Effects of Fluid Balance on Outcomes Among Children in the Intensive Care Unit at Kenyatta National Hospital

Abstract

Background For critically ill pediatric patients, fluid therapy plays a pivotal role in resuscitation. Adequate intravascular volume re-establishment using timely fluid management may be crucial in saving lives. Nevertheless, pediatric patients also receive varying amounts of fluid beyond resuscitation e.g., via medication, food, or routine fluid administration. The total fluid intake can exceed fluid output leading to a positive water balance. Pediatric practitioners experience challenges regarding the kind of fluid, and volume of intake, besides the timing of intravenous fluid administration. Fluid buildup following early resuscitation might worsen the condition and increase the risk of death, hence it is imperative to monitor fluid status and ensure close assessment of very ill pediatric patients. Broad Objective To determine the relationship between fluid balance and clinical outcomes among critically ill pediatric patients admitted to the Pediatric Intensive Care Unit/Main Intensive Care Unit (PICU/main ICU), Kenyatta National Hospital. Study design and site This was a prospective observational study that was conducted at Kenyatta National Hospital. It was carried out in adult and paediatric ICUs. Participants and methods Study participants were children aged 1 month to 12 years admitted to the two ICUs. Consecutive sampling method was used to recruit a total of 81 children. A standardized questionnaire was used to collect patient data from admission throughout the PICU/ICU stay. Data management The data that was collected included continuous, categorical, and discrete data. Data entry was done using excel spreadsheets and then exported to R software. Data was cleaned, coded, and analyzed using R Studio version 4.0.2. The first analysis included descriptive statistics i.e., median age and weight with their interquartile ranges, frequencies, and proportions for categorical data such as gender and fluid overload. Binary logistic regression was used to determine the association between the independent variables e.g., gender, weight, age xiv and dependent variables e.g., mortality. On the other hand, Poisson regression was used to determine the relationship between the factors associated with length of ICU stay and days on mechanical ventilation. The results were evaluated at 5% significance level. Tests were interpreted using odds ratios and P-values. Results This study sampled a total of 81participants. The prevalence of fluid overload was 74% (95% CI 62%, 83%). There was a mortality rate of 51.0%. Mortality in those with fluid overload was 47.1% (95% CI 33.2%, 61.4%) while mortality in those without fluid overload was 38.9% (18.3%, 63.9%). There were no factors associated with mortality, fluid overload OR 1.28 (95% CI 0.36, 4.68), p-value 0.70, inotropic use OR 0.84 (95% CI 0.27, 2.69), p-value 0.78. Inotropic use and mechanical ventilation were significantly associated with the length of ICU stay p-values 0.004 and <0.001 respectively. Conclusion Mortality and fluid overload in this study were above fifty percent. This study did not find any significant association between ICU mortality and the factors that were investigated among the study participants. Inotropic use and mechanical ventilation were associated with the length of ICU stay. There is need for more studies with larger sample sizes to further investigate the factors studied under this study.