Factors associated with mortality among medical patients admitted to Mulago Hospital with severe sepsis
Abstract
Severe sepsis is a significant cause of morbidity and mortality among hospitalized patients worldwide. Globally, mortality ranges from 30%-70% and in Mulago hospital,
mortality is 42%. The high incidence is attributed to the high infectious disease burden
including HIV infection. Diagnosis is still a challenge and relies on clinical and laboratory parameters many of which are unavailable in many health care facilities in Uganda. Though a number of prognostic factors have been shown to predict outcome of severe sepsis, many laboratory parameters are not routinely done in a number of health facilities in Uganda. This study was conducted to determine the impact of prior duration
of illness, prior antibiotic use and type of causative microorganism on mortality of severe
sepsis in Mulago hospital that could be easy distinguishable outcome predictors of severe sepsis and hence aid prompt intervention.
Objective: The main objective of this study was to determine the early and late mortality
rates and associated factors of medical patients admitted to Mulago Hospital with severe sepsis.
Methods: This was a prospective cohort study conducted on patients ≥18 years admitted
to Mulago Hospital Accident and Emergency Unit. Consenting patients who fulfilled the
inclusion criteria for severe sepsis were recruited consecutively until a sample size of 235 patients was achieved. A pre-tested questionnaire was administered to the study patients or their attendants. Approximately 20mls of blood was drawn for laboratory tests including: serum lactate, CBC, HIV test, CD4 cell counts, thick malaria smear, and aerobic culture and sensitivity.
Study patients were followed up both in hospital and at home for a total of 30 days or
until death (if death occurred before the end of 30 days from recruitment). Individual patients were identified using study numbers. Study tools were kept confidential. Data entry was done using Epi data version 3.1 and thereafter exported to STATA version 10.0 for analysis. Continuous variables were summarized as means and standard deviations and categorical variables as frequencies and percentages. For associated factors, bivariate analysis was done and factors found significant were assessed for interaction at multivariate level. The common bacterial isolates were S.pneumoniae, non-typhoidal Salmonella and E.coli while Cryptococcus neoformans was the only isolated fungus. The bacterial isolates were sensitive to ceftriaxone and erythromycin and resistant to cotrimoxazole and penicillins.
Conclusion: The overall in and out patient 30-day mortality of patients with severe sepsis
was 37%. Factors that were found significantly associated with mortality were; diarrhea
and a low serum albumin. The common bacterial isolates included; S.pneumoniae, nontyphoidal Salmonella and E.coli which were all sensitive to ceftriaxone but resistant to
cotrimoxazole and penicillins. Cryptococcus neoformans was the only isolated fungus.