High prevalence of congenital heart disease among hospitalized neonates in northwestern Tanzania; implications of maternal and environmental factors

Abstract

Introduction

There is paucity of epidemiological data on neonatal congenital heart disease (CHD) in Tanzania, hindering evidence based policy or protocol development and resources allocation to improve health care services. Consequently, neonates with CHD are subjected to late referrals, disease complications and high treatment costs. This study aimed to determine prevalence, patterns, and factors associated with CHD among neonates admitted to a tertiary hospital, northwestern-Tanzania.

Methodology

A cross-sectional study was conducted from November 2022 to March 2023 involving 513 admitted neonates. Detailed physical examination and screening echocardiography was done to all participants. Data was analyzed using STATA version 13.

Results

The overall prevalence of CHD was 20.3% (104/513) which decreased to 10.3% (53/513) after excluding small defects with high likelihood of spontaneous closure. From the overall CHD, acyanotic CHD were the commonest by 83.7% (87/104). Severe CHD accounted for 6.8% (35/513). History of residing near mining areas (OR 2.9[CI: 1.2–6.9], P-0.017) and lack of folic acid supplementation during the first trimester (OR 2.5[CI: 1.4–4.4], P- 0.001) were significant maternal risk factors. While, Pre-ductal oxygen saturation of <90% (OR 3.2[CI: 1.4–7.3], P-0.005), tachycardia (OR 2.8[CI: 1.3–5.9], P-0.007), and presence of central cyanosis (OR 3.9[CI: 1.1–13.6], P-0.031) were significant clinical predictors.

Conclusion

Strategies are needed to strengthen reproductive health services among women of child bearing age and promote environmental health policies in mining and nearby areas. Clinicians are urged to perform thorough cardiovascular examination together with diagnostic services to facilitate early detection of CHD, timely referrals and appropriate management.

1. Introduction

Globally, Congenital Heart Disease (CHD) represent a major health challenge, accounting for about 28% of all congenital anomalies [1,2]. The worldwide incidence of CHD ranges between 40 and 75 per 1000 live births and among those children, 44 per 1000 need intervention within the first year of life [3,4]. Approximately 500,000 children are born in Africa with CHD each year, the majority of whom are in sub-Saharan Africa, where access to health care is limited with shortage of specialized expertise and cardiac diagnostic facilities [5].

The presence of CHD during neonatal period cannot be thoroughly ascertained by physical cardiovascular examination or pulse oximetry alone [6,7], echocardiography remains the gold standard in determining the actual prevalence of CHD, diagnosing even the asymptomatic lesions during the neonatal period allowing for timely identification and intervention including referral. Neonatal echocardiography services requires proper resources and relevant expertise which are not readily available in most of regional and district hospitals in Tanzania. Bugando Medical Centre (BMC) is a tertiary hospital serving a large population in the Northwestern part of Tanzania with a significant number of children having CHD and other congenital malformations [8,9]. About three quarter of children attending the pediatric echocardiography clinic at BMC have CHD, with 58.9% of them being diagnosed within the first year of life and 60% requiring surgery. Among those who require surgery, 6.3% are inoperable due to irreversible complications such as pulmonary hypertension at the first presentation in the clinic. Less than a quarter of them are diagnosed within the first month of life, indicating that most of the neonates with CHD are missed during the neonatal period [9].

Unrecognized CHD are associated with heart failure, under-nutrition, poor neurodevelopmental outcomes, poor productivity, pulmonary complications such as pulmonary hypertension and recurrent pneumonia, and increased mortality [1,[10], [11], [12]]. Data on the prevalence, pattern and factors associated with congenital heart disease among neonates in our setting is scarce. Lack of this important epidemiological information hinders local and national healthcare policy makers from initiating screening programs, effectively allocating resources for specialized cardiac care services and timely referrals, and strategically implementing preventive measures. We believe data from this study will help to improve pediatric cardiac services in the region and ultimately reduce morbidity and mortality due to CHD.

2. Methodology

2.1. Study area

This study was conducted in the pediatric and child health department at BMC, which serves as a zonal consultant and teaching hospital with a 950-bed capacity located at the shores of Lake Victoria in Mwanza, Tanzania. The hospital caters to a population of about 16 million people from eight regions namely; Kagera, Geita, Mwanza, Shinyanga, Simiyu, Kigoma, Mara and Tabora. Economical activities varies from one region to the other, however fishing and mining activities are common in most of the regions. The pediatric department at BMC has a total bed capacity of around 130 beds and approximately 50 neonates are admitted to the neonatal ward and neonatal intensive care unit (NICU) weekly from in-house deliveries and referrals.

2.2. Study design and population

This was a hospital based cross-sectional analytical study conducted from November 2022 to March 2023 involving all neonates aged 0 to 28 days of life who were admitted to BMC neonatal ward and NICU. Neonates whose mothers/caretakers provided a written informed consented were enrolled into the study, however premature babies born at less than 28 weeks gestational age and as.

2.3. Data collection procedure

A standardized data collection tool was used to collect participants’ information. Demographic data such as age, sex, gestational age, birth weight, place of residency, and maternal risk factors were recorded. Clinical examination was reported based on the presence of central cyanosis, respiratory rate, heart rate, pre-ductal and post-ductal oxygen saturation, cardiac murmur, hepatomegaly and presence of edema. Information on the sex of participants was collected from the mothers which based on sex assigned at birth by appearance of the external genitalia. A bedside transthoracic echocardiography using SonoSite M-Turbo portable machine was then performed to every enrolled neonate at varying postnatal ages. Echocardiography was performed in the presence of the mother or caregiver after the neonate has been fed and while lying quietly to maximize the quality of the image. Standard two-dimensional views were obtained; a four-chamber apical view, parasternal long axis and short axis view, and subcostal and suprasternal views as well as color flow Doppler.

CHD was defined as the presence of any structural abnormalities of the heart or intra-thoracic great vessels that are actually or potentially of functional significance. The diagnosis of CHD was made in any newborn with single or multiple structural defects. As a repeat echocardiography was not performed to account for physiological patency of ductus arteriosus, neonates with hemodynamically stable, very small Patent Ductus Arterisus (≤1.4 mm) and Patent Foramen Ovale were considered to have normal echocardiographic findings in final analysis determining for overall prevalence of CHD. Furthermore to improve comparability with other studies we determined prevalence of clinically significant CHD by excluding defects with high likelihood of spontaneous closure including isolated small Patent Ductus Arteriosus (PDA), Secundum Atrial Septal Defects (ASD) and Ventricular Septal Defects (VSD). A PDA was considered small if the ranged ≥1.5 to ≤ 3 mm, moderate when ranged >3 to ≤ 5 mm and large if size exceeded 5 mm. An ASD was considered small at < 5 mm, moderate when ranged between 5 mm and 8 mm and large if size was > 8 mm. A VSD was regarded small at <3 mm, moderate when ranged between 3 and 5 mm and large if size was > 5 mm.

CHD were first classified into cyanotic or acyanotic defects, and further classified into severity categories as mild, moderate and severe cardiac lesions according to Hoffman et al. [4]. In case of multiple CHD defects observation, the most severe anomaly was considered. During the study neonates with CHD were reviewed, managed and if need be referred to the Cardiac Institute by the clinical managing team.

2.4. Data analysis

Data was entered into Microsoft excel database and analyzed by using STATA version 13 (Stata Corp, College Station, Texas, USA). Descriptive statistics were summarized in form of proportions and frequency tables for categorical variables and continuous variables were summarized using means (standard deviation) or median (interquartile range). Univariable logistic regression analysis followed by multivariable logistic regression analysis was used to determine variables that associated with overall CHD. Factors identified in the univariable analysis with a p-value <0.20 were subsequently included in the multivariable logistic regression model. A p-value of ≤0.05 at a 95% confidence interval (CI) was considered statistically significant.

3. Results

3.1. Participant's enrollment

During study period, a total of 1100 neonates were admitted to the neonatal ward and NICU at BMC and 1022 neonates were eligible to be recruited into the study i.e. they were born at gestation age ≥28 weeks. At the end of study time, 513 neonates with parental/caretakers consent and complete study procedures were enrolled and included in the final analysis. A total of 509 neonates were not enrolled or included in the final analysis due to various reasons as depicted in “Fig. 1”.

Fig. 1.

Flow chart of study Participants.

3.2. Demographic and clinical characteristics of 513 neonates

The median age was 5 [IQR 3-9] days with about 347 (67.6%) of the neonates aged ≤7 days. The majority of neonates 287 (55.9%) were males and 2 (0.4%) had ambiguous genitalia. Most of the neonates 371 (72.3%) were born at a gestation age of ≥37 weeks or more. The median birth weight was 2.9 [IQR 2.2-3.3] kilograms, with 352 (68.6%) neonates being born with ≥2.5 kg (Table 1). On the clinical characteristics, 74 (14.4%) neonates had dysmorphic features, 123 (24.0%) had obvious malformation, where, neural tube defects and gastroschisis were the commonest (Table 2).

Table 1.

Social-demographic and other baseline characteristics of 513 neonates.

Characteristics	Number (%) or Median [IQR] n = 513
Sex
Female	224 (43.7)
Male	287 (55.9)
Ambiguous	2 (0.4)
Age
≤7 days	347 (67.6)
>7 days	166 (32.4)
Age in days	5 [3 – 9]
Gestation Age at birth
<37 weeks	142 (27.7)
≥37 weeks	371 (72.3)
Gestation Age at birth in weeks	38 [36 – 40]
Birth weight
<2.5 kg	161 (31.4)
≥2.5 kg	352 (68.6)
Birth weight in kilogram	2.9 [2.2 – 3.3]
Maternal age groups
<35 years	438 (85.4)
≥35 years	75 (14.6)
Maternal Age in years	27 [23 – 31]
Marital status
Married	453 (88.3)
Not married	60 (11.7)
Maternal Educational level
≤7 years in school	261 (50.9)
>7 years in school	252 (49.1)
Maternal employment
Employed	290 (56.5)
Unemployed	223 (43.5)
Maternal Diabetes Mellitus
No	497 (96.9)
Yes	16 (3.1)
Maternal fever in the 1st trimester
No	356 (69.4)
Yes	157 (30.6)
Maternal Antidepressant use in 1st trimester
No	512 (99.8)
Yes	1 (0.2)
Maternal Anticonvulsant use in 1st trimester
No	509 (99.2)
Yes	4 (0.8%)
Maternal Alcohol use
No	489 (95.3)
Yes	24 (4.7)
Maternal Folic Acid use in 1st trimester
No	353 (68.8)
Yes	160 (31.2)
Consanguinity
No	501 (97.7)
Yes	12 (2.3)
Maternal HIV infection
No	494 (96.3)
Yes	19 (3.7)
Parity
<5 times	443 (86.4)
≥5 times	70 (13.6)
CHD in the family
No	506 (98.6)
Yes	7 (1.4)
Residency near mining areas
No	490 (95.5)
Yes	23 (4.5)

CHD; Congenital Heart Disease, HIV; Human Immunodeficiency Virus.

Table 2.

Clinical characteristics of 513 neonates.

Characteristics	Number (%) n = 513
Dysmorphic features
No	439 (85.6)
Yes	74 (14.4)
Obvious Malformation
No	390 (76.0)
Yes	123 (24.0)
Central Cyanosis
No	497 (96.9)
Yes	16 (3.1)
Respiratory Rate per minute
40-60	392 (76.4)
>60	112 (21.8)
<40	9 (1.8)
Preductal saturation
≥90%	475 (92.6)
<90%	38 (7.4)
Heart Rate per minute
100-160	458 (89.3)
>160	55 (10.7)
Cardiac Murmur
No	446 (86.9)
Yes	67 (13.1)

3.3. Prevalence and distribution of congenital heart disease among 513 neonates

In Fig. 2, we present the overall prevalence of CHD, which was 20.3% (104/513)Among them, 83.7%(87/104) were acyanotic CHD. The commonest acyanotic CHD being Patent Ductus Arteriosus (PDA), followed by Atrial Septal Defects (ASD) and Ventricular Septal Defects (VSD). Meanwhile the commonest cyanotic CHD was Transposition of Great Arteries (TGA). After excluding isolated small PDA, ASD and VSD, the prevalence of clinically significant CHD, was 10.3% (53/513). Fig. 3 presents the distribution of the CHD in percent per total study population, alongside the severity categoriesas adopted from Hoffman et al. (5).

Fig. 2.

Prevalence and types of Congenital Heart Disease among 513 neonates

Fig. 3.

Distribution of Congenital Heart Disease and classification based on severity categories. (ASD; Atrial Septal Defect, AVSD; Atrioventricular Septal Defect, HPLHS; Hypoplastic Left Heart Syndrome, PDA; Patent Ductus Arteriosus, TGA; Transposition of Great Arteries, VSD; Ventricular Septal Defect, CHD; Congenital Heart Disease).

3.4. Maternal factors associated with congenital heart disease among 513 neonates

Maternal characteristics that significantly associated with presence of CHD included; lack of folic acid supplementation during the first trimester (OR 2.5 [95% CI, 1.4-4.4] P = 0.001) and history of residing near mining areas (OR 2.8 [95% CI, 1.2-6.9] P = 0.017). Maternal consumption of alcohol in the first trimester had borderline statistical significance with OR 2.4 [95% CI, 1.0-5.6] and P = 0.050 (Table 3).

Table 3.

Demographic and other baseline factors associated with CHD among 513 neonates.

Variables	CHD		Univariable		Multivariable
	NO (N = 409) n(%)	YES (N = 104) n(%)	OR (95 % CI)	P-value	OR (95%CI)	P-value
Sex
Female	181(80.8)	43(19.2)	1.0
Male	227(79.1)	60(20.9)	1.1(0.7—1.7)	0.63
Ambiguous genitalia	1(50.0)	1(50.0)	4.2(0.3—68.7)	0.31
Gestation age
< 37 weeks	115(81.0)	27(19.0)	1.0
≥37 weeks	294(79.3)	77(20.7)	1.1(0.7—1.8)	0.661
Birth weight (kg)
<2.5	127(78.9)	34(21.1)	1.0
≥2.5	282(80.1)	70(19.9)	1.0(0.6—1.5)	0.747
Maternal Age (Years)
<35	350(79.9)	88(20.1)	1.0
≥35	59(78.7)	16(21.3)	1.1(0.6—2.0)	0.805
Residency near mining area
No	396(80.8)	94(19.2)	1.0
Yes	13(56.5)	10(43.5)	3.2(1.4—7.6)	0.007	2.8(1.2—6.9)	0.017
Maternal Education
> 7 years in school	214(82.0)	47(18.0)	1.0
≤ 7 years in school	195(77.4)	57(22.6)	1.3(0.9—2.1)	0.195
Maternal Diabetes Mellitus
No	397(79.9)	100(20.1)	1.0
Yes	12(75.0)	4(25.0)	1.3(0.4—4.2)	0.634
Maternal Fever
No	290(81.5)	66(18.5)	1.0
Yes	119(75.8)	38(24.2)	1.4(0.9—2.2)	0.142
Maternal use of Anticonvulsants
No	407(80.0)	102(20.0)	1.0
Yes	2 (50.0)	2(50.0)	4.0(0.6—28.7)	0.169
Maternal use of Alcohol
No	395(80.8)	94(19.2)	1.0
Yes	14(58.3)	10(41.7)	3.0(1.3—7.0)	0.011	2.4(1.0—5.6)	0.050
Maternal use of Folic Acid
Yes	143(89.4)	17(10.6)	1.0
No	266(75.3)	87(24.7)	2.8(1.6—4.8)	<0.001	2.5(1.4—4.4)	0.001
Consanguinity
No	401(80.0)	100(20.0)	1.0
Yes	8(66.7)	4(33.3)	2.0(0.6—6.8)	0.264
Maternal HIV infection
No	394(79.8)	100(20.2)	1.0
Yes	15(79.0)	4(21.0)	1.1(0.3—3.2)	0.931
History of CHD in Family
No	404(79.8)	102(20.2)	1.0
Yes	5(71.4)	2(28.6)	1.6(0.3—8.3)	0.586
Parity
<5 times	355(80.1)	88(19.9)	1.0
≥5 times	54(77.1)	16(22.9)	1.2(0.7—2.2)	0.563

CHD; Congenital Heart Disease, HIV; Human Immunodeficiency Virus.

3.5. Clinical predictors of congenital heart diseases among 513 neonates

Central cyanosis (OR 3.9 [95% CI, 1.1-13.6] P = 0.031), pre-ductal saturation of <90% (OR 3.2 [95% CI, 1.4-7.3] P = 0.005, and heart rate of >160 beats per minute (OR 2.8 [95% CI, 1.3-5.9] P = 0.007) were among the neonatal clinical characteristics that significantly associated with the presence of CHD (Table 4). Due to physiologic relation of these clinical variables, variance inflation factors (VIF) analysis was performed and all the values were below 5, indicating no significant multi-collinearity was present among them.

Table 4.

Clinical predictors of CHD among 513 neonates.

Variables	CHD		Univariable		Multivariable
	NO (N = 409) n (%)	YES (N = 104) n (%)	OR (95%CI)	P-value	OR (95%CI)	P-value
Dysmorphic features
No	353(80.4)	86(19.6)	1.0
Yes	56(75.7)	18(24.3)	1.3(0.7—2.3)	0.350
Obvious Malformation
No	311(79.7)	79(20.3)	1.0
Yes	98(79.7)	25(20.3)	1.0(0.6—1.7)	0.987
Central Cyanosis
No	404(81.3)	93(18.7)	1.0
Yes	5(31.2)	11(68.8)	9.6(3.2—28)	<0.001	3.9(1.1—13.6)	0.031
Pre-ductal Oxygen Saturation
≥90%	394(83.0)	81(17.0)	1.0
<90%	15(39.5)	23(60.5)	7.5(3.7—15)	<0.001	3.2(1.4—7.3)	0.005
Respiratory rate (breaths/min)
40-60	329(84.0)	63(16.0)	1.0
>60	72(64.3)	40(35.7)	2.9(1.8—4.7)	<0.001	1.3(0.7—2.4)	0.386
<40	8(88.9)	1(11.1)	0.7(0.1—5.3)	0.69	0.7(0.1—5.9)	0.764
Heart rate (beats/min)
100-160	383(83.6)	75(16.4)	1.0
>160	26(47.3)	29(52.7)	5.7(3.2—10)	<0.001	2.8(1.3—5.9)	0.007

4. Discussion

The birth prevalence of CHD has been reported to increase progressively over time. In 2011, Linde et al. report an estimated global prevalence of 9 per 1000 live births by 2009 [1], which remained relatively the same at 9.4 per 1000 live births until 2017 as reported by Liu et al. [2]. However, in 2020, GBD 2017 Congenital Heart Disease Collaborators, have reported a substantially higher global birth prevalence of approximately 18 per 1000 live births [13], further highlighting on the rising trends. Furthermore, the spectrum of CHD lesions included in the analysis has been shown to significantly widen the estimates from approximately 6 per 1000 live births when only moderate to severe lesions are considered, to as high as 75 per 1000 live births when trivial lesions are included [4]. There is a notable geographical heterogeneity in global birth prevalence of CHD, with Africa having the lowest prevalence at approximately 2.3 per 1000 live births [2], a finding that may be largely be due to limited data. However, recent individual studies among African newborns have reported substantially higher birth prevalence ranging from 14 to 81 per 1000 live births [14,15], suggesting that there could be a considerable underestimation of true burden of CHD among newborns in this region.

Reflecting on both the global and individual African estimates of CHD birth prevalence, we report a much higher prevalence in our study both before and after exclusion of small cardiac defects at 20.3% and 10.3%, respectively. This higher prevalence could largely be due to the following reasons. First, this was a tertiary level hospital based study, conducted among sick neonates admitted to neonatal wards including the NICU, comprising both in-borns and out-borns referred from lower level healthy facilities within the region and across other seven regions in the Lake zone, thus higher likelihood of over representation of CHD cases. Second, the inclusion of small cardiac defects with high chances of spontaneous closure, may have further increased the prevalence. Thus, our study prevalence need to be interpreted within the clinical context and not as population based prevalence.

Despite the challenges of reporting birth prevalence estimates across studies through variations in timing of screening and definition of clinically significant CHD; echocardiographic screening in neonatal period remains critically important for early detection of CHD, as it allows detection of both clinically significant lesions requiring timely intervention and minor defects that require follow up.

The types of CHD in our study was similar to that from other studies, with acyanotic CHD being the commonest type of lesions [[13], [14], [15], [16], [17], [18], [19], [20]]. PDA was the commonest acyanotic CHD in our study. This finding contrasts with findings from other studies in which ASD [13] or VSD [14,20] predominated. Although the exact reason for this difference across studies is unclear, the preponderance of PDA lesions in our study could likely be related to the timing of echocardiographic screening. The median age at echocardiography in our study population was 5 days, an age period where postnatal ductal closure may be incomplete particularly in preterm and critically ill neonates. Among cyanotic CHD, TGA was the most common lesion in our study is similar to other literature reporting TGA as the commonest cyanotic CHD in the neonatal period [13]. In contrast, other studies have reported TOF as the most common cyanotic CHD [9,19]. This could be explained by the differences in age distribution of study populations, as those other studies included older children. Such approaches increases the possibility of missing out patients with TGA who typically presents early in neonatal period and is associated with high mortality when intervention is delayed.

Furthermore, we explored for association between CHD and socio-demographics and maternal factors. We found that lack of folic acid supplementation during the first trimester was independently associated with increased odds of having CHD in indexed neonates. Similar findings have been reported elsewhere stating that pre-conception folic acid supplementation reduces the risk of CHD and lower dietary folate intake during pregnancy increases the risk of CHD [[21], [22], [23]]. In Tanzania folic acid supplementation is part and parcel of care provided to pregnant women attending antenatal clinics mostly to prevent or treat anemia during pregnancy, however studies have revealed low uptake of folic acid among Tanzanian pregnant mothers [24,25] influenced by various factors. Strategies to improve pre-conception intake of folic acid among women of reproductive age with continuation throughout the pregnancy as recommended by WHO, should be a way forward because of its documented benefits in preventing occurrence of congenital anomalies including CHD.

We also observed a statistically significant association between maternal residence near mining areas and presence of CHD among neonates. Environmental contamination by heavy metals from mining activities has been reported to increase the risk of CHD in the offspring of exposed mothers [[26], [27]]. Further studies are warranted to explore the mechanistic effect of toxic elements involved in mining and occurrence of congenital heart diseases in our setting, meanwhile reinforcement of safety regulations and precautions in the mining areas and nearby areas is needed to control contamination of the environment with toxic elements from mining activities.

We further observed a borderline association between maternal consumption of alcohol during pregnancy and occurrence of CHD in neonates. Despite the fact that this finding aligns with what is already documented in literature regarding the teratogenic effect of alcohol to the developing heart [28,29], larger studies are warranted to establish statistically significant association.

For the neonatal clinical characteristics, presence of central cyanosis, pre-ductal saturation below 90% and tachycardia (>160 beats/min) were the factors that independently predicted the presence of CHD among the study participants. Occurrence of clinically apparent cyanosis depends on absolute concentration of deoxygenated hemoglobin and occurs when oxygen saturation falls below 80%. On the other hand Pulse oximetry can detect hypoxemia before overt cyanosis becomes apparent. In neonates with CHD, the onset of desaturation depends on the type and complexity of CHD that could either present with inadequate mixing of deoxygenated blood with oxygenated blood e.g. in patient with TGA with intact intra-ventricular septum or present with the ductal dependent pulmonary/systemic circulations in lesions with ventricular outflow obstruction [30]. However, presence of tachycardia could reflect systemic stress beyond hypoxemia. In a neonate with CHD, tachycardia can be explained by their structure abnormalities of the heart or as a sign of compensation in those presenting with heart failure. Other non-cardiac causes presenting with tachycardia such as fever and anemia could have contribution to the findings as well. In conclusion, these significant neonatal clinical characteristics reinforces the importance of thorough and serial cardiovascular examination among newborns which when combined with use of pulse oximetry can substantially improve early detection and referral in resource constraint settings where access to echocardiography is limited.

5. Limitations

This study has several limitations. First, being a cross-sectional study the causal relationship between exposures and outcome cannot be ascertained. Second, the study could have been subjected to recall bias as some of the information obtained from mothers may have been affected by poor recalling and to selection bias considering the convenient sampling method applied. Finally, being a single center study our study findings may have limited generalizability. However, they provide important data from Sub-Saharan Africa where CHD data is scarce, thereby enhancing the global health relevance of the study and contributing to understanding the burden and spectrum of CHD in under researched populations.

6. Conclusion

The overall prevalence of CHD among neonates admitted at Bugando Medical Centre was 20.3% and that of clinically significant CHD was 10.3%. Maternal history of residing near mining areas and lack of folic acid supplementation during the first trimester were strongly associated with increased odds of CHD among neonates. Presence of central cyanosis, pre-ductal saturation of less than 90%, and heart rate of more than 160 beats per minute were among the neonatal clinical characteristics that independently predicted the presence of CHD in a neonate.

7. Recommendations

Clinicians should have high suspicion of CHD in every neonate with maternal history of residing near mining areas and lack of folic acid supplementation in the first trimester, as well as for those neonates presenting with clinical symptoms of central cyanosis, low pre-ductal oxygen saturation less than 90%, and tachycardia. Early referral for expert review and possible echocardiography should be done as early as possible to identify those with severe CHD who need urgent intervention. Promotion of folic acid supplementation before conceiving and during the first trimester of pregnancy should be done to minimize the risk of congenital malformations including CHD. The findings also highlight the need for emphasizing on implementation of environmental health policies to reduce exposure potentially teratogenic heavy metals contaminating the environment in the mining areas and nearby residencies. A follow-up study to assess temporal trends of CHD prevalence and the impact of spontaneous closure of mild CHD on prevalence is recommended.

CRediT authorship contribution statement

Martha B. William: Writing – original draft, Project administration, Methodology, Formal analysis, Data curation, Conceptualization. Florentina Mashuda: Writing – review & editing, Supervision, Data curation. Neema Chami: Writing – review & editing. Delfina R. Msanga: Writing – review & editing. Julieth Kabirigi: Writing – review & editing, Supervision. Neema Kayange: Writing – review & editing. Tumaini Mhada: Writing – review & editing. Tulla Masoza: Writing – review & editing, Supervision, Methodology, Data curation.

Ethical consideration

This study complied with the principles of Declaration of Helsinki. Ethical clearance for the research proposal was sought from the joint Bugando Medical Centre and Catholic University of Health and Allied Sciences (BMC/CUHAS) ethics and review committee with a certificate number CREC/627/2022 obtained on 15th November 2022. A written informed consent was obtained from the mothers/caretakers of the neonates who were enrolled into the study. Participants’ privacy and confidentiality was observed throughout the study period.

Availability of data and materials

The research data is not publicly available due to ethical and confidentiality reasons involving patient data, but will be made available upon request from corresponding author and with approval from joint Bugando Medical Centre and Catholic University of Health and Allied Sciences (BMC/CUHAS) ethics and review committee.

Funding

The research did not receive any specific grant for funding agencies in the public, commercial, or not-for profit sectors.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.