Correspondence

5-Year Survival of Patients with AIDS Receiving Antiretroviral Therapy in Haiti

N Engl J Med 2009; 361:828-829August 20, 2009

Article

To the Editor:

We report 5-year outcomes of 910 adults with human immunodeficiency virus (HIV) infection who consecutively initiated antiretroviral therapy according to guidelines of the World Health Organization (WHO) from 2003 through 2004. All the patients were followed at the Groupe Haïtien d'Etude du Sarcome de Kaposi et des Infections Opportunistes (GHESKIO) clinic in Port au Prince, Haiti, through May 1, 2009.1,2 Combination therapy with zidovudine, lamivudine, and efavirenz was initiated in 428 patients (47%); therapy with zidovudine, lamivudine, and nevirapine in 381 patients (42%); and other regimens in 101 patients (11%). CD4 counts were measured every 6 months, and HIV type 1 (HIV-1) RNA levels were measured when patients met WHO clinical criteria or CD4 criteria for failure of antiretroviral therapy. The study was approved by the institutional review board at the GHESKIO Centers in Haiti and at Cornell University in New York.

Of the 910 patients, 70 (8%) were lost to follow-up, and 208 (23%) died. For 738 patients who were receiving care at 6 months, 587 (80%) had an adherence level of 90% or more. According to Kaplan–Meier analysis, 79% of the 910 patients were still alive at 60 months (Figure 1Figure 1Kaplan–Meier Estimates of Survival after the Initiation of Antiretroviral Therapy in 910 Adults with Human Immunodeficiency Virus Infection in Haiti.). The rate of death in the first 6 months (25 deaths per 100 person-years) was seven times the rate after 6 months (3.3 deaths per 100 person-years) (P<0.001). Deaths during the first 6 months were associated with having an AIDS-related illness, a weight in the lowest quartile, and a CD4 count of less than 50 cells per milliliter (P<0.001 for all comparisons). Deaths after 6 months were associated with an adherence rate of less than 90% (P<0.001), an age of more than 50 years (P=0.009), and a diagnosis of tuberculosis during the first 6 months of antiretroviral therapy (P=0.02).

Of the 910 patients, 121 (13%) had a drug-related toxic effect that prompted a change in first-line therapy: anemia associated with zidovudine in 51 patients, central-nervous-system symptoms associated with efavirenz in 25 patients, gynecomastia in 20 men receiving efavirenz, rash associated with nevirapine in 15 women, and other effects in 10 patients. A total of 263 patients (29%) met the WHO clinical criteria or CD4 criteria for treatment failure. Of the 211 patients for whom data regarding HIV-1 RNA levels were available, 113 (54%) had a plasma HIV-1 RNA level of more than 50 copies per milliliter. The positive predictive value of clinical criteria (WHO stage III or IV HIV-related symptoms) for detectable HIV-1 RNA was only 48%. The positive predictive value of CD4 criteria was 77%. HIV-1 reverse-transcriptase genotyping was performed for 91 patients who had a plasma HIV-1 RNA level of more than 1000 copies per milliliter at the time of treatment failure, according to WHO criteria (see the table in the Supplementary Appendix, available with the full text of this letter at NEJM.org).

We also analyzed HIV-1 RNA levels in banked plasma from 405 control subjects in the same cohort who did not meet the WHO criteria for treatment failure. Two control subjects were selected for each patient with treatment failure with the use of incidence density sampling, with matching for age, sex, and length of follow-up. Of the 405 control subjects, 63 (16%) had a plasma HIV-1 RNA level of more than 50 copies per milliliter. The sensitivity of the WHO clinical and CD4 criteria for predicting virologic failure was 113 of 176 patients (64%), and the specificity was 342 of 440 (78%).

This report documents the long-term sustainability of antiretroviral-therapy programs in resource-poor countries, with excellent rates of retention and adherence and a survival rate of 75% at 5 years. The WHO criteria of treatment failure are not sensitive or specific. Waiting to meet these criteria may delay the recognition of virologic failure and result in the accumulation of HIV-1 drug-resistance mutations.

Paul Leger, M.D.
Macarthur Charles, M.D., Ph.D.
Patrice Severe, M.D.
Cynthia Riviere, M.D.
Jean William Pape, M.D.
Groupe Haïtien d'Etude du Sarcome de Kaposi et des Infections Opportunistes (GHESKIO), Port au Prince, Haiti

Daniel W. Fitzgerald, M.D.
Weill Cornell Medical College, New York, NY

Supported in part by a grant (AI58257) from the National Institute of Allergy and Infectious Diseases; grants (TW006896, TW006901, and TW00018) from the Fogarty International Center; and grants from the Global Fund to Fight AIDS, Tuberculosis, and Malaria and the President's Emergency Plan for AIDS Relief (PEPFAR).

2 References

1. 1

   Severe P, Leger P, Charles M, et al. Antiretroviral therapy in a thousand patients with AIDS in Haiti. N Engl J Med 2005;353:2325-2334
   Full Text | Web of Science | Medline

2. 2

   Scaling up antiretroviral therapy in resource-limited settings: treatment guidelines for a public health approach: 2003 revision. Geneva: World Health Organization, 2004.
   

Citing Articles (4)

Citing Articles

1. 1

   Severe, Patrice, Jean Juste, Marc Antoine, Ambroise, Alex, Eliacin, Ludger, Marchand, Claudel, Apollon, Sandra, Edwards, Alison, Bang, Heejung, Nicotera, Janet, Godfrey, Catherine, Gulick, Roy M., Johnson, Warren D. Jr., Pape, Jean William, Fitzgerald, Daniel W., . (2010) Early versus Standard Antiretroviral Therapy for HIV-Infected Adults in Haiti. New England Journal of Medicine 363:3, 257-265
   Full Text

2. 2

   Serena Koenig, LC Ivers, S Pace, R Destine, F Leandre, R Grandpierre, J Mukherjee, PE Farmer, JW Pape. (2010) Successes and challenges of HIV treatment programs in Haiti: aftermath of the earthquake. HIV Therapy 4:2, 145-160
   CrossRef

3. 3

   Lutgarde Lynen, Johan Van Griensven, Julian Elliott. (2010) Monitoring for treatment failure in patients on first-line antiretroviral treatment in resource-constrained settings. Current Opinion in HIV and AIDS 5:1, 1-5
   CrossRef

4. 4

   (2010) Routine versus clinically driven laboratory monitoring of HIV antiretroviral therapy in Africa (DART): a randomised non-inferiority trial. The Lancet 375:9709, 123-131
   CrossRef