Prevalence, risk factors and outcomes of neonatal dehydration among 13 hospitals in Kenya

Abstract

Background Information - Neonatal deaths currently contribute nearly 50% of under-five deaths and identification of causes of neonatal mortality is a priority. There are concerns that dehydration is a common problem in sick neonates, but few studies have documented the burden of dehydration, partly because of difficulty in making clinical diagnosis. In the neonatal period, dehydration has less pronounced clinical signs and this may lead to missed or mis-diagnosis leading to adverse outcomes and increasing cost of care. Study significance- This study described the outcomes in neonates admitted with dehydration thereby informing policy on the need for local clinical guidelines on diagnosis of dehydration in newborns. This study also sought to investigate neonatal and maternal risk factors for dehydration thereby highlight at risk patients that are most vulnerable to neonatal dehydration for early recognition and prompt treatment in efforts to reduce fatality. Study Objectives- The objectives of this study were to quantify the burden of neonatal dehydration among Kenyan hospitals, describe the outcomes of neonates presenting with dehydration, and to identify the risk factors of dehydration that can be obtained from routine hospital data. METHODS Study design – This was a cross-sectional analytic study. Study Area- This study was conducted across 13 county (first-referral level) hospitals in Kenya participating in a clinical information network (CIN). Data Collection Procedures - Data were abstracted from medical files of neonates admitted to pediatric wards of these 13 Kenyan hospitals within a 1year period (December 2015-November 2016). Newborns aged less than 28 days were identified from the CIN database and 65 patients randomly selected from each hospital from a list of unique identifiers. Medical notes for selected patients were reviewed by trained clerks located in each hospital and data were abstracted into a specially designed survey tool and synched to a central server. Data analysis – Newborns fulfilling a defined criterion for neonatal dehydration were identified from review of medical notes. Well babies and those admitted for accommodation were excluded. Inverse probability weighting was used to determine the prevalence of neonatal dehydration for the 13 hospitals. The adverse outcomes of neonatal dehydration were determined by calculating the proportion of neonates fulfilling case definition of neonatal dehydration who experienced undesirable outcomes upon discharge. The Risk factors for neonatal dehydration were identified using mixed effects logistic regression. RESULTS Study Population - A total of 810 neonates fulfilled criteria for inclusion. Dehydration Prevalence - The overall prevalence of neonatal dehydration was 16.7 % (132/810, 95% exact binomial CI [13.9 – 19.6%]) and it varied across study hospitals (range 7.8% to 26.6%). Dehydration presented alongside other conditions (comorbidities), the most common were sepsis, jaundice and pneumonia. Laboratory results for serum sodium and renal function tests for dehydration were rarely available even though tests were requested. Outcomes of Neonatal Dehydration – The major outcomes experienced by dehydrated patients (n=132) were; renal failure in 4(3%) of dehydrated newborns, neuro-disability/sequelae in 1 patient (0.8%) and mortality in 8(6%). A majority, 97(73.5%) were normal (in good health) upon discharge. Risk Factors of Neonatal Dehydration - This study identified neonatal age (Odds Ratio (OR) 0.5, 95% Confidence Interval (CI) [0.31-0.75]) and parity (OR 1.61, 95% CI [1.01-2.5])) as important risk factors for neonatal dehydration. There was minimal variation in outcome (dehydration) across study hospitals and the intra- class correlation coefficient was 1.7%. CONCLUSION The study findings indicate that neonatal dehydration is common in neonates admitted across hospitals in Kenya with proportions varying across hospitals. Main adverse outcomes experienced by dehydrated neonates in this study were kidney injury, neurologic sequelae and mortality. Age less than one week and being born to primiparous mother, were found to be significant risk factors of dehydration. Lactation support and training is recommended for primiparous mothers and mothers with neonates in their first week of life. RECCOMENDATIONS 1. Guidance for diagnosing dehydration that includes weight loss determination and laboratory examination should be included in the national paediatric guidelines. 2. Lactation support including practical training on proper breastfeeding techniques should be provided especially to first time mothers and more attention should be accorded to babies in their first week of life. 3. A study reviewing detailed medical records with improved documentation is needed.