Neonatal Umbilical Cord Infections: Incidence, Associated Factors and Cord Care Practices by Nursing Mothers at a Tertiary Hospital in Western Uganda

Background: Umbilical cord infections complicate to neonatal sepsis that significantly contributes to neonatal mortality worldwide. There is paucity of data on the incidence, factors associated with neonatal umbilical cord infections in western Uganda, yet nursing mothers continue to practice potentially dangerous cord care practices. We described the incidence, factors associated with umbilical cord infections and cord care practices by nursing mothers at a tertiary hospital in western Uganda. Methods: This was a hospital based cross sectional study at a tertiary hospital in western Uganda between the months of March and June 2019. Two hundred and forty (240) neonates aged between 2 to 14 days attending the neonatal unit were recruited in the study. Umbilical cord infection was assessed based on the World Health Organisation (WHO) clinical definition: finding of any of discharge, redness and swelling on the umbilical stump. Data on sociodemographic characteristics, maternal and neonatal perinatal factors and cord care practices of the nursing mothers were analysed by bivariate and multivariate logistic regression using STATA 13.0 to determine factors associated with umbilical cord infection. Results: Sixty-five, 65 (27.1%) neonates had at least one sign of cord infection. Majority of the dence of neonatal umbilical cord infection is high in this setting. Application of a dangerous substance to the umbilical stump and maternal secondary education level were significantly associated with umbilical cord infection. Clinicians should routinely examine the umbilical stump of every neonate for the signs of infection and encourage mothers to avoid application of dangerous substances to the cord. Maternal education on the recommended cord care practices during antenatal and postnatal period could reduce the incidence of umbilical cord infection among neonates in low resource setting.


Background
Umbilical cord infections have for centuries caused deaths in neonates. The incidence of cord infection in neonates increased in the 1940s after the introduction of nurseries for newborns in hospitals [1]. These infections now contribute the highest risk of death due to neonatal sepsis, compared to all other causes of neonatal sepsis [2] [3] [4] [5]. Umbilical cord infection is the infection of the umbilicus and/or surrounding tissues, occurring primarily in the neonatal period. It is a true medical emergency that can rapidly progress to neonatal sepsis and death, with an estimated mortality rate between 7% and 15% [6].
According to World Health Organization (WHO), umbilical cord infection is the infection of the neonatal umbilical cord stump clinically diagnosed by the presence of any of umbilical discharge, erythema or swelling [7]. Multiple complications can occur from bacterial colonization and infection of the umbilical cord because of its direct access to the bloodstream.
Umbilical cord infection risk is 62% higher in neonates receiving topical cord applications of potentially unclean substances [4] [8]. Other common risk factors for the development of umbilical cord infection include unplanned home birth or septic delivery, low birth weight, prolonged rupture of membranes, umbilical catheterization, chorioamnionitis [9], lack of knowledge on cord care, twin's delivery [10] and prematurity [11] [12].
Chlorhexidine cord cleansing and dry cord care significantly reduces neonatal mortality related to umbilical cord infection [9] [13] [14] [15]. The incidence of umbilical cord infection reported in different communities varies greatly, depending on prenatal and perinatal practices, cultural variations in cord care, and delivery venue (home versus hospital) [9].
In developing countries, where most deliveries take place at home, some traditional practices such as applying unclean substances to the cord stump, conflict with principles of cleanliness and greatly increase the risk of cord infection and tetanus [9] [16]. These substances have included plant extracts, raisins, coins, cow dung, colostrum, and, more recently, antibiotics in the form of pow-  [18]. There is paucity of data on the incidence, factors associated with neonatal umbilical cord infections in western Uganda, yet nursing mothers continue to practice potentially dangerous cord care practices. This study described the incidence, factors associated with umbilical cord infections and cord care practices by nursing mothers at a tertiary hospital in western Uganda.

Study Design and Setting
This was a hospital based descriptive analytical cross-sectional study that determined the incidence, and described associated factors of umbilical cord infections and cord care practices by nursing mothers among neonates in the neonatal section of Kampala International University Teaching Hospital (KIU-TH), western Uganda.
The KIU-TH neonatal section has three divisions; the special care baby unit (SCBU) where preterm and very sick neonates are admitted, the baby nursery at maternity ward where immediate newborns are monitored for 48 -72 hours with their mothers before discharge and the neonatal stable room where sick neonates with non life-threatening conditions are admitted for treatment. The neonatal unit provides neonatal care services including neonatal resuscitation, nasogastric feeding, intravenous fluid therapy and management of neonatal infections, prematurity, birth asphyxia and neonatal jaundice.
The section is well staffed with 3 pediatricians, 4 senior house officers (SHOs), 2 junior house officers (JHOs) and 2 general nursing officers. The doctors work on scheduled daily programs to ensure there is an attending doctor at all times. Nurses also work in shifts and are regularly trained through continuous medical education (CME) on neonatal life support and management of common newborn conditions.
The unit is equipped with ambubags, face masks, a radiant warmer, room heater, 5 incubators, 2 infusion pumps, 3 phototherapy machines, 4 oxygen concentrators and provides improvised nasal-bubble continuous positive airway pressure (CPAP)to premature neonates.

Inclusion and Exclusion Criteria
Two hundred and forty (240) late preterm and term neonates born between 34 and 42 weeks of gestation were consented and enrolled consecutively over a period of 4 months from March to June 2019. The study included stable neonates at the maternity nursery and those admitted at the neonatal unit and excluded early and extreme preterm neonates born below 34 weeks of gestation. This is because these neonates are under the care of the nurse who as well cares for the cord and also on the routine prophylactic antibiotics.

Participants' Characteristics
Socio-demographic, perinatal characteristics of the nursing mother and neonate, Open Journal of Pediatrics and cord care practices by the nursing mother were captured using a structured questionnaire designed by the principal investigator and pretested for reliability and validity of the data. Pre-testing of the questionnaires was done at the Neonatal Ward at Mbarara Regional Referral Hospital (MRRH) in the same geographical area, located in Mbarara Municipality, using ten (10) questionnaires. Clearance was sought from the MRRH administration before pre-testing the research questionnaires on the patients. Interviews and examination of the participants was done by the investigator(s) and this could compromise inter-rater reliability bias. However, a second opinion was always thought from the attending clinician for neonates who were found to have cord infection before being put on the treatment list for cord infection or sepsis. The questionnaires were translated into local language (Runyankole) and to ensure that the translations are correct, the questionnaires were back translated into the English and compared to assess if meanings have changed.

Data Collection
The study included newborns in the neonatal nursery and neonatal unit. Sick neonates first received emergency treatment as per treatment protocols. Written informed consent from the mother/father of the neonate was obtained; a study number and inpatient number was assigned to each neonate. The investigators conducted a face-to-face interview with mother/care-giver, and information regarding socio-demographic data of the neonate, perinatal factors and cord care practices documented using a structured questionnaire.
A full neonatal head to toe examination was done on every neonate with emphasis on the abdominal examination. Weight of every neonate was taken using a digital weighing scale and recorded in the respective questionnaire. Comprehensive neonatal physical examination of every neonate was done with focus on the umbilical stump for clinical signs of umbilical cord infection. Umbilical cord infection was defined as presence of any of purulent discharge, redness and/or swelling/edema around the umbilicus, according to the World Health Organization/Integrated Management of Childhood Illness (WHO/IMCI) guidelines [7].

Statistical Analysis
Data from completed questionnaires was arranged, summarized and entered using the statistical computer software package, Microsoft Excel 2016. The data was cleaned, checked for errors and corrected, then imported to STATA version 13.0 (Statacorp, College station, USA) for analysis.
Incidence and cord care practices by nursing mothers were analyzed and presented as frequencies and percentages. Factors associated with umbilical cord infection among neonates were analyzed using bivariate and multivariate logistic regression. Factors with p-value ≤ 0.2 at bivariate analysis were considered for multivariate analysis. Measures of effect were reported using odds ratio for both crude and adjusted analysis, followed by corresponding 95% CI and p-value. At multivariate analysis, factors with p-value ≤ 0.05 were considered statistically significant. Open Journal of Pediatrics

Study Limitations and Delimitations
Some mothers may have hidden information related to dangerous substances applied to the cord since it is associated with social desirability bias. However, participants were encouraged and utmost confidentiality provided so that the required information was given to achieve the research objectives.
This study was hospital based and could not explore deeper into the community perspectives about cord care practices. Interviews and examination of the participants was done by the investigator(s) and this could compromise inter-rater reliability bias. However, a second opinion was always thought from the attending clinician for neonates who were found to have cord infection before being put on the treatment list for cord infection or sepsis.

Umbilical Cord Care Practices by Mothers
One hundred seventy four, 174 (72.5%) nursing mothers had not received education by a health worker on proper cord care practices during antenatal and postnatal period, and did umbilical cord care based on advice from the attending grandmother and/or relative. Majority of the mothers 168 (70%) do not use the recommended umbilical cord care practices. Among these, 73 (30.4%) do not cleanse the cord while 95 (39.6%) cleanse with application of a dangerous substance as shown in Figure 1. The commonly applied dangerous substances in cord care included: saliva, cow ghee, cow dung, herbs (ereka, mushroom, grass, kiyondo, pumpkin flower, onion leaves), ash (papyrus, matchbox), soot (saucepan, rubber), rotten yam, soil and lizard faeces in order of reducing frequency. The main reason for applying a dangerous substance was to hasten healing and cord separation.

Incidence of Neonatal Umbilical Cord Infections
Sixty-five (27.1%) of the 240 neonates had at least one sign of infection on the cord. Sixty (92.3%) of these were aged below 7 days of life and mostly male, with no statistical significance. This is shown in Table 3.

Factors Associated with Umbilical Cord Infection
Neonates of mothers aged 20 -24 years (uOR = 6. . This is shown in Table 4 and Table 5.

Discussion
Umbilical cord infection is preventable and can be reduced by practicing clean delivery and clean cord care, by avoiding harmful practices, and by increasing tetanus toxoid immunization coverage [18]. In this study, the incidence of umbilical cord infection among neonates is high at 27.1%. The finding is similar to other studies done in least developed countries. Studies in Uganda at China Uganda Friendship Hospital Naguru 24% [19] and Mulago National Referral Hospital 17.2% [20], Pumwani Maternity Hospital in Kenya 37.6% [21], Pemba Island Tanzania 20% [4] and Pakistan 21.7% [22] found a high incidence of neonatal umbilical cord infection. Surprisingly, a community study in Nepal [8] found neonatal umbilical cord infection rate of 5.5% significantly lower than the incidence from similar least developed countries, indicating that cultural practices influence the incidence compared to the economic disparities. The incidence is however several times higher compared to the few studies done in middle income and developed countries. An epidemiological study in eastern Turkey [23] put the incidence of umbilical cord infection at 11.8% while two prospective observational studies at hospitals in Oman [12] and India [5] found the incidence was even lower at 1.8% and 2.3% respectively.
The World Health Organisation [24] strongly recommends daily chlorhexidine (4%) application to the umbilical cord stump during the first week of life for newborns who are born at home in settings with high neonatal mortality (neonatal mortality rate > 30 per 1000 live births) and clean, dry cord care for newborns born in health facilities, and at home in low neonatal mortality settings. Use of chlorhexidine in these situations may be considered only to replace application of a harmful traditional substance such as cow dung to the cord stump [24]. This study was done on neonates born in health facilities and/or at home in low mortality setting [25].
Majority of the study participants in this study were from a rural low resource setting with less information on the recommended cord care practices compared to the participants from studies in developed settings. This has great effect on the care of the umbilical cord by mothers. About 39.6% of the mothers in this study applied a dangerous substance mostly saliva and ash to the cord of the neonate to hasten healing and/or separation; this was significantly associated with umbilical cord infection.
The odds of having umbilical cord infection were 3 times higher in neonates whose mothers applied a dangerous substance to the cord compared to neonates receiving no cord application (aOR = 3, p = 0.006). Dangerous substances like saliva, ash, soil, cowdung contain pathogenic bacteria which when allowed access to the nutrient rich umbilical cord allows multiplication of the bacteria to cause umbilical cord infection.
Application of potentially dangerous substances in cord care has for long been associated with umbilical cord infection in studies done from different settings [3]

Conclusions
The Community sensitization by health care groups and leaders on the dangerous cultural cord care practices like the application of dangerous substances to the umbilical cord should be further promoted.
Maternal child health policy makers in least developed countries should promote application of chlorhexidine gel on the neonatal cord to replace the ues of dangerous substances. This is possible if every delivery kit supplied to the hospitals contained the chlorhexidine gel.

Declarations
Ethical approval and consent to participate: Ethical approval was sought from the research ethics committee (REC) of Kampala International University Western Campus, under reference number UG-REC-023/201901. Informed consent was sought from each mother and the purpose of the study well explained to the participants. The consent forms were in both English and local language, and the participants had the right to decline to participate or withdraw from the study at any time if they so wished.