| dc.description.abstract | Background
Mortality in African patients initiating ART is high in the first year of treatment, which contributes to significant ART program losses. The causes and factors associated with death are not well understood. Although cohort studies have reported on cause specific mortality in patients on ART, they have not had access to patients during hospitalization. Hospitalized patients represent the burden and spectrum of severe morbidity and mortality because when patients on ART get severe morbidities, they are likely to be admitted and some may die.
Objectives: The objective of this study was to determine inpatient diagnoses, outcomes and factors associated with mortality in HIV infected patients on ART admitted to Mulago
hospital.
Methods: Prospective case series of HIV positive patients admitted to the medical wards of Mulago hospital while receiving antiretroviral therapy for at least two weeks. Consecutive patients who fulfilled the inclusion criteria were recruited until a sample size of 201 patients A pre-tested questionnaire was administered to the study participants or their attendants at the beginning of hospitalization. Participants underwent a full clinical examination and approximately 5mls of blood was drawn for laboratory tests including: a full blood count, renal function tests, liver function tests and CD4 counts. Additional tests to aid in making the diagnosis were done at the discretion of the attending clinician. Patients were followed up in the hospital until discharge, death or for thirty days if hospitalization exceeded thirty days. Data was coded using study numbers and all study information was kept confidential. Data entry was done using Epi-data version 3.1 and then exported to SPSS version 16.0 for analysis. Continuous variables such as age, body mass index (BMI), etc were summarized using medians, and inter-quartile ranges. Categorical variables e.g. inpatient diagnoses, patients’ WHO clinical stage, ART regimens etc were summarized as frequencies and percentages. Bivariate and multivariate analysis using the Chi- square test and binary logistic regression respectively were done to determine factors predictive of death such as low BMI and Karnofsky performance status scale (KPSS) at admission.
Results
Tuberculosis (18%), cryptococcal meningitis (11%), Zidovudine (AZT) associated anemia
(10%), sepsis (5%), and Kaposi’s sarcoma (5%) constituted the most inpatient diagnoses. Twenty four (12%) diagnoses were not established. Mortality rate was 21%. Tuberculosis (24%), cryptococcal meningitis (19%), sepsis (12%) and undiagnosed neurological syndromes (21%) were significant causes of mortality. Fifty two percent of the deaths occurred within 12 weeks of ART initiation. A low baseline Karnofsky performance score of ≤ 40 (OR, 21.1; CI 1.43- 31.6), body mass index (BMI) < 18 (OR, 0.2; CI 0.42- 0.97) and age >34 years (OR, 7.65; CI 1.09- 53.8) were significantly predictive of mortality.
Conclusions and recommendations
These findings show that opportunistic infections and AZT associated anemia caused
significant morbidity and high mortality in patients on ART hospitalized in Mulago hospital. This highlights the importance of intensified routine screening for tuberculosis and cryptococcal meningitis in patients so as to treat or provide primary prophylaxis for these infections when starting ART. Patients on AZT should be closely monitored. The three predictors of death are easy to assess and can be used to triage patients in resource limited
settings. Studies are needed to explore the effect of nutritional rehabilitation on survival. | en_US |
