Current Trends in Neonatal Morbidity and Mortality: Experiences from a Tertiary Center in Lagos, Nigeria

ABSTRACT.

Neonatal mortality is a major contributor to under-five mortality, and Nigeria has the second-highest number of neonatal deaths globally. The country has introduced evidence-based interventions to improve newborn care over the years. The aim of this study was to determine the current trends in neonatal morbidity and mortality at the Lagos University Teaching Hospital, monitor progress over time, and identify areas for improvement. The admission registers and case files of all the neonatal ward admissions were reviewed from January 2018 to April 2020; the age at admission, gestational age, sex, inborn or out-born status, diagnosis, and outcome were recorded and analyzed. Of the 2,959 admissions during the study period, 68.4% were out-born and 77.9% were term gestation infants. The most common diagnoses were neonatal jaundice (NNJ; 28.4%), infection (28.0%), prematurity with associated complications (22.1%), and hypoxic ischemic encephalopathy (HIE; 18.2%). The overall mortality rate was 17.6%. Prematurity with associated complications (39.2%), HIE (24.8%), congenital anomalies (CAs; 12.7%), and NNJ (11.5%) were the most common conditions associated with mortality. Of those who died, the most common diagnoses among term infants were HIE (40.7%), CAs (21.8%), NNJ (18.9%), and infection (15.5%); respiratory distress syndrome (52.4%), infection (31.8%), and CAs (7.8%) were the most common diagnoses in preterm infants. The high risk of mortality with HIE, jaundice, infections, and CAs in this cohort reflects the national figures and trends. Efforts to improve neonatal care, especially respiratory support and education of the populace on NNJ, should be intensified to reduce neonatal mortality in the country.

INTRODUCTION

Globally in 2019, 5.2 million children younger than 5 years old died, and 2.5 million (47%) of these deaths occurred during the neonatal period (< 28 days of life).¹ As a result of improvements in the treatment of pneumonia, diarrheal diseases, malnutrition, and malaria, among other measures during the past two decades, there has been an appreciable reduction in the total number of under-five deaths from 12.5 million in 1990 to 5.2 million in 2019, amounting to a 59% reduction.¹ The contribution of neonatal mortality to under-five mortality, however, increased from 40% in 1990 to 46% in 2019, as a result of the slower decline (52%) in neonatal mortality during this period.¹ Therefore, it is important that efforts to reduce neonatal mortality are intensified to achieve the Sustainable Development Goal (SDG) of reducing neonatal mortality to 12 per 1,000 live births and reducing under-five mortality to at least as low as 25 per 1,000 live births by 2030 (goal 3, target 2).²

In sub-Saharan Africa, there is a significant burden of poor neonatal outcomes, with the region having the highest neonatal mortality rate of 27 deaths per 1,000 live births in 2019,¹ and the majority of these deaths occur in the first week of life.³ Nigeria is a major contributor to neonatal mortality globally, with the second-highest number of neonatal deaths (270,000) in the world.⁴ There have been concerted efforts to reduce neonatal mortality throughout the years, using evidence-based interventions. These include the WHO recommendations for antenatal steroids, thermal care including the use of kangaroo mother care, and treatment of infections.⁵ The Essential Newborn Care Course (ENCC) package, which was adopted in Nigeria in 2008, encompasses practices and interventions that are provided to a newborn immediately after delivery, including thermal care, prevention of infection, initiation of breastfeeding within an hour after birth, and newborn resuscitation.^6,7 The training of health-care workers in the ENCC has been harmonized in Nigeria by the Federal Ministry of Health, with assistance from partners including the American Academy of Pediatrics through the Helping 100,000 Babies Survive and Thrive Initiative,⁸ among others. Other interventions are care of the small baby, identification of small and sick babies, and determination of transport for further care.⁵ Despite these past improvements and evidence-based interventions, 53 countries including Nigeria risk missing the SDG targets for neonatal mortality because of the slow rate of reduction in these populations during the past two decades.¹ It is therefore important to review regularly the trends in neonatal mortality to identify which interventions require increased focus and attention.^9,10

The aim of this study was to determine the current trends in neonatal mortality at the Lagos University Teaching Hospital (LUTH) as well as common neonatal morbidities, to monitor progress over time, and to identify areas for improvement.

METHODS

This was a 28-month retrospective review of the hospital admission records— from January 2018 to April 2020—at Lagos University Teaching Hospital, a tertiary care center in Lagos, Nigeria. LUTH has one of the largest neonatal ward capacities in Nigeria, with 80 cot spaces for infants. There are two neonatal wards, one for infants born within the facility (inborn ward) and the other for infants born outside the facility who require further care (out-born ward). Patients admitted into the inborn ward come from the hospital’s labor ward, labor ward theater, and the postnatal wards. The out-born ward receives patients from the Children Emergency Center and outpatient clinics. The hospital has a blood bank, radiology, pharmacy, microbiology services, and chemical pathology laboratory services available 24 hours a day. The hospital has oxygen for respiratory support; however, there were no functioning mechanical ventilators at the time of the study. Continuous positive airway pressure (CPAP) was provided mainly by improvised bubble CPAP¹¹ devices, which used 100% oxygen, because only a few commercial CPAP devices were available.

We reviewed the admission register and case files of all newborn admissions in the hospital for a 28-month period—January 2018 to April 2020. Information extracted included the infants’ biodata (age at admission, gestational age, and gender), inborn or out-born status, admitting diagnosis, and outcome (death or survival). Preterm neonates were categorized based on their gestational age at birth: extremely preterm, < 28 weeks; very preterm, 28 to < 32 weeks; moderately preterm, 32 to < 34 weeks; and late preterm, 34 to < 37 weeks.^12,13 They were also classified based on their birthweight as extremely low birthweight (< 1,000 g), very low birthweight (1,000 to < 1,500 g), low birthweight (1,500 to < 2,500 g), and normal birthweight (2,500–4,000 g).¹⁴

Data extracted were entered into a Microsoft Excel version 16.20 spreadsheet (Microsoft Corporation, Redmond, WA), where it was cleaned and edited. The cleaned data were analyzed within Microsoft Excel. Analyzed data are presented as frequencies in tables and charts.

Ethical approval for the research was obtained from the Lagos University Teaching Hospital Health Research Ethics Committee (approval no. ADM/DCST/HREC/APP/4154). The extracted information was de-identified to protect and maintain patient confidentiality at the point of data abstraction.

RESULTS

Subjects.

A total of 2,959 neonatal patients (< 28 days of life on admission) were cared for at LUTH during the study period. Of the total number of patients, 2,024 (68.4%) were admitted to the out-born ward and 2,304 (77.9%) were born at term (gestational age, ≥ 37 weeks). The mean age of infants on admission was 4.4 days. The demographic characteristics of the study population are shown in Table 1.

Table 1.

Patient demographics

Patient characteristic	Cases, n	Total admissions, %
Gender
Male	1,843	62.3
Female	1,081	36.5
Unspecified	35	1.2
Gestational age, weeks
Extremely preterm, < 28	85	2.8
Very preterm, 28 to < 32	187	6.3
Moderately preterm, 32 to < 34	168	5.7
Late preterm, 34 to < 37	209	7.1
Unspecified	6	0.2
Term, ≥ 37	2,304	77.9
Birthweight, g
< 2,500	744	25.1
2,500–4,000	218	73.6
> 4,000	37	1.3
Place of admission
Inborn	935	31.6
Out-born	2,024	68.4

Morbidity.

For all infants, neonatal jaundice was the most common diagnosis (28.4%, n = 840 [738 term and 102 preterm]), followed by neonatal infections (28.0%, n = 828 [549 term and 279 preterm]), prematurity (22.1%, n = 655), and hypoxic ischemic encephalopathy (HIE) (18.2%; n = 538 [529 term and 9 preterm]), as shown in Table 2. These totals include dual diagnoses (prematurity with jaundice, prematurity with infection, and so on); thus, the total numbers exceed the 2,959 infants in the study population. Analyzing only the population of term infants, jaundice remained the most common diagnosis (n = 738, 32.0%), followed by infection (n = 549, 23.8%), then HIE (n = 529, 23.0%). Of those born preterm, neonatal infections (n = 279, 42.6%), respiratory distress syndrome (RDS) (n = 174, 26.6%), and jaundice (n = 102, 15.6%) were the most common comorbidities.

Table 2.

Neonatal admissions by primary diagnosis

Diagnosis	Cases, n	Total admissions, %
Jaundice	738	25.0
Prematurity and associated complications	655	22.1
With infection	(279)	(9.4)
With respiratory distress syndrome	(174)	(5.9)
With jaundice	(102)	(3.4)
With anomalies	(33)	(1.1)
With hypoxic ischemic encephalopathy	(9)	(0.3)
With other	(58)	(2.0)
Infection	549	18.6
Hypoxic ischemic encephalopathy	529	17.9
Congenital anomalies	217	7.3
Respiratory distress (meconium aspiration syndrome, transient tachypnea of the newborn)	106	3.6
Anemia	34	1.1
Macrosomia	26*	0.9
Hypoglycemia	15	0.5
Other	90	3.0

^*Eleven other macrosomic (birthweight, > 4,000 g) infants who were admitted for other reasons (nine with sepsis, one with jaundice, and one with hypoxic ischemic encephalopathy) were excluded.

Mortality.

Of the total 2,959 patients included in the analysis, 521 died during the neonatal period, resulting in a mortality rate of 17.6% for the population studied. Among those who died, 337 were male (64.8%), 176 (33.8%) were female and eight were of unspecified gender. Of those who died, among both term and preterm populations, prematurity with associated complications was the most common condition associated with mortality during the neonatal period (n = 204, 39.2%), followed by HIE (n = 129, 24.8%), congenital anomalies (n = 69, 12.7%), and neonatal jaundice (n = 60, 11.5%; Figure 1). Of the 317 term infants who died, HIE was the most common diagnosis associated with mortality (n = 129, 40.7%), followed by congenital anomalies (n = 69, 21.8%), neonatal jaundice (n = 60, 18.9%), and infection (n = 49, 15.5%).

Figure 1.

Neonatal mortality by diagnosis. CA = congenital abnormalities; CP = complication of prematurity; HIE = hypoxic ischemic encephalopathy.

For the 204 preterm infants who died, RDS was the most common diagnosis associated with mortality (n = 107, 52.4%), followed by infection (n = 65, 31.8%), and congenital anomalies (n = 16, 7.8%), as shown in Figure 2. Seventy-five (88.2%) of the extremely preterm, 85 (45.2%) of the early preterm, 27 (16.1%) of the moderate preterm, and 16 (7.8%) of the late preterm infants died (Table 3). One hundred two infants (50.0%) were extremely low birthweight, and 70 (34.3%) and 31 (15.2%) were very low birthweight and low birthweight, respectively. Only one late preterm weighed 2,500 g at birth. Considering the place of birth, 381 (19.0%) of the out-born infants died compared with the 140 (15.0%) of the inborn infants. A greater proportion of infants with neonatal jaundice, infection, and HIE died in the out-born group compared with the inborn group (Table 4).

Figure 2.

Mortality according to complications associated with prematurity. CA = congenital abnormalities; HIE = hypoxic ischemic encephalopathy; IVH = intraventricular hemorrhage; RDS = respiratory distress syndrome.

Table 3.

Complications of prematurity by gestational age cohort

Gestational age categories	Admissions, n	Deaths, n (% of subgroup by diagnosis)
Extremely preterm (< 28 weeks)	85	75 (88.2)
Infection	8	5 (62.5)
Respiratory distress syndrome	77	70 (90.9)
Very preterm (28–31 weeks)	187	85 (45.4)
Infection	101	44 (44.4)
Respiratory distress syndrome	68	36 (52.9
Neonatal jaundice	10	0 (0.0)
Congenital anomalies	1	1 (100.0)
Congenital heart disease	2	2 (100.0)
Intraventricular hemorrhage	1	1 (100.0)
Hypoxic ischemic encephalopathy	2	0 (0.0)
Other	2	1 (50.0)
Moderately preterm (32–33 seeks)	168	27 (16.1)
Infection	100	6 (6.0)
Respiratory distress syndrome	25	1 (4.0)
Neonatal jaundice	16	0 (0.0)
Congenital anomalies	13	12 (92.3)
Anemia	5	0 (0.0)
Congenital heart disease	0	0 (0.0)
Intraventricular hemorrhage	2	2 (100.0)
Hypoxic ischemic encephalopathy	2	2 (100.0)
Hypoglycemia	1	0 (0.0)
Other	4	4 (100.0)
Late preterm (34–36 weeks)	209	16 (7.7)
Infection	70	9 (12.9)
Respiratory distress syndrome	4	0 (0.0)
Neonatal jaundice	71	2 (2.8)
Congenital anomalies	8	1 (12.5)
Anemia	2	0 (0.0)
Congenital heart disease	1	0 (0.0)
Intraventricular hemorrhage	1	1 (100.0)
Hypoxic ischemic encephalopathy	5	2 (40.0)
Hypoglycemia	3	0 (0.0)
Other	44	1 (2.3)
Unclassified	6	1 (16.7)

Table 4.

Admitting diagnosis and mortality by place of birth

Primary diagnosis	Inborn		Out-born
	Cases, n	Mortality, n (%)	Cases, n	Mortality, n (%)
Neonatal jaundice	186	1 (0.5)	552	59 (10.7)
Infection	84	1 (1.2)	465	48 (10.3)
Prematurity with associated complications	385	101 (26.2)	270	103 (38.1)
Congenital anomalies	48	21 (43.8)	169	48 (28.4)
Hypoxic ischemic encephalopathy	79	15 (19.0)	450	114 (25.3)
Meconium aspiration syndrome/transient tachypnea of the newborn	61	0 (0.0)	45	3 (6.7)
Macrosomia	21	0 (0.0)	5	0 (0.0)
Hypoglycemia	15	0 (0.0)	0	0 (0.0)
Anemia	6	0 (0.0)	28	2 (7.1)
Other	50	1 (2.0)	40	4 (10.0)
Total	935	140 (14.9)	2024	381 (18.8)

The current causes of mortality were compared with those reported about 20 years ago from the study center,⁹ as shown in Table 5. Prematurity, HIE, and congenital abnormalities were the leading conditions associated with mortality in our study, compared with asphyxia, jaundice, and infection in this earlier study.⁹

Table 5.

Trends in causes of neonatal mortality at the study center

Ezeaka et al.9 (2003)		Current study (2021)
Disease condition	Contribution to overall mortality, %	Disease condition	Contribution to overall mortality, %
Asphyxia	35.0	Prematurity with associated complications	39.2
Jaundice	23.3	Hypoxic ischemic encephalopathy	24.8
Infection	19.4	Congenital abnormalities	13.0
Prematurity	11.7	Jaundice	11.5
Tetanus	4.9	Infection	9.4
Other	5.7	Other	2.1

DISCUSSION

Neonatal mortality remains an important metric to follow globally because it reflects the general health and well-being of the population at large. The disparities in neonatal mortality generally reflect the inequalities in society.¹⁵ Sub-Saharan Africa—and Nigeria, in particular—have both shown great success in improving neonatal mortality, but need accelerated efforts in the future to work toward the WHO SDGs.¹ The prevalence of neonatal mortality in our study was 17.6% of the study population. This has decreased somewhat from the 21.1% neonatal mortality reported by Ezeaka et al.⁹ from the same tertiary center. Globally, the three most common causes of neonatal mortality are prematurity with associated complications, asphyxia, and sepsis.^4,16 In our study, however, after prematurity with associated complications and asphyxia, congenital anomalies and jaundice exceeded infection as the third and fourth most common diagnoses associated with neonatal mortality, respectively.

Prematurity has become more prevalent in sub-Saharan Africa, including Nigeria, in the past two decades. The WHO recommends interventions for improving outcomes for preterm infants, including the use of antenatal corticosteroids, surfactant, continuous bubble positive airway pressure ventilation, provision of warmth using kangaroo mother care, and prompt use of antibiotics to treat infections.⁵ When compared with the study by Ezeaka et al.⁹ from the same center in 2003, prematurity with associated complications ranked fourth in the cause of neonatal mortality; however, it is now the highest cause of mortality in our study. The most common causes of death among preterm infants in our study were RDS, and infection. This is not surprising because most of the infants were out-born and most likely had no benefit of antenatal steroids, which has been shown to improve outcome in preterm infants. The prohibitive cost of surfactant also limits its access to the very preterm and extremely preterm infants who require it. There were no functional neonatal ventilators for infants who required additional support beyond CPAP during the study period. In addition, during the study period, there were also insufficient commercial CPAP machines at the center. Therefore, most of the infants needing CPAP were placed on improvised CPAP devices,¹¹ which may not be as effective as the standardized commercial devices. These factors, in addition to out-of-hospital deliveries, which reduce the possibility of antenatal corticosteroids for preterm births, further increase the risk of death from RDS.

The provision of improved neonatal care along with high-quality, low-cost commercial CPAP machines; ventilators; subsidized surfactant; and blended oxygen are urgently needed if we are to decrease neonatal morbidity and mortality significantly. This is important, especially for the extreme and early preterm neonates, with 88.2% and 45.4% mortality, respectively, and who jointly make up about 30% of the neonatal mortality in our study. The main causes of death in these two groups were RDS and sepsis.

Although majority of preterm admissions in our study were inborn infants, there was greater mortality (38.1%) among the out-born preterm admissions compared with inborn admissions (26.2%). The majority of these infants are usually transported to the hospital in suboptimal conditions,^17,18 and are often hypothermic and in respiratory distress on arrival. They have also been exposed to conditions that increase the risk of sepsis, such as inappropriate cord care.¹⁷ These facts may explain the increased mortality in this group of neonates. Education of mothers and health-care providers on the importance of hospital delivery and early referral to facilities that are equipped to care for these vulnerable infants will reduce mortality in this group.¹⁹ The importance of appropriate means of transport with a prior notification of the receiving health-care facility is also very important to improve the outcome for these infants.^19,20

Asphyxia, on the other hand, which was reported as the second highest morbidity 20 years ago, is now the fourth most common cause of morbidity in our study and also in a multicenter study on the burden of disease and risk factors for mortality among hospitalized newborns in Nigeria and Kenya.²¹ This is probably a reflection of the positive effect of increased training of health-care workers on neonatal resuscitation, step-down training to other health-care workers,²² and helping babies breathe training on newborn resuscitation. Possible reasons for the high number of deaths from asphyxia despite the reduced number of cases include the unavailability of facilities for therapeutic hypothermia and that only the more severe cases are referred whereas the other milder cases recover after resuscitation. It is also possible that the skills acquired by the trained health-care workers have depreciated over time. It is therefore important to review the areas from which these cases are being referred so that the health-care workers from facilities in these areas are trained in newborn resuscitation if they have not benefited from neonatal resuscitation training in the past. Other possible risk factors for the delivery of asphyxiated infants should also be explored for necessary intervention.

Congenital anomalies, which did not feature as a cause of death almost 20 years ago at the study center, was the third-highest cause of mortality. The improvement in antenatal care, delivery, neonatal care, diagnosis and treatment of infections may be responsible for the increasing number of congenital anomalies as the cause of neonatal deaths. Therefore, it is important to start measures to reduce congenital anomalies such as antenatal folic acid, focused antenatal care, and education of mothers on the need to avoid indiscriminate use of drugs, especially during early pregnancy, which may compromise the infant. The use of routine antenatal anomaly scans for all pregnant women will also aid in the early detection of anomalies and in prompt intervention for neonates with congenital anomalies.

Neonatal jaundice is common worldwide and is a common cause of readmission,²³ but rarely as a cause of death or even significant morbidity.²⁴ In stark contrast to high-income countries, where jaundice-related death is exceedingly rare, jaundice is often one of the leading causes of neonatal death in low- to middle-income countries (LMICs). This is especially true in Nigeria, where it is frequently one of the four most common causes of death. As noted in a systematic review and meta-analysis by Slusher et al.,²⁴ as well as by others, including Ogunfowora in Nigeria,²⁵ Eze in Yemen,²⁶ and Jajoo in India,²⁷ many infants still experience complications from severe neonatal jaundice and die of this preventable cause of morbidity and mortality. In a recent article by Olusanya et al.,²⁸ neonatal jaundice ranked seventh as a cause of early neonatal death.

The reasons for the severe morbidity and mortality from neonatal jaundice in LMICs are multifactorial.^29,30 They include the high rate of unsupervised home births, with more than 60% of deliveries occurring outside of hospitals³¹; lack of routine screening for neonatal jaundice before discharge for infants who are born in hospitals³⁰; a high incidence of glucose-6-phosphate dehydrogenase deficiency combined with the use of hemolytic agents and sepsis.^32,33 Additional factors include the high likelihood of ineffective phototherapy from sub-standard devices and erratic power supply; and the inability to perform emergent blood transfusions for several reasons, such as unavailability of blood.^30,34 When jaundice develops, it is discovered late, and mothers present late to the hospital after trying various ineffective home remedies to no avail.³⁵ Our finding that jaundice was the most common morbidity is consistent with reports from an earlier study at the center about two decades ago⁹ and a recent study of neonatal outcomes in secondary and tertiary facilities in Nigeria (study center included) and Kenya.²¹

There has been no widespread active training on the early recognition of neonatal jaundice over the years; however, this is now included in the essential care for every newborn (ENCC), but the effect of the addition of neonatal jaundice to these programs is yet to be seen both in term and preterm infants included in our study. One recent small study from Nigeria demonstrated that maternal education is effective in significantly reducing acute bilirubin encephalopathy.³⁶ In addition, a packaged approach to prevention and treatment has been instituted in some LMICs and has been successful in reducing severe neonatal jaundice,³⁷ and should be widely implemented in Nigeria.

The incidence of neonatal infections in our study is less than that reported in an earlier study by Lawn et al.,³⁸ in which sepsis was estimated as a cause of 26% of all neonatal deaths worldwide and the 22% reported in Nigeria.³⁹ Infections also ranked fifth in our study as a cause of mortality, compared with third in the 2003 study.⁹ Possible reasons for the decline in the contribution of infections to neonatal deaths in our study include the improved laboratory support for cultures, availability of life-saving antibiotics in the facility, the low threshold for treatment of neonatal sepsis, and the increase in infection control awareness and prevention.

When considering further the morbidities associated with mortality and place of birth, the percentage of mortality within the disease cohort was greater for all the conditions except congenital abnormalities in out-born infants. A plausible explanation for this is the referral nature of the study center, which encourages referrals of complicated cases, including those with in utero diagnosis of abnormality around the time of delivery. This precludes adequate preparation and counseling of the parents, especially in cases of anomalies that are not compatible with life.

The major limitations of our study include its retrospective nature, the single-center data source, and the resulting limited depth of analysis. That being said, LUTH is the largest teaching hospital in Nigeria, with an average of 1,200 births and 1,250 neonatal admissions (including out-born admissions) annually. Although a single center’s experience is not generalizable, given the location and the size of the neonatal unit at LUTH, we do believe our study findings are an important reflection of the current neonatal morbidity and mortality in urban centers in Nigeria. Future studies should include larger and more diverse populations, multiple centers, and more detailed analyses of patient characteristics to identify risks and possible strategies for improvement.

CONCLUSION

In conclusion, neonatal mortality is still high at LUTH and reflects the national figures and trends. Prematurity with associated complications, HIE, jaundice, infections, and congenital anomalies remain the leading causes of neonatal mortality. Efforts to improve antenatal care are crucial to prevent preterm births, and when this is inevitable, facilitation of delivery in or rapid appropriate transport to facilities staffed and equipped to handle these infants is imperative. Facilities for improved preterm care should be provided in secondary and tertiary hospitals. More awareness should be created on neonatal jaundice, with a packaged approach to prevention and effective treatment, including high-quality phototherapy at all levels of care, thus preventing severe neonatal jaundice and its sequelae. Training on neonatal resuscitation, essential newborn care, and care of small and sick infants should be intensified, in addition to the provision of low-cost medical devices⁴⁰ to improve care. These efforts will help to reduce neonatal mortality and move the country closer to achieving SDG goal 3.2.