Assefa Desalew1, Agumasie Semahgn1, Gezahegn Tesfaye2. 1. Department of Nursing, School of Nursing and Midwifery, College of Health and Medical Sciences, Haramaya University, P.O. Box 235, Harar, Ethiopia. 2. Department of Reproductive Health, School of Public Health, College of Health and Medical Sciences, Haramaya University, P.O. Box 235, Harar, Ethiopia.
Neonatal asphyxia is defined as the failure of initiating and maintaining of breathing at birth.[1,2] Worldwide, more than 1 million neonatal mortality occurred due to birth asphyxia every year.[3,4] A diagnosis of asphyxia is established, when the newborn has a <7 APGAR score at 1st or 5th min after birth.[1,5,6] The acidity of umbilical cord blood can also indicate infants’ oxygen shortage.[5,6] Birth asphyxia results in impairment of tissue perfusion and then yielding to hypoxemia and hypercarbia.[7-9] It is due to the newborn fail to breath normally, which leads to decreased oxygen perfusion to various organs.[5,10,11]Globally, an estimated 4 million newborns die in the neonatal period; 3 million of them died within 7 days.[12] This accounts for 46% of under-five mortality[13-15] and estimated to increase to 52% in 2030.[16,17] More than 99% of neonatal mortality occurs in developing countries.[18] Neonatal asphyxia is responsible for 42 million disability-adjusted life years.[1,3,10,19-26] The proportion of birth asphyxia is 2 per 1000 births in developed countries but more than 10 times higher in developing countries, where the setting with limited access to quality maternal and neonatal care.[9] Birth asphyxia contributes to a significant burden of neonatal mortality and morbidities. It may result in multi-organ dysfunctions or death. Moreover, survivors of neonatal asphyxia and its main complication (hypoxic-ischemic encephalopathy) may suffer from epileptic disorder, cerebral palsy, mental retardation, blindness, hearing, learning, and behavioral disabilities.[3,9,19,23]In developing countries, newborns had a high chance of being asphyxiated and stillbirth.[25] The available evidence on neonatal mortality rates (NMR) ranged from 0.2% to 64.4% in these settings.[27] The majority of neonatal mortality happened in Asia 39% and Sub-Saharan Africa (SSA) 38%. Around 70–80% of these neonatal deaths occur due to preventable and treatable conditions with access to simple, affordable interventions.[12-28] Ethiopia is among countries accounting for more than half of newborn deaths in developing counties.[13-18] Birth asphyxia, septicemia, and complications of preterm birth, jaundice, meningitis, and tetanus are the main cause of neonatal mortality in SSA.[29,30] According to the Ethiopian demographic and health survey, the NMR was 29 out of 1000 live births,[31] and more than 50% of neonatal deaths occurred within the 1st day of life.[17] A complication of prematurity, neonatal asphyxia, and neonatal sepsis were the three common causes of newborn death in Ethiopia.[32-36]Multiple published studies showed that poor antenatal care (ANC),[37,38] cesarean section, meconium-stained amniotic fluid (MSAF), preterm birth, preeclampsia or eclampsia, and instrumental delivery were major contributing factor for birth asphyxia.[39,40] Occurrences of more serious complications and limited access to quality intrapartum care increased the burden and magnitude of asphyxia in resources limited countries.[23]For effective health care to be achieved, attention has to be directed to reduce neonatal deaths secondary to birth asphyxia.[38] Supporting with basic newborn resuscitation alone reduce about 30% of intrapartum-related deaths.[1,2,41-43] Moreover, 1-day Helping Babies Breath training can improve the capacity of birth attendants but its implementation of the real action is uncertain.[44] Furthermore, interventions directed to birth asphyxia are less dependent on technology and commodities than trained people.[45] Therefore, improving skills of birth attendance, emergency obstetric care and retraining of this personnel with access to resuscitation equipment is crucial for reductions of mortality due to birth asphyxia.[25,28,46]Although promising advancement in maternal and childcare occur in the past 10 years, prenatal asphyxia still remains as the main cause of neonatal morbidities and mortality.[38,47-49] With an accelerated increment in facilities-based delivery, attention has to shift toward the quality of service as poor quality would further increase the burden of birth asphyxia.[2] In SSA, including Ethiopia, the main challenges to reducing birth asphyxia are the lack of skilled workers and resuscitation equipment.[46] Evidence pinpointed that birth attendance has insufficient knowledge of birth asphyxia and poor skills in newborn resuscitation.[50] To the best of our knowledge, there is a lack of compressive and solicited evidence for determinants of birth asphyxia in this country. Therefore, the main aim of this meta-analysis was to determine the pooled magnitude of birth asphyxia and its determinants in Ethiopia.The review questions were:What was the pooled estimate of birth asphyxia in Ethiopia?What were the main determining factors of birth asphyxia in Ethiopia?
Methods
This systematic review and meta-analysis process, identification, screening, and eligibility assessment of full articles were carried out according to Preferred Reporting Items for Systematic Review and Meta-analysis (PRISMA) statement[51] [Additional file 1]. The review protocol was registered in an international prospective register of systematic Review (PROSPERO ID: CRD42018105467). This can be accessed from http://www.crd.york.ac.uk/PROSPERO/display record.php?ID: CRD 4201 810 5467.
Searching strategies
The databases, including PubMed, Medline, CINAHL, EMBASE, and other relevant sources including Google search engine, Google Scholar, and World Health Organization websites were used to search relevant articles. The following keywords were served as search strings (a) population (newborn, neonate, and fetus), (b) outcome (birth asphyxia, perinatal asphyxia, intrapartum asphyxia, asphyxia neonatorum, perinatal suffocation, suffocation, APGAR score, determinants, associated factors, correlates, predictors, and risk factors), (c) study design (observational studies), and (d) setting (Ethiopia). Finally, all studies which are in agreement with the review title were retrieved and screened for inclusion criteria [Additional file 2].
Eligibility criteria
All studies with cross-sectional, cohort, and case–control study design and survey results were eligible in this meta-analysis. However, case series and reports were excluded from this meta-analysis. Articles with the main aim of determining the proportion of its determinants of birth asphyxia within Ethiopia were considered. Both facility and community-based articles were also included. Both published and unpublished studies at any time point until the last date of search (June 2, 2019), written in the English language and fulfill all other criteria were eligible in the selection process.
Studies screening and selection process
With the possible and appropriate capacity, online documents from the main dataset or directory were transferred into ENDNOTE reference manager software version X8. Then, the articles were collected into a single folder to find duplicates files and removed with the above software. After that, two authors (GT and AD) were separately screened the articles based on preset inclusion criteria. Through title screening, the studies that mentioned birth asphyxia were nominated for abstract screening. Consequently, studies that fulfill eligibility criteria with titles and abstracts were retrieved for full-text screening. Then, full-text screenings were carried out with two independent authors (AD and GT). In any disagreement between the first two authors, the third (AS) was asked to reach into the final decision. The studies screening process based on PRISMA guidelines follows the diagram, as shown in Figure 1.[51]
Figure 1
Preferred Reporting Items for Systematic Review and Meta-analysis 2009 flow diagram illustrating the screening process for the meta-analysis in identifying the determinant of birth asphyxia in Ethiopia
Preferred Reporting Items for Systematic Review and Meta-analysis 2009 flow diagram illustrating the screening process for the meta-analysis in identifying the determinant of birth asphyxia in Ethiopia
Critical appraisal of studies
Studies were critically assessed for the validity of results. To ensure the methodological and evidence quality of the studies, we used the Joanna Briggs Institute (JBI) appraisal tool for observational studies.[52] This JBI critical appraisal checklist had nine questions to assess prevalence data (Q1-Q9), which mainly addresses the methodological area of every article. The results of two authors (GT and AD) with consulting the third author (AS) (in case of discrepancy in the first two authors) were reached into the final judgment. Then, articles with positive answers (yes) for more than 50% of the checklists (i.e., yes for five or more) were included in this meta-analysis. Particular attention was focused on the appropriateness of the design, sampling techniques, data collection, objective, statistical analysis, and any sources of bias [Additional file 3].
Data extraction
Based on the inclusion criteria, two authors (AD and AS) set an extraction template in the Microsoft Excel sheet (2013). Then, the reviewers independently extracted information from all eligible publications. The study description table was formulated to summarize the study design, sample, population, aim, key finding (prevalence of birth asphyxia), and secondary outcome (determinants) [Table 1]. The extracted numerical data were documented and stored in a Microsoft Excel separate sheet.
Table 1
Describe the characteristics of included studies for outcome variables in the systematic review and meta-analysis
Describe the characteristics of included studies for outcome variables in the systematic review and meta-analysis
Data synthesis and statistical analysis
A summary table was prepared to describe the characteristics of the included studies. The quantitative data were extracted using Microsoft Excel. Then, data were moved into comprehensive meta-analysis (CMA)[53] and RevMan v5.3 statistical software for the meta-analysis. The pooled prevalence of birth asphyxia was calculated with CMA statistical software, while the factors associated with birth asphyxia were analyzed using RevMan software. The data analysis was performed by AD and GT. The Cochran Q test was applied to find out the occurrence of significant statistical heterogeneity and the level was measured using I2 statistics. When the included studies have high heterogeneity, we used a random-effects model. Sub-group analysis was also conducted considering the APGAR score at the 1st min and 5 min. Any bias related to publication was checked with a funnel plot.
Results
Search results
From 1080 articles retrieved through main databases and direct searches, 438 studies were removed due to duplication through ENDNOTE citation manager. Then, 478 studies were excluded after the title and abstract screening. Full publications of 104 articles were checked in detail for the presence of one of the outcomes variables, and 74 studies were removed. The remaining 30 studies undergone a critical appraisal and 18 studies were excluded in the final meta-analysis because of relative poor method related quality, inconsistency, and unavailability of the data. Finally, 12 publications were included in the pooled estimation of the magnitude of birth asphyxia and eight studies were considered for the analysis of factors associated with birth asphyxia [Figure 1].
Characteristics of studies
Twelve articles with 17147 newborns and 2328 cases of birth asphyxia were incorporated in meta-analysis. Among the studies included in the final analysis, seven were cross-sectional, four were case–control, and the others were prospective cohort. All included studies had sample sizes ranged from 154 to 9738. All articles were written in English. The characteristics of included articles in the meta-analysis were described in the following [Table 1].[54-64]
The prevalence of birth asphyxia
The pooled proportion of birth asphyxia in Ethiopia was found to be 22.8% (95% confidence interval [CI]: 13–36.8%), as shown in Figure 2. In sub-group analysis, the prevalence of birth asphyxia in the 1st min was 30.4% (95% CI: 24.6–37%), and at the 5th min was 14.6% (95% CI: 4.3–39.5%).
Figure 2
Overall pooled estimation of birth asphyxia using the random-effect model in Ethiopia
Overall pooled estimation of birth asphyxia using the random-effect model in Ethiopia
Determinants of birth asphyxia
The present meta-analysis found various determinants for birth asphyxia in Ethiopia. Maternal illiteracy, low birth weight (LBW), antepartum hemorrhage (APH), preterm births, newborn with MSAF, cesarean section delivery, prolonged duration of labor, instrumental delivery, non-cephalic fetal presentation, and induction or augmentation of labor were found to have a statistically significant association with birth asphyxia. However, ANC use, parity, and maternal anemia were not significantly associated with the outcome variable [Figures 3-16].
Figure 13
Association between induction and augmentation of labor with birth asphyxia in Ethiopia
Figure 16
Association between antenatal care follow-up with birth asphyxia in Ethiopia, 2019
Association between low birth weights with birth asphyxia in EthiopiaAssociation between preterm births with birth asphyxia in EthiopiaAssociation between non-cephalic fetal presentations with birth asphyxia in EthiopiaAssociation between maternal educations with birth asphyxia in EthiopiaAssociation between cesarean section with birth asphyxia in EthiopiaAssociation between instrumental deliveries with birth asphyxia in EthiopiaAssociation between hypertension during pregnancy with birth asphyxia in EthiopiaAssociation between antepartum hemorrhages with birth asphyxia in EthiopiaAssociation between duration of labor with birth asphyxia in EthiopiaAssociation between meconium-stained amniotic fluids with birth asphyxia in EthiopiaAssociation between induction and augmentation of labor with birth asphyxia in EthiopiaAssociation between parity with birth asphyxia in EthiopiaAssociation between maternal anemia with birth asphyxia in EthiopiaAssociation between antenatal care follow-up with birth asphyxia in Ethiopia, 2019
LBW
In this meta-analysis, LBW (<2.5 kg) found a statistically significant association with birth asphyxia with adjusted odds ratio (AOR; 5.17, 95% CI: 2.62–10.22). These indicated that LBW newborns were 5 times more likely to be affected with birth asphyxia compared with their counterparts. Despite the presence of heterogeneity between the studies, LBW was associated with birth asphyxia, as illustrated in Figure 3.
Figure 3
Association between low birth weights with birth asphyxia in Ethiopia
Preterm
According to this meta-analysis, preterm births were found as significant determinants of birth asphyxia. Babies born before 37 weeks of gestation (preterm) have an increased odds of experiencing birth asphyxia with three folds as compared to infants born after term gestations (AOR; 3.98, 95% CI: 3.00–5.29) [Figure 4].
Figure 4
Association between preterm births with birth asphyxia in Ethiopia
Fetal presentation
According to this meta-analysis, non-cephalic fetal presentation was significantly associated with birth asphyxia. Fetuses who present in non-cephalic ways had more risk of being affected with birth asphyxia (AOR; 4.33, 95% CI: 1.97–9.51) [Figure 5].
Figure 5
Association between non-cephalic fetal presentations with birth asphyxia in Ethiopia
Maternal education
Maternal education level has a statistically significant association with birth asphyxia. Illiterate women were more likely to give asphyxiated newborn when compared with mothers who have attended at least primary and above education level (AOR; 1.96, 95% CI: 1.44–2.67) [Figure 6] and found a statistically significant association with birth asphyxia.
Figure 6
Association between maternal educations with birth asphyxia in Ethiopia
Mode of delivery
Newborn delivered through cesarean section had about 4 times the chance of experiencing severe asphyxia than newborns delivered with spontaneous vaginal birth (AOR; 3.66 [95% CI: 1.35–9.91]) [Figure 7]. Similarly, newborns delivered by assisting instrumental delivery were 2.7 times more likely to be asphyxiated than newborns delivered through spontaneous vaginal mode (AOR; 2.74, 95% CI: 1.48–5.08) [Figure 8]. Giving birth through a cesarean section or instrumental delivery was more likely to expose their newborn for birth asphyxia as compared with newborn delivered through spontaneous vaginal delivery.
Figure 7
Association between cesarean section with birth asphyxia in Ethiopia
Figure 8
Association between instrumental deliveries with birth asphyxia in Ethiopia
Hypertension during pregnancy
Having hypertensive disorders of pregnancy showed a significant association with the outcome variable. Mothers who had preeclampsia or eclampsia have 4 times the chance of giving asphyxiated newborn than mothers without these disorders (AOR; 4.35, 95% CI: 2.98–6.36) [Figure 9].
Figure 9
Association between hypertension during pregnancy with birth asphyxia in Ethiopia
APH
The presence of APH was found to have a statistically significant association with birth asphyxia. Neonates from mothers with APH were at high risk of being asphyxiated than newborns from mothers without APH (AOR; 3.43, 95% CI: 1.74–6.77) [Figure 10].
Figure 10
Association between antepartum hemorrhages with birth asphyxia in Ethiopia
Prolonged duration of labor
The prolonged duration of labor was found as one of the determinants of birth asphyxia. Baby born after a prolonged duration of labor had about 3 times more likely to experience asphyxia than those born in the normal duration of labor (OR; 3.09, 95% CI: 1.60–5.99) [Figure 11].
Figure 11
Association between duration of labor with birth asphyxia in Ethiopia
MSAF
MSAF was found to have a statistically significant association with birth asphyxia. Newborn delivered with MSAF were 4 times more likely to be asphyxiated as compared with those delivered with clear fluids (AOR; 4.59, 95% CI: 1.40–15.08) [Figure 12].
Figure 12
Association between meconium-stained amniotic fluids with birth asphyxia in Ethiopia
Induction of labor
Newborns delivered after induction or augmentation labor to facilitate the delivery process had almost 4 times the chance of suffering from birth asphyxia as compared with their counterparts (AOR; 3.69, 95% CI: 2.26–6.01) [Figure 13].
Parity
Babies born from primipara mothers also had a higher chance of getting birth asphyxia than newborns from multipara mothers (AOR; 1.29, 95% CI: 1.03–1.62) [Figure 14].
Figure 14
Association between parity with birth asphyxia in Ethiopia
Anemia
Newborn babies from anemic mothers had a chance of giving asphyxiated babies than non-anemic mothers, but it was not statistically significant association (AOR; 2.96, 95% CI: 0.93, 9.40) [Figure 15].
Figure 15
Association between maternal anemia with birth asphyxia in Ethiopia
Use of ANC follow-up
Women who had ANC follow-up were 35% less likely to have asphyxiated babies than women who had no visit; however, there was no statistical association (AOR; 0.65, 95% CI: 0.38–1.11) [Figure 16].
Discussion
The present finding indicates that maternal education, APH, caesarian section, instrumental delivery, prolonged duration of labor and induction or augmentation, mode of delivery, being primiparous, LBW, preterm births, MSAF, and non-cephalic presentations were associated with birth asphyxia in Ethiopia. We found out that nearly one-fourths of the newborn were suffering from birth asphyxia in Ethiopia. However, in this analysis, maternal anemia and current use of ANC were not associated with birth asphyxia. The present finding provides important information because; to the best of our knowledge, this paper is the first meta-analysis with regard to determinants of birth asphyxia.The present meta-analysis provides a summary of available compressive evidence of birth asphyxia and its determinant in the country. Maternal education and prevention of APH are believed to decrease birth asphyxia.[65,66] In current circumstances specifically, in developing countries, where maternal illiteracy is high, it is clear that mothers may not use the prevention strategies as they had inadequate awareness about the burden of birth asphyxia and the determining factors. Therefore, there is a need to refocus the attention to improve birth outcomes with quality intrapartum service including proper resuscitation and early detection of the preventable factors of birth asphyxia in resource-limited countries, particularly in Ethiopia. With regard to the association of APH with birth asphyxia, this may be explained by the fact that there is a reduced blood movement from the placenta to the fetus, resulting in hypoxemia and lead to asphyxia or stillbirth if maternal transfusion is delayed at the time of delivery.According to present meta-analysis, cesarean sections, instrumental deliveries, induction or augmentations, and prolonged durations of labor were found statistically significant with neonatal asphyxia. This finding was consistent with the different studies conducted in other settings.[47,67-74] The burdens related to birth asphyxia may be related to instrumental delivery because of prolonged labor and delayed interventions so that close monitoring of labor processes, early detection of the main complications, and timely appropriates decision and avoiding unnecessary indications for cesarean section are essential to reducing the burden of birth asphyxia.Furthermore, preeclampsia or eclampsia has found a statistically significant association with birth asphyxia. The finding is in agreement with evidence in Africa such as Ghana and Egypt.[39,68,70] This may be due to the reduction of blood follow, nutrients and oxygen movement to the fetus, which may increase the risk of in intrauterine development restriction, which may result in perinatal asphyxia. In addition, MSAF was found as a determinant of birth asphyxia. This was in agreement with studies from different countries.[69,73-78] The possible reason may be related to inhalation of MSAF, which causes irritation and inflammation of the lung tissues or may obstruct the airway further inducing hypoxia and birth asphyxia.Moreover, LBW and pre-term births were found to be significant determinants of birth asphyxia, which was similar to different findings in many settings.[10,39,78-81] In fact, much of LBW newborns are more likely to be pre-term that they are not able to produce adequate surfactant and prone to multiple morbidities, including organ system immaturity, including the inability of initiation of breathing, face challenges in cardiopulmonary transition, and finally, develop birth asphyxia.[75,77] Moreover, a non-cephalic fetal presentation was found independent predictor of birth asphyxia and it is in agreement with other articles in different countries.[6,10,79] In fact, non-cephalic presentation has long been well known to face greater hazards during the process of birth including birth asphyxia, birth trauma, and death. This may be because fetuses presenting with non-cephalic way are more likely to have other associated problems such as umbilical cord prolapsed and head entrapment that predispose them to birth asphyxia.
Limitations and strengths
The present review had certain limitations. The first one was not including qualitative studies in the review, which might identify other determinants of birth asphyxia or might support the quantitative findings. Second, conducting meta-analysis despite the heterogeneity between the included studies might affect the findings. Third, the search was only limited to articles published in the English language. Finally, despite the incorporation of studies from different parts of the country, the representativeness of the population is not as strong as the studies were observational in nature and the majority of them were conducted among newborns admitted to the neonatal intensive care unit. This meta-analysis also has strengths such as the selection and inclusion of both published and unpublished literature which has the potential to minimize publication bias. Moreover, our search strategy was extensive using multiple reputable databases and search engines.
Conclusion
The pooled magnitude of birth asphyxia was very high. Maternal education, APH, caesarian section, instrumental delivery, prolonged duration of labor, induction or augmentation, MSAF, and non-cephalic presentation were factors associated with birth asphyxia. LBW and preterm births were found as fetal related determinants of birth asphyxia. Hence, to reduce birth asphyxia and associated neonatal mortality, attention should be directed to improve the quality of intrapartum service and timely communication between the delivery team. In addition, intervention strategies that aim to reduce birth asphyxia should target the identified factors.
Authors’ Contributions
AD and AS initiated and formulated this meta-analysis. AD conducted activities from initiation to finalization of the manuscript. AD, AS, and GT build-up the search strategies, meta-analysis, and interpretation of the findings. All authors thoroughly revised the manuscript.
Authors: Maria Altman; Anna Sandström; Gunnar Petersson; Thomas Frisell; Sven Cnattingius; Olof Stephansson Journal: Eur J Epidemiol Date: 2015-05-26 Impact factor: 8.082
Authors: Dan L Ellsbury; Reese H Clark; Robert Ursprung; Darren L Handler; Elizabeth D Dodd; Alan R Spitzer Journal: Pediatrics Date: 2016-03-02 Impact factor: 7.124
Authors: Stephen N Wall; Anne C C Lee; Susan Niermeyer; Mike English; William J Keenan; Wally Carlo; Zulfiqar A Bhutta; Abhay Bang; Indira Narayanan; Iwan Ariawan; Joy E Lawn Journal: Int J Gynaecol Obstet Date: 2009-10 Impact factor: 3.561
Authors: Mariya Mukhtar-Yola; Lamidi Isah Audu; Oluyinka Olaniyan; Henry T Akinbi; Adekunle Dawodu; Edward F Donovan Journal: BMJ Open Qual Date: 2018-09-04
Authors: Alexander Manu; Shams Arifeen; John Williams; Edward Mwasanya; Nabila Zaka; Beth Anne Plowman; Debra Jackson; Priscilla Wobil; Kim Dickson Journal: BMC Health Serv Res Date: 2018-07-09 Impact factor: 2.655
Authors: Selina Chow; Ronald Chow; Mila Popovic; Michael Lam; Marko Popovic; Joav Merrick; Ruth Naomi Stashefsky Margalit; Henry Lam; Milica Milakovic; Edward Chow; Jelena Popovic Journal: Front Public Health Date: 2015-10-07
Figure 1
Figure 2
Figure 13
Figure 16
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9
Figure 10
Figure 11
Figure 12
Figure 14
Figure 15