556 
Predictors of Early Mortality in Haitian Patients Treated with ART in a Community Setting
Rebecca Dillingham*1, R Pinkerton1, P Leger2, P Severe2, J Pape2,3, and D Fitzgerald2,3
1Univ of Virginia Hlth System, Charlottesville, US; 2GHESKIO Ctrs, Port au Prince, Haiti; and 3Weill Med Coll, Cornell Univ, New York, NY, US
Background: ART
became available at no cost to patients in Haiti in 2003. In initial studies,
ART improved 1-year survival for Haitian patients with AIDS from 30% in the
pre-ART era to almost 90% on ART. Most deaths on ART occurred in the first 6
months of therapy. Identifying simple clinical characteristics that predict
early mortality and that are available in almost all resource poor countries
may allow for interventions to improve survival at one-year.
Methods: This
study is a retrospective analysis of the survival of a cohort of 622 adult
Haitian patients who initiated HAART at the GHESKIO Centers between March 2003
and June 2005.
Results: We
followed 622 patients for 4308 patient-months. The mean age at study enrollment
was 40 years (IQR 33 to 46 years); 53.7% of patients were female. The mean weight
for females was 112.5 lb (IQR 98 to 125), while for
males it was 126.1 lb (IQR 110 to 140). The mean CD4 was 129 (IQR 36 to 199).
The mean hemoglobin was 10.5 (IQR 9 to 12). Common presenting complaints included
diarrhea in 279 (45%) patients, and pulmonary tuberculosis in 44 (7%); 69
(11.1%) patients died, 51(74%) within the first 6 months. By survival analysis,
the 1-year survival was 86%. Statistically significant univariate
hazard ratios for mortality included: the
presence of diarrhea at the initiation of ART (1.9; 95%CI 1.2 to 3.1, p = 0.008); the presence of wasting,
defined as weighing less than the 25th percentile for gender (3.1;
95%CI 1.9 to 5.0, p = 0.0001);
hemoglobin <9.5 (2.5; 95%CI 1.5 to 3.9, p
= 0.0001), and CD4 count <200 (2.2; 95%CI 1.1 to 4.3, p = 0.023). Sex, age, and TB at
presentation were not significant (p >0.05)
predictors of survival. Multivariate Cox regression analyses show that wasting
(OR = 2.4; 95%CI 1.4 to 3.9, p = 0.001)
and low hemoglobin (OR = 2.1; 95%CI 1.3 to 3.5, p = 0.003) were the only independent risk factors for death in the
first year. Hemoglobin level and wasting account for the univariate effects of CD4 level and the presence of
diarrhea at therapy initiation.
Conclusions: Interventions that provide early intensive
nutritional rehabilitation, micronutrient supplementation, and treatment of
anemia may improve survival of patients started on ART in resource-limited
settings. Prospective studies to further characterize these risk factors and to
plan interventions are warranted.
|
