Discussion:
In the current multicentre collaborative study on 384 neonates with MAS, 347 were discharged home successfully. Previous studies have made efforts to identify predictors of mortality 2,3. With improved survival there is limited evidence on length of stay in the hospital based on associated morbidities, although few studies have addressed the predictors of severe MAS9-11. The present collaborative study attempted to devise a predictive model/ scores to predict prolonged stay. To our knowledge, this is the first study to determine the predictors of prolonged stay in the hospital in MAS neonates.
The requirement of primary support beyond O2, FiO2 beyond 30% by 1 hour of admission, moderate-severe PPHN and HIE stage 2 or 3 were predictive of prolonged hospital stay (>7 days) in our cohort. The final weighted score in the model was obtained from beta-coefficients and is known to be superior to the traditional way of devising using odds ratio18. The final score thus obtained from the regression had excellent discriminatory power to predict the outcome [AUC 0.82, 95% CI (78-87%)], figure 2. The optimal cutoff (J-point) >21 with varying combinations of any of the 4 predictors had a positive predictive value of 2.6, i.e. 2.6 times more likely to have prolonged hospital stay. The OSI was used to classify the severity of PPHN in the study. The OSI, as evidenced in previous studies, correlates with OI when SpO2>70% and is a non-invasive reliable marker allowing for continuous monitoring for oxygenation status15,19. The conventional definition proposed by Cleary and Wiswell to define severity of MAS involves: (a) mild MAS, requires <40% oxygen for <48 hours, (b) moderate MAS, requires >40% oxygen for >48 hours with no air leak, and (c) severe MAS, requires assisted ventilation for >48 hours and is often associated with PPHN20. The conventional definition has been in use for more than two decades. The severity is primarily based on respiratory support and does not provide guidance on mortality or length of stay in hospital. Moreover with the increasing use of CPAP as primary support8 in the current era, the applicability of the classification is a debatable. The index study identified additional morbidities in the current era with better management of MAS and moreover determined the predictive scores for prolonged stay.
The predictors of severe MAS had varied with previous two published studies9-10. Hofer et al. studied 55 neonates with severe MAS from a cohort of 205 MAS neonates and found acute tocolysis, fetal distress and moderate birth asphyxia as significant risk factors associated with severe MAS on univariate analysis. The independent predictors by regression analysis were not available. The majority of neonates had a long stay with a median stay of 12.5 (1-144) days. Another retrospective cohort by Oliveira et al. enrolled 15 neonates with severe MAS from a cohort of 29 MAS neonates and found the need for surfactant therapy as a predictor of severity. These neonates had a hospital stay of a median (range) of 8 (2-29) days. Previous studies were limited by smaller samples and thus limited predictors of severity. Although the length of stay was similar to the index study, predictors of prolonged stay were not specifically addressed.
The associated morbidities in the current study were similar to other published studies5, 9, 10-12 in MAS, except Oliveira et al10 who had higher surfactant requirement in the sample of 29 neonates. CPAP is increasingly used as the primary respiratory support in MAS5,8. A randomized trial by Pandita et al.5 evaluated the role of CPAP as primary respiratory support in MAS and found a reduction in the need for mechanical ventilation, sepsis, shock and PPHN in the CPAP group. CPAP as primary respiratory support was used in 184 (53%) neonates with invasive ventilation in 64 (19%) in our study. The current study also had a lower incidence of late-onset sepsis in 2 (0.6%). Despite the use of CPAP as primary support, there was no increased incidence of air leaks compared to similar studies5, 8. Antibiotics are commonly prescribed empirically in neonates with MAS owing to the sickness. However, antibiotics have not affected the duration of hospital stay or mortality in MAS as per meta-analysis of 4 RCTs22. The current study had 7 (2%) neonates with early-onset sepsis and 2 (0.6%) neonates with late-onset sepsis. However, antibiotics were started for 322 (93%) neonates empirically. Considering the low incidence of sepsis, there is a need for quality improvement strategies for optimizing antibiotic usage in MAS. Steroids reduce pulmonary inflammation on histology and improve oxygenation in animal studies23, 24. Two recently published systematic reviews25, 26 have evaluated the role of steroids in MAS. There was no mortality benefit, but reduced hospital stay with nebulised budesonide was observed (low certainty of evidence). The current study used dexamethasone in post-extubation settings, and nebulised steroids or methylprednisolone were not used. There is a need for further large well-designed study evaluating the role of steroids in severe MAS.
The strengths of the study are large sample size from a multicentric cohort, standardized criteria for respiratory management and devising a simple clinical prediction score. Difference in the level of nursing care, doctor:staff ratio, different patient demographics could have altered outcomes in MAS at each centres. Despite these limitations, the study has proposed a prediction score for a prolonged stay in MAS applicable across low and middle-income countries. This scoring could help in optimizing the strategies for treating MAS in the initial few hours. Validation of the scoring would be required in different settings before broader application of the scoring. The prediction score can serve as a tool for risk stratification for future clinical trials. There is a need for QI strategies on optimizing antibiotic use and respiratory management. To conclude, more than two-thirds of neonates with MAS had prolonged stay. The primary support beyond oxygen, Fio2 requirement >30%, Moderate to severe PPHN, HIE stage 2 or 3 were predictive of prolonged stay in neonates with MAS.