Encephalopathic syndrome of trypanosoma brucei rhodesiense: Incidence, risk factors and predictive cerebrospinal fluid leukocyte count at Lwala Hospital, Uganda
Background: Encephalopathic syndrome (ES) of HAT is defined as occurrence of CNS signs and symptoms in a patient previously without them (or development of new ones, or exacerbation of existing ones), after initiation of treatment for late stage human African trypanosomiasis. It has been studied in West African form of the disease where it has been found to be associated with increased cerebrospinal fluid leukocyte count ≥100 per ml. However, not much is known about the risk factors and the predictive cerebrospinal fluid levels in the East African form. Objective: To determine incidence, risk factors, and predictive cerebrospinal fluid leukocyte count for encephalopathic syndrome of trypanosoma brucei rhodesiense in patients treated at Lwala Hospital, Uganda. Methods: Retrospective cohort, based on records of patients treated for late stage Trypanosoma brucei rhodesiense at Lwala Hospital in Eastern Uganda from 1st January 2004 to 31st December 2012. A structured data extraction form was used to obtain information from patient records. Logistic Regression was used to assess the risk factors for ES, while a ROC curve was plotted to determine the predictive CSF leukocyte count. Results: Incidence of ES was 16.1% (95% CI=2.1 – 20.2). Suramin pretreatment (before the melarsoprol course) was found to be protective of ES, (OR= 0.51, 95% CI 0.28 – 0.96) while treatment from August to October increased the risk, (OR=3.83, 95% CI 1.43 – 10.22). Weight, admission state and the HIV sero-status were found to be confounders. Weight (OR=0.98, CI 0.96 – 0.99) and a positive HIV serology (OR=1.39, CI 1.08 – 1.80) increased the risk of ES while a fair admission state (OR=0.36, CI 0.18 – 0.72) reduced the risk. CSF leukocyte count was not associated with ES and was also found to be neither accurate nor reliable in predicting its occurrence. Area under ROC curve was 0.504 and point nearest to the top left corner corresponded to sensitivity and specificity of 28% and 80% respectively and cut off at 36 leukocytes per ml. Conclusion: The incidence of ES in Lwala is about four times higher than that observed in other foci of both T.b gambiense and T.b rhodesiense. Suramin pre-treatment protected patients from ES and reduced its risk by 50% while treatment between August and October increased the risk fourfold. Recommendation Early detection of cases before progression to late stage is recommended since the risk factors cannot be modified.