Stanley Mwita ·Mary Jande ·Deogratias Katabalo ·Benjamin Kamala ·Deborah Dewey

Abstract Background The most common cause of death among preterm infants in low-and middle-income countries is respiratory distress syndrome.The purpose of this review was to assess whether antenatal corticosteroids given to women at risk of preterm birth at ≤ 34 weeks of gestation reduce rates of neonatal mortality and respiratory distress syndrome in low-and middle-income countries.Methods Two reviewers independently searched four databases including MEDLINE (through PubMed),CINAHL,Embase,and Cochrane Libraries.We did not apply any language or date restrictions.All publications up to April 2020 were included in this search.Results The search yielded 71 articles,10 of which were included in this review (3 randomized controlled trials,7 observational studies,36,773 neonates).The majority of studies reported associations between exposure to antenatal corticosteroids and lower rates of neonatal mortality and respiratory distress syndrome.However,a few studies reported that antenatal corticosteroids were not associated with improved preterm birth outcomes.Conclusions Most of the studies in low-and middle-income countries showed that use of antenatal corticosteroids in hospitals with high levels of neonatal care was associated with lower rates of neonatal mortality and respiratory distress syndrome.However,the findings are inconclusive because some studies in low-resource settings reported that antenatal corticosteroids had no benefit in reducing rates of neonatal mortality or respiratory distress syndrome.Further research on the impact of antenatal corticosteroids in resource-limited settings in low-income countries is a priority.

Keywords Antenatal corticosteroids·Low-and middle-income countries·Neonatal mortality·Respiratory distress syndrome

Introduction

Preterm birth is defined as a live birth that occurs before completing 37 weeks of gestation.According to the World Health Organization,every year about 15 million babies are born prematurely around the world.The prevalence of preterm birth ranges from 5 to 18% with more than 60% of preterm births occurring in low-and middle-income countries (LMICs) in Africa and South Asia [1].Preterm birth complications are responsible for almost 28% of neonatal deaths [2-4].In 2015,one million neonatal deaths worldwide were due to complications associated with preterm birth [5].

The most common cause of deaths among preterm infants in LMICs is respiratory distress syndrome (RDS)[6],a lung disease caused mainly by immaturity of the lungs and surfactant [7].Symptoms of RDS include cyanosis,tachypnea,chest retractions,grunting,and nasal flaring [8].RDS-specific mortality in LMICs is high [9].A study conducted in Brazil found that preterm infants with RDS were three times more likely to die compared to those without RDS [10].

Administration of antenatal corticosteroids (ACS) (betamethasone or dexamethasone) to women at risk of preterm birth is among the most effective hospital-based intervention for reducing rates of neonatal deaths and RDS because ACS accelerates lung maturation [11,12].ACS have also been associated with reduced risk of necrotizing enterocolitis(NEC) and intraventricular hemorrhage (IVH) [11].A single course of ACS is currently recommended to be administered between 24 and 34 weeks of gestation for women who are at risk of preterm birth within seven days.Specifically,these include cases of threatened preterm labor,preeclampsia,preterm premature rupture of membranes and antepartum hemorrhage [13-15].

An updated meta-analysis of randomized controlled trials (RCTs) that included evidence primarily from hospital settings in high-income countries found that treatment with ACS (compared with placebo or no treatment) was associated with a reduction in neonatal mortality and RDS [16].Previous reviews and meta-analyses have reported on:(1)the use of ACS given at ≥ 34 weeks of gestation for increasing lung maturity in term and near-term fetuses [17];(2)the use of ACS in preventing neonatal respiratory morbidity after elective caesarean section at term [18];(3) the effect of repeat doses of ACS [19]and (4) the effects of different ACS types and regimens [20].However,most of these studies have focused on research conducted in high-income countries.Limited research has investigated the use of ACS for improving neonatal health outcomes in preterm babies (≤34 weeks of gestation) in LMICs.Therefore,we conducted a scoping review to determine the available studies and to identify and analyze knowledge gaps related to research on the effects of ACS given to women at risk of preterm birth at ≤ 34 weeks of gestation on neonatal mortality and RDS in LMICs.

Methods

We utilized a scoping review framework to summarize the extent and nature of the research and to map the evidence for impact of ACS given to women at risk of preterm birth in reducing neonatal mortality and RDS in LMICs.The current study was guided by the adapted Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA)extension for scoping reviews (Supplementary Fig.1) [21].

Search strategies

The search strategies were drafted by an experienced librarian (Lorenzetti DL) and were refined further through team discussion.The following bibliographic databases were searched from inception to April 2020:MEDLINE (through PubMed),EMBASE,CINAHL and Cochrane Library.The final search strategy for MEDLINE can be found in additional Supplementary Table 1.The final search results were exported into EndNote,and duplicates were removed.The electronic database search was supplemented by searching relevant reviews and studies for additional published articles.

Study eligibility criteria

Eligible studies included RCTs and observational studies that compared ACS administration of either betamethasone or dexamethasone with placebo or no treatment in women at risk of preterm birth.In addition,studies that compared an intervention/experimental group with a control group were included in this review.We excluded qualitative studies,reviews,commentaries,comment letters,and study protocols.Participants under study were neonates born between 24 and 34 weeks of gestation or infants that weighed less than 2500 g if weight was used as a proxy for preterm birth.

The outcomes of interest were neonatal mortality (death during the first 28 days of life) and RDS prior to discharge from the hospital.We included studies that defined RDS based on chest radiographic findings,i.e.,the presence of severe or generalized microreticulogranular infiltrates and air bronchogram.Studies that focused only on knowledge and utilization of ACS were excluded from this review.We did not apply any language or date restrictions.Gray literature and unpublished works were excluded.

Screening process

Using the above-predefined eligibility criteria,studies were identified for potential inclusion in this review.To determine whether studies were eligible for inclusion,two reviewers,MS and DD,independently read the titles and abstracts obtained from the database searches to determine if they met the inclusion criteria.Differences in opinion were resolved by discussion among MS and DD.

Data extraction and analysis

Using a structured data extraction form,MS extracted and recorded data on study characteristics and DD reviewed the extraction process.Extracted data included the following:authors,publication year,study location,time frame,preterm definition,study design,setting,participants,and main findings.The criteria for a country to be classified as LMIC were based on the Gross National Income (GNI) per capita as per data from the World Bank.Low-income economies are defined as those with a GNI per capita of less than $995;lower middle-income economies are those with a GNI per capita between $996 and $3895;and upper middle-income economies are those with a GNI per capita between $3896 and $12,055 [22].

Results

Search results

Our initial literature search identified 71 publications,61 from electronic database and 10 from the references within review articles.After screening for duplications,37 publications were retained.After reviewing the titles and abstracts,20 publications were excluded based on our predefined eligibility criteria.Seventeen articles were assessed for eligibility based on their full text.Four papers were found to be ineligible due to lack of data on the effects of ACS on neonatal mortality and/or RDS.Three additional publications were excluded because they were duplicate publications on the same study cohort (Supplementary Table 2 and Fig.1).Hence,ten studies (36,773 neonates) were included in this review (Fig.1).

Characteristics of sources of evidence

The included studies were published between 1999 and 2018.Of the ten studies included,seven were observational studies (30,100 neonates):three secondary analyses of medical records data,three retrospective studies,and one study included both secondary and retrospective analyses.Three studies were RCTs (6673 neonates;3473 in intervention and 3200 in control arms).More than twothirds of the studies were conducted in South America and Africa,two studies were from Asia,and one study was conducted in the Caribbean.Of the ten studies,three defined preterm births based on weight only,three used gestational age only,and four used both weight and gestational age.There were variations in the birth weight ranges that were considered as proxies for preterm birth.Less than 2500 g was used in two studies,less than 1500 g was used in three studies,one study used greater or equal than 750 g,and another study used 1000 g to 2000 g.Signifi-cant variation also was observed in the gestational ages of the pregnancies,ranging between 24 and 34 weeks.

Fig.1 Flow diagram of literature search

Almost all studies were conducted in hospital settings with the exception of one study,which was conducted in both community and hospital settings.Table 1 describes the characteristics of the studies included in this scoping review.

Summary of the results

Table 2 summarizes results on neonatal mortality,survival rates and RDS as reported from the studies included in this scoping review.

Neonatal mortality and neonatal survival rate

Eight studies reported on neonatal mortality [23-25,27,28,30-32].Four studies reported that the use of ACS was associated with a significant decrease in neonatal mortality [24,25,31,32],two studies reported non-significant associations between ACS use and neonatal mortality [23,27],and two studies reported that ACS were not associated with improved outcomes in neonatal mortality [28,30].In a prospective,double-blind randomized trial conducted among 218 pregnant women with severe pre-eclampsia in Brazil,neonatal mortality risk was found to be 50% lower in the ACS group(14%) compared to the placebo control group (28%),yielding a risk ratio (RR) of 0.5 [95% confidence interval (CI)0.28-0.89][24].Also,an investigation of outcomes in very low birth weight infants treated in 11 neonatal intensive care units in four South American countries found that lack of ACS was associated with increased odds of neonatal mortality [odds ratio (OR) 0.38 (95% CI 0.22-0.66),P＜ 0.01].An observational study conducted in nine hospitals of four South East Asian countries during 2005 revealed that infants exposed to ACS were less likely to die compared to those not exposed to ACS (RR 0.92,P=0.05) [27].Retrospective analysis of a very low birth weight cohort recruited from 26 tertiary care and teaching sites in South America reported that lower rates of neonatal mortality were associated with ACS use [OR 0.62 (95% CI 0.52-0.74),P＜ 0.001][31].

In contrast,a multi-center,double blind,placebo-controlled,randomized trial conducted in South Africa that involved preterm infants born between 28 and 34 weeks gestational age or with birth weights between 1000 and 2000 g found that use of ACS was associated with a non-significant reduction in 28-day neonatal mortality [OR 0.37 (95% CI 0.09-1.34),P=0.16][23].In addition,a large randomized controlled trial conducted in six countries by Althabe et al.[28]reported that the 28-day neonatal mortality of low birthweight infants did not differ between the intervention and control groups [RR 0.96 (95% CI 0.87-1.06),P=0.65].An observational study conducted in Tanzania by Massawe et al.[32]revealed that ACS provided as part of maternal and infant care bundle that included neonatal antibiotics and maternal antibiotics were associated with lower rates of neonatal mortality [RR 0.74 (95% CI 0.60-0.91),P=0.005].

A retrospective analysis of medical records data in a small observational study that included 40 infants in a rural hospital in Tanzania born to women with eclampsia and severe pre-eclampsia reported higher rates of neonatal mortality among infants exposed to ACS;62% of the infants exposed to ACS died compared to 15% of those who were not exposed (P＜ 0.01) [30].

Two studies reported neonatal survival rate and found that use of ACS was associated with higher survival rates[26,29].A secondary analysis of medical records data of preterm neonates born in hospitals located in China assessed infant survival and major morbidities prior to discharge.The results revealed improved neonatal survival in infants of mothers who received ACS [OR 1.615 (95% CI 1.233-1.901),P=0.033][29].An observational study in very low birth weight infants conducted at a university hospital in Jamaica found improved neonatal survival in infants whose mothers received ACS [OR=7.9 (95% CI 1.4-44.6),P=0.02][26].

Respiratory distress syndrome

Of the ten studies identified,only four reported the impact of ACS in reducing RDS [23,24,27,31].Two studies reported that use of ACS was associated with lower rates of RDS,whereas the other two studies reported no difference in risk of RDS in ACS-exposed and non-exposed infants in LMICs.An RCT conducted by Amorim et al.in Brazil found that the group who received ACS was significantly less likely to suffer from RDS compared to the placebo group [RR=0.53 (95% CI 0.35-0.82)][24].Grandi et al.[31]conducted a study which revealed that use of ACS was associated with lower odds of RDS [OR 0.62 (95% CI 0.53-0.74),P＜ 0.001].In contrast,an RCT conducted by Pattinson et al.[23]in South Africa found no differences between ACS-exposed and placebo groups with regard to RDS [OR 1.24 (95% CI 0.64-2.38),P=0.59].An observational study by Pattanittum et al.[27]conducted in four South East Asian countries also revealed that ACS-exposed and not-exposed infants did not differ in risk of RDS [RR 1.06,(95% CI 0.83-1.34),P=0.64].

Discussion

This scoping review identified ten eligible studies from LMICs that had investigated the associations between administration of ACS (betamethasone or dexamethasone)to women at risk of preterm birth and neonatal mortality and RDS.The studies included both RCTs and observational studies.The majority of the studies were conducted in hospital settings.Many of the studies reported that use of ACS was associated with lower rates of neonatal death and with improving survival of neonates [24-26,29,31,32].These reports are consistent with the results of a systematic review conducted by Mwansa-Kambafwile et al.[33]that included four studies from middle-income countries.Specifically,they reported that use of ACS was associated with signifi-cantly lower rates of neonatal deaths.In the present review,the poorest outcomes were reported by Mooij et al.[30]in a small study conducted in a rural hospital in Tanzania with inconsistent maternal and neonatal care.They found signifi-cantly more death among infants exposed to ACS.Massawe et al.[32]found that ACS use was associated with lower rates of neonatal mortality when included as part of a maternal and infant care bundle that also included implementation of the helping babies breathe protocol and avoidance of hypothermia.Despite the benefits of ACS,unnecessary treatment (e.g.,in infants with high birthweights) could be harmful.Althabe et al.[28]reported that ACS use had no effect on neonatal deaths among low birthweight infants.Nevertheless,among all births,there were increased risks of neonatal mortality [RR=1.12 (95% CI 1.02-1.22)].The authors stated that the increased risks in mortality were apparently due to increased mortality rates in infants with high birthweight.

This review also examined the association between ACS administration to pregnant women at risk for preterm birth and RDS in their infants in studies conducted in LMICs.To date,only four studies were identified,and the results were inconsistent.Two studies found that the use of ACS was associated with lower rates of RDS in infants [24,31].These findings are similar to those which have been reported in high-income countries;specifically,the use of ACS was associated with significantly lower rates of RDS [16].However,a clinical trial of women with premature rupture of membranes [23]and a large observational study [27]both found no difference in risk of RDS in ACS-exposed and non-exposed infants in LMICs.The limited number of studies and the inconsistent findings indicate a need for welldesigned studies that investigate the effect of ACS administration in women at risk for preterm birth on RDS symptoms in infants in LMICs.

Resources like ultrasound technology can be scarce in LMICs;as a result it may be difficult to accurately determine the gestational age of the fetus.This could lead to over treatment of women who deliver at term and under treatment of those who deliver prematurely,both of which are associated with increased risk of neonatal mortality and RDS [34].Moreover,poor diagnosis and management of both maternal and neonatal infections,inadequately trained staff,and lack of equipment for neonatal care compromise the impact of ACS in resource-limited settings [28,30,34].

Neonates who were exposed to ACS in countries such as China,Argentina,Chile and Guatemala were at asignificantly lower risk of neonatal death;however,it is noteworthy that the level of care provided to those neonates and their mothers approached that seen in high-income countries.Secondary analyses from RCTs reported high quality of perinatal care in Guatemala including the use of vitamins,delivery at a hospital or clinic,and delivery by a physician [35].China is an upper-middle-income country with a relatively high GDP [36].Further,most of the studies included in this review that reported positive findings on impact of ACS in reducing risk of neonatal mortality and RDS in LMICs were conducted in university affiliated hospitals,tertiary hospitals and neonatal intensive care units.The highest levels of care and interventions,such as oxygen therapy,continuous positive airway pressure,thermal care,nutritional support,mechanical ventilation and surfactant,are used in such facilities in LMICs and have been associated with lower rates of neonatal mortality and RDS in preterm infants [37].Thus,the combination of improved care with ACS use could potentially account for the observed reductions in risk of neonatal mortality and RDS reported[36,38].

Moreover,in a recently multi-country randomized trial,it was consistently reported that administration of ACS to women who were at risk for early preterm birth in lowresource countries resulted in significantly lower risk of neonatal death but no difference in risk of RDS [39].Future research is required to clarify the specific factors associated with lower rates of neonatal mortality and RDS in health facilities in LMICs that provide high levels of care.It is also essential that research be conducted to determine if the lower rates of neonatal mortality and RDS associated with ACS use that have been reported in facilities in LMICs that provide a high level of care can be generalized to health facilities with limited resources.Moreover,future reviews should focus on examining associations between ACS and other neonatal morbidities,such NEC,IVH and neonatal hypoglycemia,in low-resource settings.Further,to the best of our knowledge,no long-term follow-up studies of infants exposed to ACS in LMICs have been conducted.Future research is needed to examine if use of ACS in LMICs has beneficial effects on neurodevelopment and general health in childhood.

A strength of our scoping review was the inclusion of recent large randomized clinical trials conducted in LMICs.Also,we included data from LMICs in Africa,Asia,South America and North America (i.e.,Caribbean) and provided a comprehensive overview of the research to date that has investigated the associations between exposure to ACS and neonatal mortality and RDS in LMICs.However,our review has some limitations including an inability to assess the effects of complete and incomplete exposure to ACS.None of the included studies reported on the effect of the time interval between administration of ACS treatment and birth on neonatal outcomes.Further,to provide a complete picture of the state of the literature,this review included a number of smaller observational studies,which are prone to bias even though they provide information on local conditions that could influence the impact of ACS neonatal outcomes in rural and regional hospitals,particularly in low-income countries.In addition,many of the studies in this review used birth weight as a proxy for preterm birth.Thus,there is a chance that term babies may have been misclassified as preterm.As scoping reviews are not intended to assess the quality of the source of evidence,the conclusions of this review are based on the available literature regardless of the quality of the studies.

In conclusion,based on the research studies included in this review,we conclude that the use of ACS in hospitals with high levels of neonatal care in LMICs is associated with lower rates of neonatal mortality and RDS.However,the findings are inconclusive in terms of the benefit of ACS,particularly in hospitals and health care centers with limited resources,because some studies reported that administration of ACS to pregnant women at risk for preterm delivery was not associated with lower rates of neonatal mortality or RDS.Additional research on the impact of ACS in resource-limited settings in low-income countries is a priority.

AcknowledgementsWe would like to acknowledge Dr.Diane L.Lorenzetti (PhD),Librarian-Director Health Sciences Library,University of Calgary for her assistance during the literature search.

Author contributionsMS conceptualized and designed the study,checked all abstracts and went through each of the full texts,prepared the first draft of the manuscript.JM conceptualized and designed the study.KD and KB prepared the first draft of the manuscript.DD conceptualized and designed the study,checked all abstracts and went through each of the full texts.All authors critically reviewed the draft versions of the manuscript and approved the final version of the manuscript.

FundingNo financial or nonfinancial benefits have been received or will be received from any party related directly or indirectly to the subject of this article.

Compliance with ethical standards

Ethical approvalNot needed.

Conflict of interestThe authors declare that they have no conflict of interest.