Factors associated with poor clinical outcomes in children admitted with blackwater fever in eastern Uganda :

Abstract

Background: Severe malaria remains an important public health problem in sub-Saharan Africa. Blackwater fever, a complication of malaria has in the past been considered a rare complication of malaria in children living in high transmiSSion settings. More recently, however, a growing number of paediatrics case-series of Blackwater fever have been published from Africa. In Uganda, in particular eastern Uganda, clusters of Blackwater fever cases have been reported by local researcher, Olupot-Olupot, Engoru [34]. In this region B\VF still remains an important cause of hospital admissions with a high burden of mortality and morbidity causing worse outcomes among children admitted to hospital. Objective: This study seeks to determine the factors associated with poor clinical outcomes in children admitted with Blackwater fever in east em Uganda. Method: A retrospective quantitative review of hospital-based records using patient case files was done. The admission registers in the two tertiary hospitals were used to identify all the children who presented to the Paediatric Acute Care Unit (P ACU) of MRRH and SRRH within the period of study (2018) with a diagnosis of Blackwater fever/dark urine syndrome. Their case records were retrieved and the necessary information were obtained using a structured questionnaire. These included their demographic, clinical characteristic and outcome status. A diagnosis of BWF/dark urine syndrome was made at or during admission with the aid of the Hillmen Colour Chart (HCC). Bivariate and multivariate analysis using logistic regression was used to analyse the association between the predictors and poor clinical outcomes (prolonged hospitalisation and mortality) of admission with BWF. Results: Of the 9578, 1241 (13.0%) children were admitted with BWFIDUS to the Paediatric Acute Care Umt (PACU) of Mbale and Soron regional referral hospital in the year 2018 The age of the study participants ranged from 4 months to 180 months with a median age of 60 months; interquartile range (IQR) 36 - 90) and a male preponderance (1.5" I male to female ratio). 559 (45.04%) <5 years and 682 (54.96%) were ~5 years Besides all patients presenting with passing dark urine /tea-coloured unne, most of the patients also commonly presented with high fever 1109 (94.30%), vomiting 599 (53.01%) and abdominal pam 494 (45.11%). In addition, they presented with common clinical signs of pallor 742 (67.33%), clinical jaundice 369 (3442%), prostration 231 (21.94%) and abdominal signs of abdominal tenderness 120 (9.67%) and splenomegalyl22 (9.83%). Presumed sepsis and sickle cell disease were the leading co-morbidity. 426/1241 (34.3%) of the patients received at least one blood transfusion.There was significant difference in the clinical characteristics observed between <5 years and ~5years and to a small extent between the males and females 251 (27 3%) patients had prolonged hospitalization, which was defined as a baselme hospitalization stay longer than 5 days (>75 percentile of hospitalization duration) aud 40 11241patients (3.22%) had mortality during the study period. Multivariate logistic regression analysts indicated that the independent predictors for prolonged hospuahzation were abdominal pam (aOR 1.91, 95% CI 1.04 -3.49; p=0.037), the presence of "any pain" (either of chest pain, hand pain or foot pain) (aOR 2.19; 95% CI 1.41 - 3.39; p<O.OOO) and delayed capillary refill time of>3 seconds (aOR 1.84; 95% CI 0.93 - 3.67; p=0.082). Conclusion: This study affirms that the independent predictors for prolonged hospitalizanon are abdominal pain, the presence of "any pain" (either of chest pain, hand pain or foot pain) and delayed capillary refill time of >3 seconds. Therefore, recognition of these predictors warrants more vigilance by clinicians to improve clinical examination during the tnage of patients with BWF/dark urine syndrome to avert these unfavourable outcomes in children admitted with BWF/dark urine syndrome Keywords: Blackwater fever, dark urine syndrome, severe malaria, predictors, prolonged hospitalization, Mortality.