Prevalence and Risk Factors of Hypothermia at Admission to New Born Unit in Kenyatta National Hospital

Abstract

Background: World health organization (WHO) defines neonatal hypothermia as axillary temperature <36.5⁰C. Neonatal hypothermia is a common problem that coexists with other health problems like severe neonatal sepsis, asphyxia, and prematurity. Neonatal hypothermia leads to decreased cardiac output, decreased cerebral blood flow and has been associated with increased risk of neonatal mortality and morbidity in low and middle-income countries. Neonatal hypothermia is preventable if measures are put in place to identify and bridge the gaps in the warm chain that is used for thermoprotection of the newborn. Objective: To determine the prevalence and risk factors of hypothermia at admission to the newborn unit in KNH. We also sought to determine the neonatal outcome(mortality) of hypothermic neonates at 28 days of life and to assess the capacity to prevent neonatal hypothermia under the warm chain in KNH Methodology: A Cross-sectional study conducted in the newborn unit, labour ward and maternity theatre in KNH. All neonates being admitted to NBU-KNH were recruited and their axillary temperature taken. Consecutive sampling was done until the sample size was achieved. Risk factors were abstracted from the KNH NBU transfer form to the newborn data abstraction tool. The newborns were followed up for 28 days to determine their outcome (mortality or discharge). A warm chain assessment checklist was used to assess the maternity theatres, labour ward and NBU admission room. This was done on the same day of the week for the 2 months that the study was ongoing. Data management and analysis: Data was entered and checked for errors daily in Microsoft excel sheet. Cleaned data was analysed by STATA version 13. Clinical characteristics that were continuous (birth weight, temperature) were analysed and presented using means and standard deviation, as well as medians. Categorical data (gender) was analysed and presented as

frequencies and proportions. Tables and graphs were used to present the results. Statistical significance was considered when the p value <0.05 Results: The prevalence of neonatal hypothermia was 67.2%. The mean birth weight was 2400g (±998.0). More than half of the neonates (54.1%) were males. About 83.2% of all the neonates admitted were less than 1-day old at the time of admission. The mean gestational age was 35.3 weeks (±4.39). Most of the newborns (76.2%) were inborn (newborns born in KNH) and 70.9% of the admitted newborns had received resuscitation after birth. Maternal factors were not associated with neonatal hypothermia. In multivariable logistic regression, age <1day (p=0.002), gestation<37weeks(p=0.001) and very low birth weight(p=0.003) were found to be significant as risk factors for neonatal hypothermia at admission. Newborns admitted with hypothermia were 2 times more likely to die (AOR 2.34, 1.15-4.76) p=0.011. The NBU admission room had better compliance (88%) compared to labour ward and maternity theatre (50% and 60% respectively) Conclusion: The prevalence of neonatal hypothermia was 67.2%. Gestational age <37 weeks, age<1 day old and very low birth weight were independent factors associated with hypothermia. Therefore, attention is needed for thermal care of preterm newborn and newborns who require admission on their 1st day of life. There should be enhancement of thermal protection principles of warm chain.