standard care group. Both the groups were comparable for all the baseline variables including gestation, birth weight, sex, growth restriction, median Apgar score at 1 and 5 minutes, and maternal characteristics (Table 1). The proportion of infants vigorous at birth was comparable between the 2 groups. The mean (SD) postnatal age at randomization was 1.2 (1.0) hours in both the groups. Arterial potential of hydrogen, base excess, and severity of respiratory distress (median [interquartile range] Downe score 6 [5-7] vs 5 [5-7]; P = .76) at randomization were similar between the NCPAP and standard care groups, respectively (Figure).
The need for MV in the first 7 days of life was significantly lower in the infants randomized to the NCPAP group (2 [3%] vs 17 [25%]; odds ratio, 0.09; 95% CI, 0.02-0.43); P = .002). For every 5 newborns with MAS started on early NCPAP compared with hood oxygen, one would avoid ventilation in 1 neonate (risk difference = 22.0%; number needed to treat = 4.5). In the NCPAP group, 1 infant (at postnatal age of 49 hours) was ventilated for increasing oxygen requirement and the other (at postnatal age of 7 hours) for pneumothorax. In the standard treatment group, 2 infants received MV without a prior trial of NCPAP and 17 infants were placed on NCPAP (15 eventually required MV) for worsening respiratory failure. A total of 4 patients among the standard care group further required high-frequency ventilation for failure of conventional ventilation. None of the infants in the NCPAP group required high-frequency ventilation. The indications for ventilation in the standard care group were increasing oxygen requirement (11 infants) and sepsis with shock (6 infants). Among the infants whose hood oxygen failed (n = 19), the mean (SD) age of shifting to NCPAP or MV was 4.5 (5.0) hours of birth and for ventilation in the NCPAP group (n = 2) was 7.5 (6.0) hours of birth.
A total of 1 term infant randomized to standard treatment group died of massive pulmonary hemorrhage, secondary to severe sepsis and disseminated intravascular coagulopathy. In the NCPAP group, 1 infant needed MV for pneumothorax. More infants randomized to the standard care group needed surfactant and had culture-positive sepsis. Duration of oxygen was significantly higher in the standard care group (Table 2).
In this trial, we randomized neonates with MAS to NCPAP or hood oxygen at admission to neonatal intensive care unit. The mean (SD) age of starting NCPAP was 1.2 (1.0) hours after birth. Starting NCPAP early reduced the subsequent need for MV. Many recent studies report limited use of NCPAP for newborns with MAS. In the study on the delivery room management of vigorous newborns born with MSAF, among the 62 newborns with MAS, only 6 infants (10%) received NCPAP.11 In a trial on oropharyngeal suction at delivery of the head in MSAF, Vain et al12 did not report use of NCPAP for MAS. Understanding the pathophysiology of MAS, early NCPAP would improve functional residual capacity leading to better recruitment of alveoli; prevention of secondary surfactant deficiency; and open airways (avoids atelectasis and air trapping) to prevent subsequent development of hypoxia, acidosis, and pulmonary hypertension. All these may have contributed to the reduced need for MV in the group supported with early NCPAP. However, one may argue that nearly 75% of newborns with MAS would not require any respiratory support apart from oxygen as in the standard care group. We can only counter this argument by stating that increased or excess use of NCPAP is acceptable to avoid even a few newborns going on to MV especially in resource-limited settings. The reduced need for MV in the NCPAP group in comparison with our pilot observation13 may be owing to the very early use of NCPAP in infants with MAS.
The need for surfactant was higher in the standard group. Delay in initiation of NCPAP may have resulted in secondary surfactant deficiency, persistent respiratory distress and/or increased work of breathing, and may be the reason for higher use of surfactant in this group. Increased need for MV and prolonged hospital stay in the standard care group may be the reasons for increased incidence of sepsis. The reported incidence on the use of NCPAP or MV for newborns with MAS varies from 30% to 50%.3-5 In the trial by Vain et al12 among the 99 infants with MAS, 42 infants (42%) required MV, and in the trial on vigorous infants with MAS by Wiswell et al,11 the incidence of MV was 39% (24 of 62 infants with MAS). The lesser need for MV in the standard care group in our trial (25%) may be explained by exclusion of infants with asphyxia and inclusion of predominently inborn infants (90%). Prespecified protocols on the use of NCPAP or MV may also have reduced individual variations among the pediatricians and the institutes.
Reducing the need for MV in newborns with MAS has far-reaching implications, especially in low- and middle-income countries. Perinatal asphyxia and MAS are the common reasons for MV in these low- and middle-income countries.14 High mortality (approximately 40%), high incidence of sepsis, and nonavailability of quality ventilation services are the major bottlenecks for the use of MV in developing countries.15,16 When applied early, NCPAP is a safe option to manage infants with MAS and to prevent up-transfers to already overburdened level III and/or tertiary care centers and also reduce cost of care.17 However, dependence on imported NCPAP devices, lack of an ideal interface, nonavailability of round-the-clock air or oxygen supply, backup ventilation, lack of awareness and expertise among physicians, and inadequately trained nursing staff are the major challenges.18
Exclusion of infants with severe perinatal asphyxia and preponderance of inborn newborns may be reasons for lesser mortality in our study cohort compared with that reported by others. Although 1 infant in the NCPAP group had pneumothorax, the overall incidence of pneumothorax in this group is lower than that reported by others. Vain et al12 reported 6% (6 of their 99 patients) and Bhat and Rao19 reported 27% incidence of air leaks in their patients with MAS.
A relatively large sample size, multicenter enrollment, predominantly inborn infants, and standardized criteria for respiratory management are the main strengths of this study. Initiation of NCPAP or MV at the discretion of the managing clinicians, inability to blind the intervention, and use of short-term hospital-based outcomes are the main limitations of this study. The results of this study need to be replicated in other settings and with studies recruiting outborn infants.
Starting early low-level NCPAP in comparison with hood oxygen in neonates with MAS reduces the subsequent need for MV. For every 5 newborns with MAS started on NCPAP, 1 newborn is protected from MV.
Corresponding Author: Srinivas Murki, DM, Fernandez Hospital, Hyderabad, India (firstname.lastname@example.org).
Accepted for Publication: August 30, 2017.
Published Online: December 4, 2017. doi:10.1001/jamapediatrics.2017.3873
Author Contributions: Drs Pandita and Murki had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Murki, Pandita, Tandur.
Acquisition, analysis, or interpretation of data: Murki, Pandita, Oleti, Kiran, Narkhede, Prajapati.
Drafting of the manuscript: Murki, Pandita, Oleti, Prajapati.
Critical revision of the manuscript for important intellectual content: Murki, Pandita, Oleti, Tandur, Kiran, Narkhede.
Statistical analysis: Murki, Pandita, Oleti, Kiran, Prajapati.
Administrative, technical, or material support: Murki, Pandita, Tandur, Narkhede.
Study supervision: Murki, Oleti, Prajapati.
Conflict of Interest Disclosures: None reported.
Create a personal account or sign in to: