Determinants of Necrotizing Enterocolitis Among Neonates in Referral Hospitals of East and West Gojjam Zones of Amhara Regional State, Northwest Ethiopia, 2020.


 Background: Necrotizing enterocolitis is one of the most common serious surgical disorders of neonates. It is a life-threatening emergency of the gastrointestinal tract in the neonatal period that causes morbidity and mortality of newborns. Although the burden of the disease is high and it is the major cause for neonatal death, there is limited information’s about identifying the determinant factors that may help to reduce the neonatal deaths. Thus this study aimed to identify the determinants of necrotizing enterocolitis among admitted neonates in neonatal intensive care unit at referral hospitals of East and West Gojjam Zones of Amhara regional state,Northwest Ethiopia, 2020. Methods: Unmatched case control study design was conducted among 246 neonates (82 cases and 164 controls) in neonatal intensive care unit from February 24 to April 24, 2020. Data was collected through face to face interview and reviewing medical charts of the neonates. Data was entered into Epi data and exported to SPSS for analysis. Bivariable logistic regression analysis was used to identify the candidate variables at p≤0.25 and multivariable logistic regression analysis was employed to identify significant determinants at p value <0.05. Adjusted odds ratio with 95% CI was used to show the strength of association between exposure and outcome variables. Result: A total of 246 neonates were included in the study. Duration of rupture of membrane>18 hours [AOR=4.287; 95%CI (2.157-8.518)], low birth weight [AOR= 3.592; 95% CI (1.742- 7.407)], neonatal sepsis [AOR=3.553; 95% CI (1.76-7.174)] and types of enteral feeding (formula milk only [AOR=3.604; 95% CI (1.548-8.39)] and mixed milk (AOR=2.416; 95% CI (1.103-5.290)]) were determinants for necrotizing enterocolitis.Conclusions: Duration of rupture of membrane, low birth weight, neonatal sepsis and enteral feedings (formula milk and mixed milk) were significantly associated with necrotizing enterocolitis. Encouraging exclusive breastfeeding, providing adequate care for low birth weight babies and using septic precautions to avoid neonatal sepsis are the strategies to prevent necrotizing enterocolitis.


Introduction
Necrotizing enterocolitis (NEC) is the most common serious surgical disorder (1). It is a lifethreatening emergency of the gastrointestinal tract in the newborn period that causes neonatal morbidity and mortality among newborns. It is characterized by various degrees of mucosal or transmural necrosis of the intestine (1,2). The manifestations of necrotizing enterocolitis are abdominal distention, feeding intolerance, vomiting, blood in stool, abdominal wall erythema and systemic instability. Both clinical signs and radiographic findings are important to diagnose necrotizing enterocolitis. Plain abdominal radiographs are essential to make a diagnosis of NEC.
NEC is the most common one of the gastrointestinal emergencies in premature infants in the neonatal intensive care unit (NICU) (5). Premature infants are accounts to 90 % to develop NEC, while, full-term and near-term infants also develop the disease when they expose to risk factors (5,6).
The exact pathogenesis of NEC remains unclear, but prematurity, enteral feeding, bacterial products, and intestinal ischemia have all been shown to cause activation of the inflammatory cascade. Awareness of the risk factors for NEC, early trophic feeding with breast milk and 4 following the established feeding guidelines have been shown to reduce the incidence of NEC (7).
The first 28 days of life (neonatal period) is the most vulnerable time for a child's survival.
Children face the highest risk of dying in their first month of life at a global rate of 18 deaths per 1,000 live births. Globally, an estimated 2.5 million newborns died in the first month of life, approximately 7,000 every day (8).
A study in China showed that the incidence of NEC was 2.50% and 4.53% in the low birth weight and very low birth weight infants, respectively (9). The mortality rates of very low birth weight infants and moderately low birth weight infants with confirmed NEC were 50.2% and 36.3%, respectively. Another study in Canada stated that NEC affects 5-12% of very low birth weight babies and resulted 50 % of fatal cases. The number of cases undergoing to surgery in this study was 20-40% of cases and leads health care cost more expensive (10).
World Health Organization report showed that among the sustainable development goal (SDG) regions, sub-Saharan Africa had the highest neonatal mortality rate in 2018 at 28 deaths per 1,000 live births, followed by Central and Southern Asia with 25 deaths per 1,000 live births.
These findings suggest that focusing on the critical periods before and immediately following birth is essential to saving more newborn lives (8). NEC in Nigeria contributes to the commonest acquired and emergency surgical conditions that contributing for neonatal death of 11.8% (11).
Another study in Cairo, Egypt, showed that 16.4% neonates were diagnosed with NEC and resulted 38.6% neonatal deaths (12).
According to the 2016 Ethiopian Demographic Health Survey (EDHS) report, the neonatal mortality rate was 29/1000 live births, which had no significant reduction from the 2011 EDHS report which was 37/1000 live births(13). About 4 % neonates admiited in NICU with the case of 5 NEC and the fatality of NEC was 46.2%. It indicates that, it is a common fatal disease among neonatal morbidities (14). Identifying the risk factors and early initiation of therapy can significantly reduce the burden of neonatal death and illness related to NEC. The major causes of neonatal death in premature infants in Ethiopia are gestational age of 28-31 weeks,32-34 weeks and 35-36 weeks with NEC 2(22.2%),5(55.6%) and 2(22.2%), respectively (15). This shows that, NEC is one of a contributing factor for neonatal mortality in Ethiopia.
There is a limited information regards NEC in the study area, Ethiopia as a whole and Africa. In addition, there are some contradicting or inconsistent findings on some factors for NEC, like prematurity, low birth weight and enteral feedings around the world. It gives a little attention although it is a very fatal case of neonates. Therefore, this study was aimed to identify determinants of NEC in neonatal intensive care unit.

Study area and period
This study was conducted in East and West Gojjam zone referral hospitals, Amhara Region, Ethiopia, from February 24 to April 24, 2020. There are three referral hospitals in these zones (Felege Hiwot compressive referral hospital (FHCRH), Tibebe Gion Specialized teaching hospital (TGSTH) and Debre Markos referral hospital (DMRH)). FHCRH and TGSTH are found in Bahir Dar city, the town of Amhara regional state in the Northern direction of 565 Kilometer far from Addis Ababa (capital City of Ethiopia), whereas DMRH is found in Debre Markos, the town of East Gojjam Administrative Zone,299 Kilometers far from Addis Ababa. According to information obtained from administrative offices, FHCRH, TGSTH and DMRH have annual neonatal admission of more than 3500 (monthly 292), 1800 neonates and 2500 neonates, respectively.

Study design
Institutional based unmatched case control study design was conducted among neonates admitted in NICU.

Source population
The source population was all admitted neonates in NICU from February 24 to April 24, 2020 at referral hospitals of East and West Gojjam Zones of Amhara regional state, Northwest Ethiopia.

Study population
The study population was all randomly selected neonates from February 24 to April 24, 2020 who fulfill the inclusion criteria for cases and controls among admitted neonates in NICU.

Study units
The study units were mothers and neonates on which the data was collected during data collection period.

Inclusion criteria for cases
All neonates who were admitted in the NICU of the three referral hospitals with clinical signs, radiographic finding of NEC and who being exclusively diagnosed with NEC by the attending physician at the time of data collection were included in the study.

Inclusion criteria for controls
All neonates who were admitted with other health problems to NICU during data collection period were included as controls.

Exclusion criteria for cases
Neonates who were firstly diagnosed with NEC but recovered from NEC and developed other problems during data collection were excluded.

Exclusion criteria for controls
Neonates who were diagnosed with NEC and other health problems as the same time during data collection were not included as controls in the study.

Dependent variable
Necrotizing enterocolitis

Independent variables
Socio-demographic variables: Maternal age, sex of the neonate, age of the neonate, residence, religion, maternal marital status, maternal occupation and educational status.  Cases: neonates who presented with clinical signs and radiographic findings of NEC and who diagnosed by attending physician during data collection period.
Controls: neonates who were not diagnosed as NEC (do not fulfill clinical signs and radiographic findings of NEC) and who admitted with other health problems during data collection period.
Multiple gestations: a pregnancy with more than one fetus. It includes twins, triplets, quadruplets, or more(16).
Prolonged rupture of membrane: the time from membrane ruptured to delivery which is more than 18 hours(17).

Low birth weight: neonate's weight <2500grams at birth.
Enteral feedings: Feedings that given to neonate though the mouth or through a tube that goes directly to the stomach(18).
Formula milk: artificial milk (manufactured milk) which is prepared by mixing the powder milk with water.

Sample size determination
The sample size was calculated by using Epi-Info version 7 software by considering the proportion of mixed milk of cases (31%), the proportion of controls exposed (15%), ratio (1:2), 80% power , and 95% confidence level [36-38] . Then, it gives a total estimated sample size of 234, (78 cases and 156 controls). By adding a 5% non-response rate, a total of 246 (82 cases and 164 controls) participants were included in this study. 9

Sampling procedures and techniques
Proportionate allocation was made for each referral hospitals to select cases and controls.
Cases and two corresponding controls were selected randomly (lottery method) among those neonates admitted in the NICU of the three referral hospitals. One hundred eleven neonates (37 cases 74 controls), 57 neonates (19 cases 38 controls) and 78 neonates (26 cases 52 controls) were selected from FHCRH, TGTSH and DMRH, respectively.

Data Collection procedures and instruments
An interviewed structured questionnaire [36-38] was used to collect the data. Six BSc nurses collected the data. Medical cards and feeding charts of the case and control groups were used in addition to face to face interview. Mothers were interviewed about their socio-demographic characteristics and maternal factors where as neonatal and enteral feeding factors were collected from medical cards.

Data quality control
Training was given for data collectors and supervisors. A pretest was conducted in Addis Alem hospital which is found in Bahir Dar city by taking 5 % of the sample size before actual data collection to check consistency and any ambiguous of the language. A clear explanation about the purpose and objective of the study was provided for the respondents at the beginning of the interview. Close supervision was carried out by the principal investigator and supervisors during data collection period.

Data processing and analysis
The data was coded and entered into EPI data statistical software version 3.1, then exported to SPSS statistical software version 20 for data cleaning and analysis. Bivariable analysis was done to observe the candidate variables at p-value ≤0.25 and then multivariable logistic regression 10 analysis was employed to identify the independent variables that associated with NEC at p-value <0.05. Adjusted odds ratio (AOR) with 95% CI was used to show the strength of association between exposure and outcome variables. Frequency distribution tables and statistical graphs were used to describe some variables.

Socio-demographic characteristics
This study was intended to identify determinants of necrotizing enterocolitis among neonates admitted in NICU at referral hospitals of East and West Gojjam Zones. A total of 246 neonates (82 cases and 164 controls) who were admitted in NICU with their mothers were included with the overall response rate of 100%. According to this study, the median age of the mothers of cases and controls was 26.5 and 27 (25-34) years, respectively. More than half of the participants 47(57.3%) cases and 85(51.8%) controls were from urban areas. Fifty six (68.3%) cases and 86(52.4%) controls were males (Table-1).  Regarding to mode of delivery, majority of mothers 47(57.3%) of cases and three fourth 123(75.0%) of controls had spontaneous vaginal delivery (Figure-1). Zones of Amhara regional state, Northwest Ethiopia.

DISCUSSION
Neonatal enteroclitis is the major problems of neonatal deaths. Thus, identifying the determinant factors is crucial to enhance the health of neonates.
Duration of rupture of membrane (>18hours) was significantly associated with NEC. Neonates born from mothers with duration of rupture of membrane >18 hours were more likely to suffer from NEC as compared to those neonates born from mothers with duration of rupture of membrane ≤18 hours. This is consistent with studies conducted in USA (19) and Sweden (20).
The possible explanation might be due to birth canal is colonized with aerobic and anaerobic pathogens (from amniotic fluid) that might also cause colonization of the neonate's intestine at birth when duration of labor is greater than eighteen hours.
Birth weight was significantly associated with NEC. Neonates with low birth weight (<2500grams) were more likely to have NEC as compared to neonates with normal birth weight (≥2500grams). This finding is in line with findings of studies conducted in Sweden (20), Malaysia (21) and South Africa (22). This is due to the fact that low birth weight is developed due to prematurity and maternal complication like hypertension that present pre-conception or antepartum (2). Another neonatal variable in this study which was found to be significantly associated with NEC was neonatal sepsis. The present study revealed that neonates with sepsis were more likely to have NEC compared to those neonates without sepsis. This finding is also consistent with studies conducted in Sweden (20), USA (19) and China (23). The possible reason to this is; in the fact that neonate's immune system is not yet fully developed and they are susceptive to infection during invasive procedures in the NICU and also when a neonate has septicemia (sepsis), bacteria reproduce in the blood and produce toxins. The immature intestinal tract is exposed to these factors, resulting in a large number of cytokines that mediate an 22 inflammatory cascade. NEC is thought to be precipitated by an inflammatory cascade that causes cytokine release and damage to the immature intestine (1,2) In addition to these, multivariable logistic regression modeling demonstrated that enteral feeding was significantly associated with NEC. This study showed that specifically neonates who fed formula milk only were more likely to have NEC compared to those who fed breast milk only and neonates who fed mixed milk were also more likely to have NEC compared to those neonates fed breast milk only. Similar finding was reported from studies conducted in England (24), Scotland (25), Netherlands (26) and South Africa (22) also revealed that neonates who fed formula milk only were more likely to have NEC compared to those who fed breast milk only.
The possible reasons might be immaturity of bowel function of the neonates for digestion of formula milk (1). In addition to this, formula milk lacks immune protective factors that available in breast milk and it might contain food additives(5). Besides, feeding equipment that used to prepare the milk might always not be thoroughly cleaned before and after used.

Conclusion and recommendation
Duration of rupture of membrane (>18hours), low birth weight (<1500 grams), neonatal sepsis and enteral feedings (formula milk and mixed milk) were significantly associated with NEC.
Thus, using aseptic precautions for low birth weight, encouraging exclusive breastfeeding, and avoiding delay of labor are recommended in order to prevent neonatal enterocolitis. College of medicine and health sciences. Informed consent was obtained from all study participants (mothers) or parents for those have age less than 18 years-old. Written consent was obtained from participants after they were informed about the objectives, expected outcomes, benefits and the risks associated with it. Confidentiality and privacy of every respondent's information were ensured. All methods were performed in accordance with the relevant guidelines and regulations

Consent for publication
Not applicable

Availability of the data and materials
The datasets generated during and analyzed during the current study are not publicly available due to confidentiality of the participants' information but are available from the corresponding author on reasonable request.