Surfactant therapy for meconium aspiration syndrome in neonates: a systematic overview of systematic reviews and recent clinical trials

Meconium aspiration syndrome (MAS) is defined as respiratory distress in a neonate born through a meconium-stained amniotic fluid, having characteristic radiological changes whose symptoms cannot be otherwise explained [1]. Although MAS is as yet a noteworthy cause of mortality and morbidity in developing countries [2], in a study conducted in an urban Pakistanis, 27.3% of neonatal mortality was estimated to have a history or evidence of meconium passage during delivery [3]. A study conducted in China found that among all neonates, MAS becomes a source for 10% of cases of respiratory failure and is associated with up to 39% of morbidity and high mortality [4]. The pathophysiology of this condition is characterized by acute airway obstruction, surfactant dysfunction or inactivation, chemical pneumonitis by aspirated meconium and pulmonary hypertension induced right-to-left extra-pulmonary shunting [5]. Treatment of MAS in neonates involves supportive care-oxygen therapy, assisted ventilation, inhaled nitric oxide and, if available, extracorporeal membrane oxygenation (ECMO) [6,7]. A Cochrane review assessed bolus surfactant (BS) therapy in MAS, which indicated that the surfactant therapy decreases progressive respiratory failure requiring ECMO [8]. A study with three case reports found that lung lavage with diluted surfactant appeared to improve the outcomes among infants with MAS [9]. Our literature suggested that there are several systematic reviews (SRs) published in the past decade, citing more than ten randomized controlled trials (RCTs) and observational studies. When evidence is depressed among multiple SRs on relevant aspects of a topic of interest, systematic overviews of SRs are offer a solution to the information overload in literature, enabling enhanced access to targeted information of interest. Within the context of guiding the decision making on the use of surfactants for MAS, there is a need to systematically bring together, assess and summarize overall evidence available so far, especially as there is no available robust and comprehensive evidence to support the effective surfactant modality. Here, in this review, we aimed to summarize all the existing SRs and the RCTs that were not part of SRs, and to assess their quality.

This is a systematic overview that follows the preferred reporting items for systematic reviews and meta-analyses (PRISMA) checklist recommendations for reporting (Supplementary Material 1).

We searched EMBASE, PROQUEST and PubMed databases with the variations of the key terms of ‘meconium’, ‘MAS’ and ‘surfactant’ since inception to January 2020 for identifications of literature on this topic. Search strategies can be seen in Supplementary Material 2. No restrictions were imposed on the search. In addition, to identify potentially missed relevant literature, we have searched Google Scholar and references of relevant reviews (grey literature).

Studies that were SRs or comparative clinical trials on neonates with MAS that were treated with surfactants and assessed for mortality and morbidity were considered for inclusion. Any included SR, identified as SR or meta-analysis, was one that systematically identified the evidence about using the surfactants; summarized the different outcomes from different sources and; synthesized summative evidence about each of the different outcomes. Included RCTs are the ones that were not included in any of the included SRs in the current study.

We have excluded studies such as expert opinions, previous SRs of current/updated ones, narrative reviews, conference abstracts and editorials.

Two reviewers independently screened title/abstracts for inclusion and exclusion and then the eligible studies were subjected to the full-text screening based on the aforementioned definition of an SR. The same two reviewers conducted full-text screening and any discrepancies were resolved by consulting the senior author.

Two authors independently extracted the data of interest from each included SR. The extracted data related to the study characteristics, literature search strategies, patient characteristics, intervention, comparator, outcome measures, duration of follow-up, effect estimates, surfactant type, formulation and the number of doses. In addition, we also extracted the relevant data from recently published RCTs that are not included in the most recent SR.

Our literature search yielded 1797 studies from all databases. After the removal of duplicates (n = 591), we screened the titles and abstracts and removed the irrelevant articles (Figure 1). The relevant articles were subjected to full-text screening and finally, we identified three SRs [8,13,14] and two RCTs [15,16].

The main characteristics of the included SRs are presented in Table 1 and the recent RCTs are presented in Table 2. All three SRs included randomized trials that used accepted methods of randomization. However, the SR conducted by Choi et al. also included eight nonrandomized studies (NRSs) [13]. Studies in one SR [13] used surfactant lung lavage (SLL) and were not blinded, whereas three of the studies that were included in the Cochrane SR [8], which used BS, were blinded, except one RCT [17]. Three of the RCTs in the third SR [14], by Natarajan et al. were on antibiotic use in MAS [18–20]. The remaining studies in this SR, which studied the surfactant use, used both SLL and BS.

The identified SRs included 13 unique RCTs (1106 patients) and the eight NRSs [21–28] (178 patients). All these studies reported hospital admission outcomes.

Out of the ten RCTs on surfactant for MAS, four used SLL and the remaining studies tested BS [15,16,29–36]. Among the four RCTs that tested SLL, one tested a synthetic surfactant (Surfaxin), while the other studied bovine surfactant for lavage. In contrast, all BS studies evaluated only natural or animal surfactants that were made from either bovine or porcine.

An SR conducted in 2012 by Choi et al. [13] had included both RCTs and NRSs studies that assessed the effects of surfactant lavage therapy for MAS. This review included studies that were clinically heterogeneous in terms of the severity of the disease, method of surfactant lavage, initial intervention time and combined treatment modalities. Meta-analysis of two RCTs indicated that the surfactant lavage significantly decreased death or the need for ECMO (risk ratio [RR]: 0.34; 95% CI: 0.11–0.99) with no heterogeneity, however, no statistical significant difference was found for the pneumothorax outcome (RR: 0.39; 95% CI: 0.08–1.95).

This review also included eight NRSs and the meta-analysis of studies with available data indicated that the surfactant lavage had a significant effect on air leaks (RR: 0.52; 95% CI: 0.28–0.96; six studies), pneumothorax (RR: 0.45; 95% CI: 0.23–0.89; five studies) and death or the need for ECMO (RR: 0.35; 95% CI: 0.13–0.94; six studies). These results are inconsistent with those from RCTs. However, the allocation methods in these studies with concurrent control may be prone to selection bias.

A Cochrane review [8] was conducted to determine the efficacy of surfactant administration in the treatment of late preterm and term infants with MAS in RCTs. The meta-analysis of four trials (326 infants) showed no statistically significant effect on mortality (RR: 0.98; 95% CI: 0.41–2.39; typical risk difference: 0.00; 95% CI: 0.05–0.05) with no heterogeneity. There were no statistically significant reductions in the secondary outcomes of duration of assisted ventilation, supplemental oxygen, pneumothorax, air leaks, chronic lung disease and the need for oxygen at discharge, but not for the hospital stay (median duration: 8 days; 95% CI: 14–3) and need for ECMO (RR: 0.64; 95% CI: 0.46–0.91; typical risk difference: 0.17; 95% CI: 0.30–0.04; two RCTs).

A recent SR conducted by Natarajan et al. [14] studied surfactant therapy and antibiotics in neonates with MAS. Out of the 11 RCTs, eight studies assessed the effects of the use of surfactant and the other three were on the use of antibiotics. Both SLL and BS methods did not reduce the risk of mortality (RR: 0.38; 95% CI: 0.09–1.57; two studies; and RR: 0.80; 95% CI: 0.39–1.66; five studies, respectively), however, both methods reduced the duration of hospital stay (mean difference: 2.0; 95% CI: 3.66– −0.34; one study; and RR: 4.68; 95% CI: 7.11–2.24 days; four studies) and duration of mechanical ventilation (mean difference: 1.31; 95% CI: 1.91–0.72; two studies; and mean difference: 5.4; 95% CI: 9.76–1.03 days; five studies). There was no significant reduction with the use of antibiotics for MAS in the risk of mortality (RR: 1.72; 95% CI: 0.22–13.31; three studies), sepsis (RR: 1.31; 95% CI: 0.34–5.07; three studies) and duration of hospital stay and duration of oxygen therapy.

Among the studies included in these SRs, only the Chinese Study Group 2005 [17] and Lotze et al. (1998) [32] studied incidence of complications, which indicated no significant difference between the studied groups. The complications monitored were technical, neurologic, pulmonary, hemorrhagic, cardiac complications and proven sepsis.

An RCT [15] conducted in Turkey included newborns with MAS (diagnosed according to the criteria of: evidence of meconium passage at or before delivery, presence of respiratory distress after <2 h birth, chest radiography typically suggesting aspiration of meconium). Patients in lavage group (n = 17) received 30 ml/kg of diluted porcine surfactant and bolus group (n = 16) received porcine surfactant (100 mg/kg) in repetitive dose endotracheally. Lung lavage did not show any advantage over bolus therapy on duration of respiratory support, high-frequency ventilation or inhaled nitric oxide requirement. A total of three deaths occurred (two in lavage and one in bolus group). The quality score of this RCT based on the critical appraisal skills programme checklist was 9/11.

Another RCT [16] conducted in India included term infants with MAS who had moderate to severe respiratory distress (Downes score >4) and were randomized to bovine surfactant (Survanta), diluted using normal saline to a phospholipid concentration of 20 ml/kg (n = 31), or to no lung lavage (NLL) (n = 29). The median duration of respiratory support was 34 h in SLL group and 44 h in NLL group. The duration of oxygen therapy post-respiratory support decreased by 78% in SLL as compared with NLL group. There was no significant difference between both groups for duration of hospital stay and oxygen therapy, death, ECMO, incidence of clinical sepsis, pneumothorax, persistent pulmonary hypertension, discharge and death. The quality score of this RCT based on critical appraisal skills programme checklist was 9/11.

The three SRs were found to be of high, low and critically low quality (Table 1). Out of the seven critical items, only item ten was satisfied by all three SRs and protocol registration before the commencement of SR and the reporting of list of excluded studies with justification were not satisfied by two SRs. The majority of the SRs satisfied the remaining noncritical domains, which were; items 1, 3, 6, 8 and 16 (n = 3), items 5, 12 and 15 (n = 2), and none of the SR satisfied the item 10 (funding for the included studies).

SRs were found to be of high [14], unclear [13] and low [8] risk of bias on the assessment of ROBIS. The sequence of domains that contributed to high risk of bias in the Natarajan et al. SRs were: domain 3 and domain 4. For unclear risk of bias in the Choi et al. SRs, the contributing domain was number 2. The major SQs that were contributing to the rating of high risk of bias were in domain 3 (3.4, 3.5) and domain 4 (4.5 & 4.6). The major SQs that were contributing to rating the unclear risk of bias were in domain 2 (2.3, 2.5; Figure 2).

In this overview of SRs, we examined three SRs and two RCTs that were published recently and not included in any SR. All these included SLL and BS methods using natural and synthetic products for MAS treatment.

To add to the comprehensive reporting of the top sources of evidence in literature that assessed surfactant therapy (bolus and lavage) in MAS, we went beyond the published SRs and included recently published RCTs that are not included in any SRs. Each meta-analysis outcomes in SRs were summarized separately, including in a structured table, which can help readers realize or review interesting outcomes easily.

There is a clear inconsistency in the reported outcomes of SRs, which disable a straightforward interpretation about the usefulness of surfactants for MAS. A meta-analysis of RCTs of SLL in the low-quality SR, by Choi et al. [13], concluded surfactant effectiveness against death and ECMO use, but not against the pneumothorax as a result of MAS. In the same SR, a meta-analysis of NRSs concluded effectiveness against all the aforementioned, including the pneumothorax and air leaks outcome. This, however, was all contradicted by a high-quality Cochrane meta-analysis [14] on the use of BS. This Cochrane review suggested decreased progressive respiratory failure requiring ECMO. It, however, concluded no effect against death, pneumothorax, air leaks, and MV duration. With the critically low-quality meta-analysis that looked at both SLL and BS together [8], an effectiveness was not concluded against death, but only against the need for ECMO and the hospital stay.

To add to the inconclusively of surfactant effectiveness, based on the results of Choi et al. and El Shahed et al. meta-analyses, one might conclude that SLL are more effective than the BS. This, however, is negated by the two high-quality and most recent RCTs that conclude no difference between SLL and BS in effectiveness [15,16].

An important safety concern is that the instillation of large volume of fluid into a newborn’s lung might be a burden, especially in cases of severe MAS, which leads to mortality [37]. The El Shahed et al. Cochrane review assessed the efficacy of BS and indicated no improvement on morbidity or mortality [8]. Nevertheless, while two RCTs [38,39] assessed a surfactant made of porcine administered in small lavage quantity to eight infants and showed improvements in oxygenation, those effects were improved with large volume of diluted porcine surfactant in a recently published RCT [15].

In vivo studies indicated that there was more release of proinflammatory cytokines that occurred in male fetuses when stimulated with lipopolysaccharide stimulation compared with female fetuses [40,41]. A recent study (n = 95), conducted in Japan, found that male neonates were at a higher risk of developing MAS than female neonates. However, further studies are needed to confirm the role of sex on MAS development [42].

The quality assessment of SRs indicated that none of the SR reported funding source and the two other non-Cochrane reviews did not have prior protocol and did not provide excluded studies list. Hence, based on AMSTAR-2 assessment, the quality of included SRs were rated as high [8], low [13] and critically low quality [14]. In addition to that, we have found a similar pattern in risk of bias assessment results based on the ROBIS tool. Among three reviews, only the Cochrane review was found to be of low risk of bias.

The overview has some limitations. The restriction to English language might have excluded some studies published in other languages. The authors in the current study however do not have the resources to translate the non-English research literature that may generate from a non-restricted search. Furthermore, searching additional index terms to those in the study or additional combinations of them is always possible and may generate additional studies. In addition, the fact that a primary article could have been included in more than an SR may contribute to double counting of data within reported meta-analyses. Not exploration of such overlaps tool place in this study.

Overall, our overview of SRs and recent RCTs considered all the available literature to summarize and critically appraise the evidence using the available tools. However, while limited evidence of effectiveness against death, ECMO, pneumothorax, air leaks and MV duration was reported in literature, this was provided via critically low quality, with high risk of bias, reviews. Higher quality SRs, with low risk of bias, concluded a lack of surfactant effectiveness. A similar lack of clarity trend is observed when one tries to draw overall conclusion about how different surfactant modalities compare. Larger trials are needed to determine the effectiveness of BS administration and/or SLL to treat MAS effectively and safely.

To view the supplementary data that accompany this paper please visit the journal website at: Supplementary Material

D Al-Badriyeh conceived and designed the study. All authors contributed equally in data collection and analysis, result interpretation and revising the manuscript. M A Abdelaal wrote the first manuscript draft. All authors read and approved the final manuscript.

This research was funded by a Qatar University grant number: QUST-1-CPH-2019-15. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.