Downes Scoring for Respiratory Distress [4]

All babies were treated according to existing standard management protocol, and requisite investigations were done whenever needed. They were started on intravenous fluids, with the amount and composition depending on the weight and postnatal age. Enteral feeds were introduced when appropriate using expressed breast- milk. All ventilated babies were started on antibiotics. Cefotaxime and gentamicin were used as the first-line antibiotics. Antibiotics were changed according to the clinical condition and culture reports. Surfactant (Survanta) of bovine origin was administered to preterms with hyaline membrane disease (HMD) in doses as recommended by the manufacturers.

Statistical analysis was done using SSPS software (Statistical Packages for the Social Sciences) version 17 for Windows (SPSS Inc., Chicago, IL, USA). Chi-square test was used for categorical variables and Student’s unpaired t test for continuous variables. p<0.05 was considered significant for all the tests. The variables identified as significant by univariate analysis were further analyzed using multivariate logistic regression.

An analysis of ventilatory requirements revealed that the maximum peak inspiratory pressure (PIP) requirement during the course of ventilation was significantly higher in the nonsurvivors compared to that of the survivors. Although the non-survivors required higher FiO2 and maximum positive end-expiratory pressure (PEEP) than the survivors, these differences did not reach statistical significance. When maximum ventilatory requirements during the course of ventilation were analyzed in relation to outcome in individual conditions, it was found that increased FiO2 requirement had a significant difference in outcome in cases ventilated for pneumonia (p=0.007) and those ventilated for sepsis (p=0.044). In babies ventilated for other indications, there was no significant difference (Table III).

Table I. Percentage Survival in Relation to the Studied Parameters

Shock, sepsis/pneumonia, disseminated intravascular coagulation (DIC), pulmonary hemorrhage, air leak syndromes, and persistent pulmonary hypertension in newborns (PPHN) were the complications noted to have occurred during the course of ventilation. Of these, shock was the commonest followed by sepsis/ pneumonia. While the presence of shock, ventilator-associated pneumonia (VAP)/sepsis and PPHN did not have a significant bearing on the outcome, DIC, pulmonary hemorrhage and pneumothorax significantly hampered survival (Table IV). On applying multivariate logistic regression, none of the factors identified as significant by univariate analysis was found to be an independent predictor of mortality.

A total of 25 cases had HMD, among whom 23 received surfactant. Both the babies who did not receive surfactant expired, resulting in 100% mortality in that group.

Table II. Profile of Survivors and Non-Survivors

The commonest indication for ventilation in our series, MAS, had the third best outcome, with a survival of 60.7%. MAS had the best outcome in the series by Malhotra et al.[5] and Riyas et al.[1], with 100% and 63.6% survival, respectively. Poorest outcome was seen in the series by Singh et al.[7] and Karthikeyan et al.[9], where all the babies who were ventilated for MAS expired. Babies with apnea of prematurity in our series had a survival rate of only 60%, as many of them succumbed to superimposed infections during ventilation.

Survival of 52% for HMD in the present series is comparable to most other series. However, Singh et al.[7] and Schreiner et al.[12] had reported better survival in HMD in their series. Natural surfactant (bovine origin) was used in our study selectively for infants with established RDS as rescue therapy and in most cases as late rescue. Whilst natural surfactant is associated with a marginally higher risk of intraventricular hemorrhage compared to synthetic surfactant, it better reduces mortality and the incidence of pneumothorax[13]. Compared to selective use in babies with RDS, prophylactic administration of surfactant produces a better reduction in the incidence of RDS, pneumothorax, pulmonary interstitial emphysema and bronchopulmonary dysplasia, and death[14],[15]. Because of resource constraints, we were unable to use surfactant prophylactically, which might also explain the higher mortality for HMD in our series compared to western data. Treating expectant mothers antenatally with corticosteroids when a preterm birth is anticipated has been shown to reduce mortality, RDS, intraventricular hemorrhage, necrotizing enterocolitis, need for ventilatory support and intensive care, and infections in the first 48 hours of life[16]. The use of antenatal corticosteroids has been shown to reduce the need for prophylactic and early rescue surfactant[17]. Furthermore, combined use of antenatal steroids and surfactant has been proven to be more beneficial than either therapy alone[18]. In our study, the influence of antenatal steroids plus surfactant versus surfactant alone without antenatal steroids, on the outcome, was examined. Contrary to our expectation, the difference was not statistically significant, probably because of the small number of cases (only 5 babies) who had received both antenatal steroids and surfactant.

Sepsis caused the worst outcome, with only 46.1% survival. Maiya et al.[6] reported 100% survival in sepsis cases while Karthikeyan et al.[9] and Mittal et al.[2] reported 66.6% and 64.9% survival, respectively. Sepsis had a uniformly poor outcome in all other studies[7],[11].

Females were found to have better survival (65.3%) than males (52%), but the difference was not statistically significant. A similar trend was observed in the series by Riyas et al.[1] and Lindroth et al.[19]

We found that the mean gestational age of the newborns who survived was significantly higher than that of the non-survivors, which is consistent with the findings of most other authors[8],[9],[11]. It was observed that increasing birth weight was associated with better survival. This was consistent with the findings of several other authors[1],[5]-[9],[11]. Rich et al.[20] in their study on very low birth weight infants found low birth weight as the best predictor of longer duration of ventilatory support.

The Downes score provides an objective way of assessing any improvement or deterioration in the respiratory status of the baby. In regions with limited resources, where objective measurements are not available, this scoring system provides immense advantage over unsystematic clinical evaluation. Downes and co-workers found that, although the initial score did not bear any relation to the outcome, the score at 12-18 hours provided an estimate of the prognosis, with a higher score indicating poorer prognosis[4]. The majority of the neonates in our series were ventilated immediately after respiratory distress was recognized. Thus, we were not able to study the Downes score at 12-18 hours and beyond. Hence, we tried to correlate the initial score at admission with the outcome. The score was higher among non-survivors, but the difference was not statistically significant.

Arafa et al.[21] in their study found a 5 minute Apgar score of <7 to be significantly associated with mortality. In our series, the mean 5 minute Apgar of survivors was 7.1, whereas that of the babies who expired was 6.52, but the difference was not statistically significant. We compared the pH on admission with the outcome in all the ventilated babies, and found that the mean pH of the survivors was slightly higher than that of the non-survivors and also that babies with a pH of >7.3 had a better survival than those with a pH of <7.299, but the difference was not significant. This was similar to the observation by Mathur et al.[22]

Table III. Comparison of Ventilatory Parameters Between Survivors and Non- Survivors

Table IV. Analysis of Survival in Relation to Complications

Maximum ventilatory requirements during the course of ventilation were analyzed with the outcome as a whole and separately with each individual indication. The mean of the maximum PIP requirement of the non-survivors was significantly higher than that of the survivors. Mathur et al.[22] observed a similar trend in their study, but the difference was not statistically significant. When we compared the maximum FiO2 requirements with the outcome, we found that the non-survivors had actually required a greater mean maximum FiO2 than the survivors, but the difference was not statistically significant. In babies ventilated for pneumonia, a maximal FiO2 requirement of 60-80% was associated with significantly higher mortality compared to those who had required 40-60%. In babies with sepsis, a requirement of 40-60% resulted in significantly higher mortality than FiO2 of 20-40%. Mathur et al.[22] in their series identified an initial FiO2 requirement of more than 60% as a significant independent predictor of mortality.

Pulmonary hemorrhage occurred in 7 babies, among whom only 1 survived, and the difference was statistically significant. Four of the seven were cases of HMD and developed pulmonary hemorrhage following surfactant administration, a recognized complication of surfactant treatment. Six babies in the series by Karthikeyan et al.[9] developed pulmonary hemorrhage, 5 of them following surfactant administration.

Among all complications, the one that had the greatest association with mortality was air leak syndrome. All 5 babies who had air leaks in our series died. In the 1970s, it was a very common complication, occurring in 20% of all cases and 31% of cases with MAS[12]. In our series, the incidence of air leak syndrome was significantly lower compared to most other studies[6]-[8],[10],[11] despite the presence of more babies with MAS (who are more prone to air leaks), which can probably be attributed to the judicious use of pressures and early attempts at weaning. Although our center uses high-frequency ventilation in selected neonates, in this study, we had included only those babies ventilated by conventional modes. Studies have shown that there is no significant difference in outcome between the conventional and high-frequency modes, provided an optional lung volume strategy is used[23],[24]. Utilizing noninvasive modes of ventilation like nasal continuous positive airway pressure (CPAP), nasal intermittent positive pressure ventilation (NIPPV) and methods like mandatory minute ventilation have been shown to decrease some of the long-term complications associated with mechanical ventilation as well as the incidence of post-extubation failure[24],[25].

In conclusion, our study reaffirmed that increasing gestational age and birth weight are associated with a better outcome. The mean maximum PIP requirement was significantly higher in the non-survivors compared to the survivors. Although the incidence of DIC, pneumothorax and pulmonary hemorrhage was significantly higher in the non-survivors, none of these factors was found to be independently predictive of mortality. Earlier referral of highrisk pregnancies to tertiary care facilities might increase the use of antenatal steroids, enable better planning of deliveries, provide better outcomes in HMD, and reduce the incidence of MAS in developing nations. Better resource availability allowing earlier and prophylactic use of surfactant might have an impact on the outcome of HMD in our region of the world.