|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||
Background The one-year survival rate of adults and children with the acquired immunodeficiency syndrome (AIDS), without antiretroviral therapy, has been about 30 percent in Haiti. Antiretroviral therapy has recently become available in Haiti and in other developing countries. Data on the efficacy of antiretroviral therapy in developing countries are limited. High rates of coinfection with tropical diseases and tuberculosis, along with malnutrition and limited laboratory monitoring of therapy, may decrease the efficacy of antiretroviral therapy in these countries.
Methods We studied the efficacy of antiretroviral therapy in the first 1004 consecutive patients with AIDS and without previous antiretroviral therapy who were treated beginning in March 2003 in Port-au-Prince, Haiti.
Results During a 14-month period, three-drug antiretroviral therapy was initiated in 1004 patients, including 94 children under 13 years of age. At enrollment, the median CD4 T-cell count in adults and adolescents was 131 per cubic millimeter (interquartile range, 55 to 211 per cubic millimeter); in children, a median of 13 percent of T cells were CD4-positive (interquartile range, 8 to 20 percent). According to a Kaplan–Meier survival analysis, 87 percent of adults and adolescents and 98 percent of children were alive one year after beginning treatment. In a subgroup of 100 adult and adolescent patients who were followed for 48 to 56 weeks, 76 patients had fewer than 400 copies of human immunodeficiency virus RNA per milliliter. In adults and adolescents, the median increase in the CD4 T-cell count from baseline to 12 months was 163 per cubic millimeter (interquartile range, 77 to 251 per cubic millimeter). In children, the median percentage of CD4 T cells rose from 13 percent at baseline to 26 percent (interquartile range, 22 to 36 percent) at 12 months. Treatment-limiting toxic effects occurred in 102 of the 910 adults and adolescents (11 percent) and 5 of the 94 children (5 percent).
Conclusions This report documents the feasibility of effective antiretroviral therapy in a large number of patients in an impoverished country. Overall, the outcomes are similar to those in the United States. These results provide evidence in support of international efforts to make antiretroviral therapy available to patients with AIDS in developing countries.
We report the outcomes for the first 1000 patients with AIDS and without previous antiretroviral therapy who were treated consecutively at a clinic in Port-au-Prince, Haiti, beginning in March 2003, when international funding for antiretroviral therapy first became available.13 Haiti is the poorest country in the Western Hemisphere and has suffered from nearly constant political unrest for the past 20 years.14 AIDS was recognized in Haiti in the early 1980s.15 HIV infection occurs primarily through heterosexual transmission, and the prevalence is currently estimated at 3 percent in the adult population.16,17
Methods
Patients
Since 1983, the clinic of the Haitian Study Group for Kaposi's Sarcoma and Opportunistic Infections (GHESKIO) has provided the Port-au-Prince population, which is currently estimated at about 2 million people, with free HIV counseling, testing for HIV infection, and services for prevention of HIV infection, as well as care for patients with AIDS.18
Between March 2003 and December 2004, the CD4 T-cell count was determined for all adult and adolescent patients coming to the GHESKIO clinic with symptoms of HIV infection. Antiretroviral therapy, according to World Health Organization (WHO) guidelines and without regard to the perceived likelihood that the patient would adhere to the therapy, was then initiated for patients with an AIDS-defining illness or a CD4 T-cell count under 200 per cubic millimeter.1,19 The diagnosis of HIV infection in children was based on serologic testing, clinical findings, and the percentage of T cells that were CD4-positive; antiretroviral therapy was initiated in infected children according to WHO pediatric guidelines.1
The full complement of personnel serving the cohort of 1004 patients described in this report consisted of five community health workers, two AIDS peer counselors, one social worker, one pharmacist, five nurses, and three physicians. The patients were routinely seen in the clinic by a physician every two weeks during the first three months of treatment and by a nurse monthly thereafter. Medications were dispensed directly to the patients on a monthly basis.
Treatment
First- and second-line antiretroviral-therapy regimens followed WHO guidelines, as recommended by the Haitian government.1 The first-line antiretroviral-therapy regimen for adults and adolescents consisted of zidovudine, lamivudine, and efavirenz. Single-drug substitutions were permitted: stavudine could be substituted for zidovudine, and nevirapine could be substituted for efavirenz. Other single-drug substitutions were permitted, according to WHO guidelines. The first-line regimen for children under 3 years of age substituted nevirapine for efavirenz.
WHO-approved generic medications constituted about 90 percent of the supply. When generic drugs were not available, brand-name medications were purchased. The cost of the three antiretroviral medications per patient per year ranged from $550 for generic medications to $750 for brand-name medications. We budgeted an additional $1,000 per patient per year to cover other costs, including personnel ($450), laboratory monitoring ($300), medications other than antiretroviral drugs ($75), data monitoring ($75), and miscellaneous other costs ($100). We estimated the overall cost per patient per year as about $1,600.
Extrapulmonary tuberculosis, but not pulmonary tuberculosis, was considered an AIDS-defining illness.16 HIV-infected adults and adolescents with pulmonary tuberculosis and a CD4 T-cell count of more than 200 per cubic millimeter were treated for tuberculosis, and initiation of antiretroviral therapy was deferred. In patients with tuberculosis who had a CD4 T-cell count between 50 and 200 per cubic millimeter, antiretroviral therapy was initiated after the completion of two months of tuberculosis therapy. In patients with a CD4 T-cell count of less than 50 cells per cubic millimeter, tuberculosis treatment and antiretroviral therapy were started simultaneously. The tuberculosis regimen consisted of two months of isoniazid, rifampin, ethambutol, and pyrazinamide daily, followed by four months of isoniazid and rifampin daily.20
Adherence to therapy was encouraged by home visits, provision of free telephone cards for patients to call clinic staff, peer counseling by people with AIDS, pill counts, and social support programs. If it was indicated, patients were referred for nutritional aid, and counseling was offered to pregnant women, parents of HIV-infected children, and victims of domestic violence. Directly observed antiretroviral therapy, which has been very effective in rural Haiti,21 was not feasible for our urban patients, who have limited social networks and frequently change their addresses.
Clinical Measurements
Body weight was measured at every visit. The z score for weight for children was reported as the number of standard deviations above or below the median weight for age. Laboratory monitoring included the baseline CD4 T-cell count by flow cytometry (Becton Dickinson) and measurement of hemoglobin. The CD4 T-cell count was determined every six months. Follow-up hemoglobin measurements, liver-function tests, and serum chemical analyses were performed if clinically indicated. The level of HIV RNA in plasma at 12 months was determined in a subgroup of all available adult and adolescent patients who had been followed up for 48 to 56 weeks in December 2004. The Amplicor HIV-1 Monitor PCR Test (Roche), with a lower limit of detection of 400 copies of HIV RNA per milliliter, was used.
Statistical Analysis
The institutional review boards at GHESKIO and at Weill Medical College of Cornell University approved this study. We collected data from an electronic medical record and from the charts of patients with AIDS in whom three-drug antiretroviral therapy had been initiated between March 1, 2003, and April 30, 2004. Follow-up data collected through December 31, 2004, were included. The data were analyzed by an intention-to-treat approach with the use of SAS software. Proportions were compared by the chi-square test with Yates' correction or, for expected cell values of less than five, by Fisher's exact test. Means and medians were compared by Student's t-test and the Wilcoxon rank-sum test, respectively. Kaplan–Meier survival analyses were used to estimate the time from the initiation of antiretroviral therapy to death and the time from initiation to the first treatment-limiting toxic drug effect. For patients who did not reach the end point, the data were censored at the date of the last visit. The log-rank test was used to compare survival times between strata. The Cox proportional-hazards model was used for multivariate analysis. Variables associated with mortality in previous publications or in our clinical experience were included in the initial model. Variables were removed from the model by a backward selection procedure if the value of alpha was greater than 0.05. Confounders causing a 10 percent change in another predictor were left in the model.
Results
Enrollment
Between March 1, 2003, and April 30, 2004, GHESKIO provided HIV counseling, testing, and primary care to 23,394 patients, of whom 3978 (17 percent) were HIV-seropositive. Of these HIV-infected patients, 1004 (25 percent) who had not previously received treatment met the criteria for antiretroviral therapy, and therapy was initiated in these patients; 2778 (70 percent) had early-stage HIV disease and did not require antiretroviral therapy; 57 (1 percent) met the criteria for antiretroviral therapy but were lost to follow-up before therapy could be initiated; 129 (3 percent) had previously received antiretroviral therapy; and 10 (<1 percent) died after they were tested for HIV and before antiretroviral therapy could be initiated (Figure 1). The mean enrollment rate was 80 new patients per month (range, 32 to 141). Enrollment continued after April 2004, and approximately 2500 patients are currently receiving antiretroviral therapy at GHESKIO. The baseline characteristics of the first 1004 patients and their initial antiretroviral-therapy regimens are provided in Table 1.
|
|
At the time of data analysis, 800 of the 1004 patients (80 percent) were still being followed up, 75 (7 percent) had been lost to follow-up, and 129 (13 percent) had died. There was no difference in baseline characteristics (including age, sex, CD4 T-cell count, body weight, and stage of HIV infection) between patients who were lost to follow-up and those who were not. The most common reason for patients' becoming lost to follow-up was their leaving Port-au-Prince to return to their rural villages. The median follow-up time for the 1004 patients was 13 months. The cohort profile is shown in Figure 1.
Survival
Of the 910 adult and adolescent patients, 127 (14 percent) died. Of the 127 deaths, 55 (43 percent) were due to persistent wasting syndrome, 20 (16 percent) to tuberculosis, 6 (5 percent) to bacterial pneumonia, 5 (4 percent) to toxoplasmosis, 4 (3 percent) to cancer, 4 (3 percent) to cryptosporidiosis, 4 (3 percent) to sepsis syndrome, 3 (2 percent) to congestive heart failure, 3 (2 percent) to trauma, and 23 (18 percent) to unknown causes. One hundred of the 127 deaths (79 percent) occurred within six months after the initiation of antiretroviral therapy. According to survival analysis, 90 percent of the patients were alive at 6 months and 87 percent at 12 months (Figure 2). The factors present at the initiation of antiretroviral therapy that were predictors of death were the presence of an AIDS-defining illness, a CD4 T-cell count under 50 per cubic millimeter, and a body weight in the lowest quartile for sex (Table 2).
|
|
Other Measures of Treatment Response
Virologic Response
Plasma HIV RNA was measured in adult and adolescent patients who had a 1-year follow-up visit (at 48 to 56 weeks) in December 2004. Plasma was available from 100 of the 117 adult and adolescent patients who had been followed for one year (85 percent). Viral load was less than 400 copies of HIV RNA per milliliter in 76 of the 100 patients tested. There were no significant differences in baseline characteristics between the 76 patients with fewer than 400 copies of HIV RNA per milliliter and the 24 patients with 400 or more copies per milliliter.
CD4 T-Cell Response
In adults and adolescents, the median increase in the CD4 T-cell count at six months was 128 per cubic millimeter (interquartile range, 62 to 197 per cubic millimeter). The CD4 T-cell count at six months was greater than the baseline value in 459 of 504 patients (91 percent) and remained the same or decreased from baseline in 45 (9 percent). The median increase in the CD4 T-cell count at 12 months was 163 per cubic millimeter (interquartile range, 77 to 251 per cubic millimeter). The CD4 T-cell count at 12 months was greater than the baseline value in 360 of 397 patients (91 percent) and remained the same or decreased from baseline in 37 (9 percent). In children, the median CD4 T-cell percentage rose from a baseline value of 13 percent (interquartile range, 8 to 20 percent) to 21 percent (interquartile range, 16 to 29 percent) at 6 months and to 26 percent (interquartile range, 22 to 36 percent) at 12 months.
Weight Gain
At six months, adult and adolescent patients had gained a median of 4.0 kg (interquartile range, 0.9 to 7.7). At six months, 639 of 759 patients (84 percent) had gained weight, and 120 patients (16 percent) remained at the same weight or had lost weight. At 12 months, adult and adolescent patients had gained a median of 5.5 kg (interquartile range, 1.4 to 10.5). At 12 months, 396 of 466 patients (85 percent) had gained weight, and 70 (15 percent) remained at the same weight or had lost weight. In children, the z score for weight increased from a median value of –1.8 (interquartile range, –2.6 to –1.1) at enrollment, to –1.3 at 6 months (interquartile range, –1.9 to –0.6) and to –1.2 at 12 months (interquartile range, –1.6 to –0.4).
Tuberculosis
Of the 910 adult and adolescent patients, 113 (12 percent) received concurrent treatment for tuberculosis while receiving antiretroviral therapy, 72 (8 percent) began treatment for tuberculosis before beginning antiretroviral therapy, and 41 (5 percent) began treatment for tuberculosis after beginning antiretroviral therapy. Two patients with CD4 T-cell counts of 50 to 200 per cubic millimeter started tuberculosis treatment but died in the two-month period before they would have begun antiretroviral therapy (Figure 1). Of the 113 patients who received concurrent treatment for tuberculosis while receiving antiretroviral therapy, 89 (79 percent) were cured of tuberculosis, 20 (18 percent) died, 1 (1 percent) did not respond to tuberculosis therapy, and 3 (3 percent) were lost to follow-up. There was no significant difference in survival between patients with and patients without tuberculosis.
Of the 72 patients who started tuberculosis treatment before antiretroviral therapy, 11 (15 percent) had a suspected immune reconstitution syndrome with temporary worsening of tuberculosis symptoms after beginning antiretroviral therapy, including recurrent fevers, increasing cough, and a draining lymph-node fistula; 4 were treated with corticosteroids. No patient stopped antiretroviral therapy because of the immune reconstitution syndrome.
Forty-one patients were found to have tuberculosis after beginning antiretroviral therapy; symptoms of tuberculosis commenced within three months after the initiation of antiretroviral therapy in 26 (63 percent). A weight loss of more than 5 percent at month 3 was associated with the diagnosis of tuberculosis after the initiation of antiretroviral therapy (odds ratio, 2.62; 95 percent confidence interval, 1.17 to 5.86; P=0.04).
Among the 94 children, 13 (14 percent) received antiretroviral therapy with concurrent treatment for tuberculosis. Eleven of these 13 children (85 percent) appeared to have been cured of tuberculosis, and 2 (15 percent) were lost to follow-up; none of them died. There was no significant difference in survival between children with and without tuberculosis.
First-Line Medication Changes and Toxic Effects
Of the 910 adult and adolescent patients, 229 (25 percent) required a change in a first-line medication. The reasons for the change were toxic effects in 102 patients (11 percent), disruption in the medication supply in 66 patients (7 percent), sexual activity by women of reproductive age in 29 patients (3 percent), suspected treatment failure in 11 patients (1 percent), and tuberculosis in 21 patients (2 percent). The disruptions in medication supply occurred when international suppliers were months late in delivering medications.
Table 3 shows the treatment-limiting toxic effects of first-line medications. Anemia and central nervous system symptoms were the most common. Two patients had the Stevens–Johnson syndrome, one of whom died; nevirapine was the suspected cause in both cases. Gynecomastia, in several cases accompanied by lactorrhea, occurred in 15 men receiving efavirenz. According to Kaplan–Meier survival analysis, 14 percent of the adult and adolescent patients required a change in first-line medication because of a toxic effect during the first 12 months of antiretroviral therapy.
|
Discussion
Antiretroviral therapy was initiated in 1004 patients with AIDS within 14 months after international funds for therapy became available in Haiti. This large cohort was consecutively enrolled, had high rates of poverty, malnutrition, and tuberculosis, and was similar to clinic populations that would be found in other urban areas severely affected by HIV disease in the Caribbean or sub-Saharan Africa. The one-year survival was 87 percent for adults and adolescents and 98 percent for children. In comparison, the one-year survival without antiretroviral therapy for adults and children with AIDS in Haiti and other developing countries is about 30 percent.22,23,24 This rapid and effective large-scale introduction of antiretroviral therapy in the poorest country in the Western Hemisphere, even during times of political unrest, provides evidence in support of international efforts to make antiretroviral therapy available to patients with AIDS worldwide.
The virologic response rate of 76 percent (with a response defined as <400 copies of HIV RNA per milliliter) and the median increase in the CD4 T-cell count of 163 per cubic millimeter at one year in adults and adolescents are similar to results from the United States. In a meta-analysis of clinical trials involving adults not previously treated with antiretroviral therapy who received a three-drug antiretroviral-therapy regimen that included a nonnucleoside reverse-transcriptase inhibitor, the virologic response rate (<400 copies of HIV RNA per milliliter) at 48 weeks was 72 percent and the mean increase in the CD4 T-cell count was 174 per cubic millimeter.25,26 An analysis of the 12-month outcomes of patients beginning antiretroviral therapy in 2001 and 2002 in a Baltimore clinic found a virologic response rate (<400 copies of HIV RNA per milliliter) of 68 percent and a mean increase in the CD4 T-cell count of 139 per cubic millimeter.27,28
The rates of treatment-limiting toxic effects in our patients were similar to those reported for patients from developed countries who were treated with similar regimens.29,30 Despite recent concern about nevirapine toxicity,31 less than 1 percent of the patients in our cohort who were treated with nevirapine had hepatitis. Nevirapine-induced hepatitis has been associated with a CD4 T-cell count of more than 250 per cubic millimeter, but the count in most of our patients was less than 200 per cubic millimeter. Among the patients who began therapy with efavirenz, treatment-limiting toxic effects occurred in 10 percent, a rate higher than that reported from the United States. Most Haitians are of African descent, and a recent study demonstrated that people of African descent are at greater risk than others for central nervous system side effects from efavirenz.32 An additional 6 percent of patients who began therapy with efavirenz stopped because of the risk of teratogenicity in sexually active women. Alternatives to first-line regimens containing efavirenz may be especially important for populations of African origin with large numbers of sexually active women.
The challenges involved in providing antiretroviral therapy in developing countries include high rates of poverty, malnutrition, and tuberculosis. The majority of our patients earned less than $1 a day, the World Bank's international poverty line.33 Poverty affects all aspects of care, including patients' ability to buy food, obtain access to clean water and housing, and pay for transportation to the clinic. We have therefore striven to integrate our antiretroviral-therapy program with existing social programs. There is a growing body of evidence that malnutrition is a critical cofactor in AIDS progression in resource-poor countries.34,35 In our cohort, low baseline body weight was an independent predictor of death. We therefore provided a daily multivitamin supplement to all of our patients who were receiving antiretroviral therapy and a monthly stock of rice, beans, and vegetable oil to the most undernourished patients. Twelve percent of our patients who were receiving antiretroviral therapy received concurrent treatment for tuberculosis. Data from this study suggest that a failure to gain weight after three months of antiretroviral therapy should prompt an evaluation for tuberculosis. Patients with tuberculosis had a survival rate similar to that of patients without tuberculosis.
The single greatest logistic challenge was maintaining the supply of antiretroviral drugs. Delays in delivery resulted in a change in the antiretroviral-therapy regimen for 7 percent of our patients. Developing reliable manufacturing and distribution systems for antiretroviral drugs is an urgent international priority. Other logistic challenges may arise in different developing countries.
We estimated the overall cost of treating a patient with antiretroviral therapy to be about $1,600 per year, with antiretroviral medications accounting for 35 to 45 percent of the total. The care of the 1004 patients receiving antiretroviral therapy was supported by the Global Fund to Fight AIDS, Tuberculosis and Malaria and the U.S. President's Emergency Plan for AIDS Relief (PEPFAR). However, there are a number of other, intangible factors that significantly contributed to the successful implementation of this program, including continuity of leadership, international collaboration, and dedication of GHESKIO personnel who provided care to poor patients with AIDS in an urban slum, even during times of violent political unrest. These factors are the product of infrastructure development, personnel training, and mentorship, which are critical for scaling up antiretroviral therapy. The treatment outcomes achieved in Haiti were similar to those achieved in U.S. clinics, providing evidence in support of international efforts to make antiretroviral therapy available in developing countries. Finally, the sustainability of programs in Haiti and in other impoverished countries is absolutely dependent on continued international support.
Supported in part by grants from the National Institute of Allergy and Infectious Diseases (AI38858, AI64021, AI58257, AI51966, and AI 07613), the Fogarty International Center (TW006896, TW006901, TW00018, and TW00002), the Global Fund to Fight AIDS, Tuberculosis and Malaria, and PEPFAR.
Dr. Gulick reports having received consultation fees, lecture fees, or grant support from Abbott, Bristol-Myers Squibb, Boehringer Ingelheim, GlaxoSmithKline, Merck, and Pfizer. Dr. Wright reports having received consultation fees, lecture fees, or grant support from Bristol-Myers Squibb, GlaxoSmithKline, MedImmune, Merck, and Pfizer.
We are indebted to Marie-Eugene Beaulieu, Sandra Apollon, Claudine Nolte, Shilla Nassi, Jean Duperval, Abdias Marcelin, Marshall Glesby, and David Haas.
Source Information
From the Groupe Haitien d'Etude du Sarcome de Kaposi et des Infections Opportunistes, Port-au-Prince, Haiti (P.S., P.L., M.C., F.N., G. Bonhomme, G. Bois, E.G., S.K.-P., J.W.P.); the Division of International Medicine and Infectious Diseases, Weill Medical College of Cornell University, New York (M.C., E.G., R.G., W.D.J., J.W.P., D.W.F.); and the Division of Pediatric Infectious Diseases, Vanderbilt University, Nashville (P.W.).
Address reprint requests to Dr. Fitzgerald at the Division of International Medicine and Infectious Diseases, Weill Medical College of Cornell University, Room A-421, 1300 York Ave., New York, NY 10021 or at dfitzgerald{at}gheskio.org.
References
| |||||||||||||||||||||||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2010 Massachusetts Medical Society. All rights reserved. |
Previous
