Article Outline
I. Summary
II. 1. Introduction
III. 2. Methods
A. 2.1. Study subjects
B. 2.2. Statistical analysis
IV. 3. Results
V. 4. Discussion
VI. Acknowledgements
VII. Appendix A. Supplementary data
VIII. References
IX. Copyright
Summary
Objective
To retrospectively investigate the outcomes of patients with AIDS-associated Kaposi sarcoma (AIDS-KS) after initiation of antiretroviral therapy (ART), under routine practice conditions, at a university-affiliated hospital in urban Zimbabwe.
Background
While studies from developed nations have demonstrated excellent outcomes for AIDS-KS patients treated with ART, few studies have examined the outcomes of African AIDS-KS patients after starting therapy.
Methods
A retrospective cohort of 124 AIDS patients initiating ART under routine practice conditions was studied. Thirty-one patients with AIDS-KS were matched 1:3 to 93 AIDS patients without KS (non-KS). The primary outcome was loss-to-care after initiation of therapy. Multivariate analysis was performed to identify significant predictors of loss-to-care. The percent change in weight at 6 months after starting ART was a secondary outcome. A sub-group analysis evaluated differences in pre-treatment variables between AIDS-KS patients retained-in-care compared to those lost-to-care.
Results
AIDS-KS patients had significantly greater 2-year proportional loss-to-care than matched non-KS AIDS patients (26.4% vs. 9.5%; p = 0.01) after initiation of ART. In multivariate analysis, the presence of KS (p = 0.02) was the only significant predictor of loss-to-care. AIDS-KS patients had significantly less weight gain after starting ART than non-KS AIDS patients (+3.4% vs. +6.4%; p = 0.03). AIDS-KS patients retained-in-care had significantly higher pre-treatment CD4+ lymphocyte counts than AIDS-KS patients lost-to-care (223 vs. 110 cells/mm3; p = 0.04).
Conclusions
In this retrospective study, AIDS-KS patients experienced significantly worse outcomes than matched non-KS AIDS patients after initiation of ART. AIDS-KS patients with higher pre-treatment CD4+ lymphocyte counts were more likely to be retained in care.
Keywords: Kaposi sarcoma, HIV, AIDS, Antiretroviral therapy
1. Introduction
AIDS-associated Kaposi sarcoma (AIDS-KS) results from co-infection with HIV-1 and Kaposi sarcoma-associated herpes virus (KSHV) and it is the most common AIDS-related malignancy worldwide.1, 2, 3, 4 The high prevalence of HIV-1/KSHV co-infection in Sub-Saharan Africa places 30–50% of HIV-1-infected individuals from this region at risk for development of AIDS-KS.5, 6, 7, 8, 9, 10 The introduction of multi-drug antiretroviral therapy (ART) regimens in developed countries has been associated with a significant decrease in the incidence of AIDS-KS11 as well as improved patient survival. Studies from North American and European cohorts have demonstrated significant tumor regression and 5-year AIDS-KS survival rates near 90% after initiation of ART in early-stage disease.12, 13, 14, 15 However, similar outcome data for AIDS-KS patients from Sub-Saharan African populations, where AIDS-KS remains a significant cause of morbidity and mortality resulting in 10–14% of deaths in ART programs,16, 17 are lacking. Several recent ART studies in African AIDS-KS patients have not shown the same improvements in outcomes as those reported in studies from developed nations.18, 19, 20
Due to co-morbid conditions, limited access to antiretroviral medications, initiation of treatment at later disease stage, and, potentially, different host and viral genetic factors, the response of AIDS-KS patients to ART in Sub-Saharan Africa may be different than that reported in developed countries. In order to improve the treatment of AIDS-KS in African settings, it is necessary to better understand if and how AIDS-KS may lead to worse ART outcomes. The objective of the present study was to improve that understanding by investigating the outcomes of AIDS-KS patients after initiation of ART at a university-affiliated hospital in urban Zimbabwe, as well as the factors associated with those outcomes.
2. Methods
2.1. Study subjects
This study followed the US Department of Health and Human Services guidelines for human experimentation and was approved by the Colorado Multiple Institutional Review Board, the Joint Parirenyatwa Hospital College of Health Sciences Research Ethics Committee, and the Medical Research Council of Zimbabwe. The study was conducted by retrospective review of medical records from the Parirenyatwa Hospital Kaposi Sarcoma Clinic and the Family Care Centre (also known as the Opportunistic Infections/ART Clinic) in Harare, Zimbabwe. At that time, the Family Care Centre provided ART for patients with: CD4+ lymphocyte count <250 cells/mm3, history of World Health Organization (WHO) HIV stage 3 or 4 disease, current pregnancy, or current use of ART at enrolment. All patients received an intake assessment where available pre-treatment demographic and clinical data were recorded. Clinical assessment included determination of AIDS Clinical Trials Group (ACTG) KS disease stage21, 22 for AIDS-KS patients. Pre-treatment laboratory data typically included HIV serology, CD4+ lymphocyte count, complete blood count, basic metabolic panel, and liver function testing; however, complete baseline laboratory tests were not obtained in all cases. The standard initial ART regimen was stavudine, lamivudine, and nevirapine. Patients receiving concomitant treatment for tuberculosis had efavirenz substituted for nevirapine. Patients typically received antiretroviral medications in 1-month allotments and were required to refill prescriptions at the Family Care Centre pharmacy monthly.
All AIDS-KS patients enrolled in the Kaposi Sarcoma Clinic at Parirenyatwa Hospital, receiving ART through the Family Care Centre during the study enrolment period (January 1, 2004 to November 1, 2006), were screened for inclusion. Patients with less than 6 months of follow-up in the Kaposi Sarcoma Clinic and/or the Family Care Centre, or documented transfer of care to another facility, were excluded from the study to minimize inclusion of patients not committed to care through these clinics. Patients who had received more than 2 months of previous ART prior to enrolment at the Family Care Centre were also excluded. Included AIDS-KS patients were matched 1:3 to HIV-positive patients meeting a diagnosis of AIDS based on a CD4+ lymphocyte count <250 cells/mm3, or WHO HIV stage 4 disease, without known KS (non-KS), also receiving ART through the Family Care Centre. Matching was based on gender, date of initiation of ART, and age. All patients were gender-matched exactly. Non-KS AIDS patients of the appropriate gender who had initiated ART within 2 months of the date an AIDS-KS patient had started treatment were then identified. Of these identified patients, the non-KS AIDS patient closest in age to the AIDS-KS patient being matched was subsequently included in the study. Patients were matched for gender and age because Zimbabwean AIDS-KS patients are more often male and have an older age distribution than non-KS AIDS patients.23, 24 Pregnancy was not accounted for in matching, as few (n = 5) gravid patients were included.
2.2. Statistical analysis
The primary outcome was loss-to-care, defined as failure to attend clinic or refill prescriptions for 3 months or longer during the study period. Loss-to-care was confirmed by follow-up review of the available medical records after study termination. Loss-to-care was used as the primary outcome because patient deaths were not accurately recorded in Zimbabwe at the time of the study and available data suggest that in African settings, loss-to-care for more than 3 months is associated with clinically relevant outcomes including severe debilitation and death.25, 26 Loss-to-care was analyzed by standard univariate Kaplan–Meier methodology with termination of the study (August 1, 2007) as a censoring event. A two-step model-selection strategy was used to identify independent predictors of loss-to-care. Pre-treatment variables included in an initial screening univariate Cox proportional hazards regression analysis were: presence of KS, gender, age, marital/partner status (married/partner vs. no partner), having private medical aid (yes vs. no or not reported), employment status (full- or part-time employment vs. unemployed, self-employed or not reported), education level (O level education or higher vs. secondary education or less), housing type (own or rent vs. living with family or homeless), WHO performance status (1 vs. >1), pre-treatment CD4+ lymphocyte count (<50 vs. ≥50 cells/mm3), pre-treatment hemoglobin (g/dl), and pre-treatment albumin (g/dl). A final multivariate Cox proportional hazards regression analysis was performed on all variables with p < 0.15 in the initial univariate analysis, with statistical significance defined as p < 0.05.
The percent change in weight at 6 months after initiation of ART was a secondary outcome. The absolute change in CD4+ lymphocyte count and the total CD4+ lymphocyte count within 1 year of initiating ART were also defined secondary outcomes; however, only five AIDS-KS patients and 46 non-KS AIDS patients had follow-up CD4+ cell counts recorded and therefore assessment of these outcomes was underpowered to detect a significant difference. Secondary outcome variables were non-parametrically distributed and were analyzed using the Mann–Whitney rank sum test. A subgroup analysis was performed comparing the pre-treatment variables of AIDS-KS patients retained-in-care to those AIDS-KS patients lost-to-care. Pre-treatment demographic and clinical variables were analyzed by Fisher's exact test for categorical variables and Student's t-test for continuous variables. Statistical significance was defined as p < 0.05 and all applicable tests were two-tailed.
3. Results
The medical records of 110 AIDS-KS patients were reviewed and
การแปล กรุณารอสักครู่..
