Two randomized controlled trials in HIV-infected adults in Abidjan, Côte d'Ivoire provided evidence of benefit from trimethoprim-sulfamethoxazole (TS) prophylaxis. In the first study, TS prophylaxis reduced the incidence of serious adverse events in persons with World Health Organization (WHO) stage II or III HIV disease, including pneumonia, isosporiasis, malaria, and acute unexplained fever, but did not significantly reduce mortality.1 In a second study, TS prophylaxis in HIV-positive patients coinfected with tuberculosis reduced mortality by 46% as well as reducing hospital admission rates, particularly for enteritis, nontyphoidal Salmonella (NTS) sepsis, and isosporiasis.2 The incidences of Pneumocystis jiroveci pneumonia (PcP) and Toxoplasma encephalitis (TE) were low in both studies, with the exception of TE in the group with CD4 counts less than 100 cells/mm3. Based on these 2 studies, the WHO and UNAIDS made a provisional recommendation of TS prophylaxis for all HIV-infected adults and children in Africa who are symptomatic or asymptomatic with a CD4 count less than 500 cells/mm3, as well as for all infants born to HIV-infected mothers.3
These recommendations have not been widely implemented throughout the region for a number of reasons. There are logistic and financial barriers to their introduction in resource-poor settings, and the potential drawbacks of TS prophylaxis have not been adequately assessed. Concerns have been raised that the prophylactic efficacy of TS may be lower elsewhere in sub-Saharan Africa than was reported in Côte d'Ivoire because of much higher levels of bacterial resistance to TS.4,5
An additional concern is that widespread TS prophylaxis may accelerate the development of malaria parasites resistant to antifolate antimalarial drugs, such as sulfadoxine-pyrimethamine (SP). High rates of chloroquine failure have led several African countries to move to SP as the first-line antimalarial drug; in 1993, Malawi became the first African country to make this change. We have recently shown that SP retains approximately 80% clinical efficacy against Plasmodium falciparum in Malawi.6 In vitro studies have demonstrated evidence of P. falciparum cross-resistance between trimethoprim and pyrimethamine (conferred by mutations in P. falciparum dihydrofolate reductase [DHFR]) and between sulfamethoxazole and sulfadoxine (conferred by mutations in dihydropteroate synthase [DHPS]).7,8 TS prophylaxis may thus increase the spread of SP resistance in populations. If individuals on TS prophylaxis develop malaria, the risk of SP treatment failure may be higher. Because of these concerns, health policy makers in Malawi do not recommend routine TS prophylaxis for HIV-infected individuals and have called for further research into the risks and benefits of TS prophylaxis.
Little is known about the disease burden of TS-preventable infections and the associated antibiotic resistance patterns in persons living with HIV in Malawi and other settings with relatively high rates of bacterial resistance to TS. One nonrandomized study reported a mortality benefit for TS prophylaxis, but the comparison group had significantly higher CD4 cell counts and less advanced HIV disease than the prophylaxis group.9 Another study using historical controls observed lower mortality among Malawian tuberculosis patients offered voluntary counseling and testing (VCT) for HIV and TS prophylaxis.10 One prospective controlled trial did not show benefit from TS prophylaxis in adults from Senegal with CD4 counts less than 400/mm3, but this trial was stopped prematurely after results from the Côte d'Ivoire studies were presented.11
To assess the potential role of TS prophylaxis for people living with HIV in Malawi, we conducted a prospective observational cohort study in Blantyre, the commercial center of the country. Outcomes of interest were the incidence of severe events (hospitalizations and deaths) and the incidence of clinical illness episodes that could potentially be prevented by TS prophylaxis: bacteremia or sepsis, isosporiasis, bacterial enteritis, bacterial pneumonia, PcP, otitis media, sinusitis, bacterial meningitis, TE, urinary tract infection, skin and soft tissue infections, and malaria. We determined the prevalence of TS resistance in bacterial isolates and the in vivo efficacy of SP for treating falciparum malaria.
The study was conducted at Ndirande Health Center, which is situated in a large township on the outskirts of Blantyre, from September 2002 through December 2003. Approximately 100,000 inhabitants of various Malawian ethnic backgrounds populate the township. Other than private dispensaries and traditional medical practitioners, the Ndirande Health Center is the only source of medical care serving this relatively poor population. At the time of this study, antiretroviral drugs were not readily available in Malawi. The study staff provided basic medical care at the health center; patients needing hospitalization were referred to the Queen Elizabeth Central Hospital, a large public teaching hospital in Blantyre affiliated with the University of Malawi College of Medicine.
After VCT for HIV at the Ndirande Health Center, informed consent was sought from any HIV-infected person aged 16 years or older if he or she was willing to have HIV test results disclosed to study investigators and intended to remain in the study area for at least 8 months. Chronic prophylaxis and treatment with antifolate drugs were the only exclusion criteria. A full history and physical examination were carried out on recruitment, with staging of HIV infection and AIDS according to WHO classifications.12 Clinical evaluations were performed at scheduled monthly follow-up visits. Participants were encouraged to report to the clinic whenever they developed symptoms. If they missed a monthly appointment, study nurses actively traced them at work or home. If they did not attend the clinic on 2 consecutive scheduled visits, they were considered lost to follow-up.
Determine HIV rapid diagnostic strips (Abbott Laboratories, Abbott Park, IL) and UniGold (Trinity Biotech, Wicklow, Ireland) were used for HIV testing. Discrepancies were resolved by a third rapid test (Hemostrip HIV; Saliva Diagnostics, Vancouver, WA). On recruitment, CD4 and CD8 counts (FacsCount; Becton-Dickinson), measurements of hemoglobin concentrations (HemoCue; HemoCue AB, Angelholm, Sweden), measurement of IgG antibody titers against Toxoplasma (Captia Toxoplasma gondii IgG; Trinity Biotech), and blood films for malaria (thick and thin films stained with Field's stain) were conducted. CD4 and CD8 counts were repeated every 4 months to help guide diagnostic evaluations for opportunistic infections associated with specific CD4 count ranges. Hemoglobin concentrations and blood films for malaria were repeated monthly. Blood for aerobic bacterial culture (BacT/Alert 3D; BioMerieux, Marcy l'Etoile, France) and malaria smears was obtained for all febrile episodes. All pathogens recovered by culture of blood, cerebrospinal fluid, or pus were tested for antimicrobial susceptibility by the Standardized Disc Diffusion Testing Method.13 SP therapeutic efficacy was assessed in patients with uncomplicated falciparum malaria according to a standard WHO protocol.14 Sputum induction by means of an ultrasonic nebulizer was performed, and immunofluorescent staining for P. jiroveci cysts was done for suspected cases of PcP. A nested polymerase chain reaction (PCR) for P. jiroveci was performed on sputum samples from the probable PcP cases (those with clinical suspicion and good response to PcP treatment but with negative immunofluorescence staining of induced sputum samples).15 Stool examinations were performed with epifluorescence microscopy for Isospora belli, auramine staining for I. belli and Cryptosporidium spp. oocysts, and modified Ziehl-Neelsen stains for Cryptosporidium spp. and I. belli oocysts.16-18 Other diagnostic tests were performed, including basic biochemistry, hematologic, microbiologic, and radiologic investigations.
Detailed standardized case definitions and diagnostic and treatment guidelines were designed for this study and followed by study clinicians at the Ndirande Health Center and at Queen Elizabeth Central Hospital. A chest radiograph consistent with pneumonia was required for a diagnosis of bacterial pneumonia. Diagnoses of confirmed bacterial pneumonia, confirmed bacterial enteritis, and bacteremia or sepsis were assigned when pathogenic bacteria were isolated from the blood culture. A diagnosis of PcP was considered in patients on the basis of a consistent clinical picture, a chest radiograph, a positive oxygen saturation exercise test,19 and a recent CD4 count <200 cells/mm3. Each case was reviewed by a senior clinician. The diagnosis of TE was considered in any patient with new-onset convulsions, focal neurologic deficits, lowered consciousness, or signs of raised intracranial pressure who had a positive Toxoplasma IgG and for whose illness there was no other identifiable explanation. Supportive evidence included a low CD4 count (<200 cells/mm3), compatible findings on a computed tomography (CT) scan with contrast of the brain (when available), and a clinical response to toxoplasmosis treatment.
This study was approved by Institutional Review Boards at the University of Malawi College of Medicine, University of Maryland, and Michigan State University. Written informed consent was obtained from all participants in their preferred language (Chichewa or English).
For each event, an incidence rate of all episodes during the study period (including first and subsequent episodes in 1 individual, with the exception of Kaposi sarcoma, which only contributed once to incidence rates) per 100 person-years of follow-up was calculated with 95% confidence intervals based on Poisson distribution. The incidence rates of severe events were stratified by CD4 cell count and WHO clinical HIV stage.
A total of 4739 persons underwent VCT, of whom 1267 (27%) were found to be HIV-1-seropositive. Six hundred sixty participants were enrolled in the study and accumulated 4082 person-months of follow-up. Sixty-six percent of participants were female. This proportion was consistent with that observed in the population presenting for VCT (data not shown). One hundred fifty-five patients (24%) were withdrawn from the study (48 migrated, 17 withdrew consent, 17 started TS prophylaxis elsewhere, and 73 were lost to follow-up), and 52 participants died during the follow-up period. Data collected before death or withdrawal were included in the analysis.
On enrollment, 284 (43%) of the subjects had a CD4 count less than 200 cells/mm3 and 156 (24%) had a WHO clinical stage of III or IV (Table 1). There were 260 severe events (208 hospitalizations and 52 deaths), and the associated incidence rate was 76 per 100 person-years of follow-up. Severe events were correlated with the initial WHO clinical stage and CD4 cell count (see Table 1). Only 34 (5%) of the participants received antiretroviral therapy (ART) at any time during the study period, because this was only available in the Blantyre area at a price that was prohibitive for most participants. The incidence of potentially TS-preventable infections and the incidence rates of other HIV-related events are shown in Table 2.
Severe and invasive bacterial infections were common. The bacteria cultured most often were nontyphoidal salmonellae and Streptococcus pneumoniae (Table 3), and rates of in vitro resistance to TS were high (Fig. 1).
There were 57 episodes that were clinically suspicious for PcP, and induced sputum samples were collected and analyzed. In 1 patient who died despite being treated with TS and prednisolone, P. jiroveci cysts were detected using immunofluorescent staining. Four patients were treated for PcP on clinical grounds, despite negative stains of induced sputum samples, and improved. Their subsequent nested PCR test results for P. jiroveci were negative, making these diagnoses of PcP doubtful.
The overall seroprevalence of Toxoplasma in this patient population was 28%. We suspected TE in 2 individuals with unexplained focal neurologic deficits, but the diagnoses were rejected because of negative Toxoplasma IgG titers and normal CT scans of the brain. Most (74%) cases of nonspecific diarrhea were acute self-limited episodes. Among the confirmed cases of parasitic enteritis were giardiasis (n = 14), strongyloidiasis (n = 6), cryptosporidiosis (n = 1), and isosporiasis (n = 1).
SP therapeutic efficacy was assessed in 73 clinical episodes of uncomplicated falciparum malaria. Fifty-three (73%) of the participants completed the therapeutic efficacy study. Seventy-seven percent had an adequate clinical response to SP, and early and late treatment failures occurred in 17% and 6% of the population, respectively. One patient was hospitalized because of hyperparasitemia (>10% of circulating erythrocytes infected), but none met other criteria for severe malaria.20
This study represents a comprehensive evaluation of opportunistic infections and other AIDS-related syndromes affecting a large population of persons living with HIV in sub-Saharan Africa. Malaria was the most commonly diagnosed illness, highlighting the vulnerability of this population to any decrease in antimalarial drug efficacy and the potential benefit of a prophylactic regimen that prevents malaria. The therapeutic efficacy of SP for the treatment of malaria in our cohort was 77%, similar to that recently reported for children at Ndirande Health Center during the period from 1998 through 2002.6 In Malawi, where most P. falciparum infections carry multiple resistance-conferring DHFR and DHPS mutations,21 TS would be likely to select for the most highly resistant parasites, increasing the individual risk of SP treatment failure and the overall rate of antifolate-resistant malaria in the population. In addition to impairing SP efficacy, widespread TS prophylaxis may threaten the efficacy of chlorproguanil-dapsone, a new, inexpensive, and effective antifolate antimalarial drug.22 Clinical trials of the impact of TS prophylaxis on SP efficacy are needed. If it is shown to impair SP efficacy, alternative antimalaria therapy needs to be made available to persons taking TS prophylaxis.
Seventy-six percent of all the cultured pathogenic bacteria were resistant to TS. In the most serious infections with positive blood cultures, the rate of TS resistance was even higher. Our community-based resistance results are comparable to those found in studies from the tertiary level hospital in Blantyre.23 Although these data raise concerns about the efficacy of TS for prophylaxis of bacterial infections in HIV-infected persons in Malawi, it is possible that the in vivo efficacy of TS in preventing infection may not be accurately reflected in the in vitro susceptibility results. Experience in Côte d'Ivoire after the introduction of TS prophylaxis indicates that despite increasing bacterial resistance, TS seemed to maintain its prophylactic efficacy.24 In a recent report from Uganda, morbidity and mortality decreased in a cohort of adults with HIV infection after they began TS prophylaxis, despite bacterial resistance to TS that was higher than in the Côte d'Ivoire studies but generally lower than we report here.25
Many clinical events could not be precisely diagnosed: episodes of nonspecific diarrhea, unspecified respiratory illnesses, and acute fevers of unknown cause were all common. These episodes generally resolved spontaneously or after empiric therapy with antibiotics not active against PcP and reflect a variety of viral and unidentified bacterial infections and other syndromes. Acute fever of unknown cause and nonspecific diarrhea were among the events that seemed to be prevented by TS prophylaxis in the Côte d'Ivoire studies.1,2
PcP is now recognized as a common problem in African children less than 1 year of age.26 The situation in adults is uncertain: the reported incidence of PcP in African adult populations varies greatly.27 The unavailability of bronchoalveolar lavage could have resulted in underestimating the true incidence of PcP in our study. The sensitivity of induced sputum samples compared with bronchoalveolar lavage is reported to be 67%.28 Thus, even with a conservative estimate of 50% sensitivity, the incidence of PcP would still be low in our cohort.
We diagnosed no cases of TE, consistent with the low incidence rates found in most other African studies.29 The incidence of I. belli infection was also low in this community-based cohort in contrast to earlier data generated from selected patient groups in tertiary hospitals.30,31
Are there other antibiotics that might offer better prophylactic efficacy than TS without compromising SP resistance? Azithromycin, a macrolide antibiotic drug, is available in generic formulations; is safe and well tolerated in AIDS patients32; is active against PcP, Toxoplasma,33 and Salmonella typhi34; and is effective as malaria prophylaxis.35 In our study, azithromycin had good in vitro activity against the gram-positive organisms, Salmonella, and other gram-negative rods, with an overall sensitivity of 73%. This in vitro activity may underestimate the in vivo efficacy for intracellular pathogens such as Salmonella spp., because the drug achieves levels up to 100 times serum levels in phagocytes.36 It acts outside the folate pathway and is thus not expected to interfere with the therapeutic efficacy of SP. An important concern is that widespread use of azithromycin might accelerate the development of bacterial resistance to macrolide drugs,37 including erythromycin, which is frequently prescribed for respiratory illnesses in Malawi.
Many countries in sub-Saharan Africa have started ART programs and are preparing to treat large numbers of patients. The benefits and risks of TS prophylaxis need to be assessed in persons receiving ART, with careful attention to the potential impact on adherence and the ability to assess side effects common to both interventions. With increasingly clear evidence for the benefit of TS even in areas with relatively high bacterial resistance,25,38 the question of the impact of TS prophylaxis on SP efficacy for treating malaria needs to be settled definitively.
The authors thank the Malawi Ministry of Health, National AIDS Commission, and Blantyre District Health Office for their support. They also thank Steven R. Meshnick and Miriam J. Alvarez of the University of North Carolina for doing the nested PCR for PcP.
1. Anglaret X, Chene G, Attia A, et al. Early chemoprophylaxis with trimethoprim-sulphamethoxazole for HIV-1- infected adults in Abidjan, Cote d'Ivoire: a randomised trial. Cotrimo-CI Study Group. Lancet
2. Wiktor SZ, Sassan-Morokro M, Grant AD, et al. Efficacy of trimethoprim-sulphamethoxazole prophylaxis to decrease morbidity and mortality in HIV-1-infected patients with tuberculosis in Abidjan, Cote d'Ivoire: a randomised controlled trial. Lancet
4. Boeree MJ, Harries AD, Zijlstra EE, et al. Co-trimoxazole in HIV-1 infection. Lancet
5. Anglaret X. Trimethoprim-sulfamethoxazole prophylaxis in sub-Saharan Africa. Lancet
6. Plowe CV, Kublin JG, Dzinjalamala FK, et al. Sustained sulfadoxine-pyrimethamine efficacy for uncomplicated falciparum malaria in Malawi after ten years as first line therapy. BMJ
7. Iyer JK, Milhous WK, Cortese JF, et al. Plasmodium falciparum
cross-resistance between trimethoprim and pyrimethamine. Lancet
8. Triglia T, Menting JGT, Wilson C, et al. Mutations in dihydropteroate synthase are responsible for sulfone and sulfonamide resistance in Plasmodium falciparum. Proc Natl Acad Sci USA
9. Badri M, Ehrlich R, Wood R, et al. Initiating co-trimoxazole prophylaxis in HIV-infected patients in Africa: an evaluation of the provisional WHO/UNAIDS recommendations. AIDS
10. Zachariah R, Spielmann MP, Chinji C, et al. Voluntary counselling, HIV testing and adjunctive cotrimoxazole reduces mortality in tuberculosis patients in Thyolo, Malawi. AIDS
11. Maynart M, Lievre L, Sow PS, et al. Primary prevention with cotrimoxazole for HIV-1-infected adults: results of the pilot study in Dakar, Senegal. J Acquir Immune Defic Syndr
13. BSAC Disc Diffusion Method for Antimicrobial Susceptibility Testing. Available at: http://www.bsac.org.uk/uploads/may
%202003susceptibility1.pdf. Accessed November 18, 2004.
14. World Health Organization. Assessment of therapeutic efficacy of antimalarial drugs for uncomplicated Falciparum malaria in areas with intense transmission. 1996. Geneva: World Health Organization, Division of Control of Tropical Diseases. Available at: http://www.who.int/csr/drugresist/malaria/Assessment_malaria_96.pdf
. Accessed November 18, 2004.
15. Olsson M, Stralin K, Holmberg H. Clinical significance of nested polymerase chain reaction and immunofluorescence for detection of Pneumocystis carinii
pneumonia. Clin Microbiol Infect
17. Fayer R, Morgan U, Upton SJ. Epidemiology of Cryptosporidium: transmission, detection and identification. Int J Parasitol
18. Casemore DP. Laboratory methods for diagnosing cryptosporidiosis. J Clin Pathol
19. Smith DE, McLuckie A, Wyatt J, et al. Severe exercise hypoxaemia with normal or near normal X-rays: a feature of Pneumocystis carinii
20. World Health Organization. Management of Severe Malaria. A Practical Handbook
. 2nd ed. Geneva: WHO; 2000. Available at: http://mosquito.who.int/docs/hbsm.pdf
. Accessed June 11, 2004.
21. Plowe CV, Cortese JF, Djimde A, et al. Mutations in Plasmodium falciparum
dihydrofolate reductase and dihydropteroate synthase and epidemiologic patterns of pyrimethamine-sulfadoxine use and resistance. J Infect Dis
22. Amukoye E, Winstanley PA, Watkins WM, et al. Chlorproguanil-dapsone: an effective treatment for uncomplicated falciparum malaria. Antimicrob Agents Chemother
23. Gordon MA, Walsh AL, Chaponda M, et al. Bacteraemia and mortality among adult medical admissions in Malawi-predominance of non-typhi salmonellae and Streptococcus pneumoniae. J Infect
24. Anglaret X, Messou E, Ouassa T, et al. Pattern of bacterial diseases in a cohort of HIV-1 infected adults receiving cotrimoxazole prophylaxis in Abidjan, Cote d'Ivoire. AIDS
25. Mermin J, Lule J, Ekwaru JP, et al. Effect of co-trimoxazole prophylaxis on morbidity, mortality, CD4-cell count and viral load in HIV infection in rural Uganda. Lancet
26. Graham SM. HIV and respiratory infections in children. Curr Opin Pulm Med
27. Fisk DT, Meshnick S, Kazanjian PH. Pneumocystis carinii
pneumonia in patients in the developing world who have acquired immunodeficiency syndrome. Clin Infect Dis
28. Cruciani M, Marcati P, Malena M, et al. Meta-analysis of diagnostic procedures for Pneumocystis carinii pneumonia in HIV-1-infected patients. Eur Respir J
29. Holmes CB, Losina E, Walensky RP, et al. Review of human immunodeficiency virus type 1-related opportunistic infections in sub-Saharan Africa. Clin Infect Dis
30. Hunter G, Bagshawe AF, Baboo KS, et al. Intestinal parasites in Zambian patients with AIDS. Trans R Soc Trop Med Hyg
31. Cranendonk RJ, Kodde CJ, Chipeta D, et al. Cryptosporidium parvum
and Isospora belli
infections among patients with and without diarrhoea. East Afr Med J
32. Oldfield EC III, Fessel WJ, Dunne MW, et al. Once weekly azithromycin therapy for prevention of Mycobacterium avium
complex infection in patients with AIDS: a randomized, double-blind, placebo-controlled multicenter trial. Clin Infect Dis
33. Dunne MW, Bozzette S, McCutchan JA, et al. Efficacy of azithromycin in prevention of Pneumocystis carinii
pneumonia: a randomised trial. California Collaborative Treatment Group. Lancet
34. Frenck RW, Jr, Nakhla I, Sultan Y, et al. Azithromycin versus ceftriaxone for the treatment of uncomplicated typhoid fever in children. Clin Infect Dis
35. Anderson SL, Berman J, Kuschner R, et al. Prophylaxis of Plasmodium falciparum
malaria with azithromycin administered to volunteers. Ann Intern Med
36. Rakita RM, Jacques-Palaz K, Murray BE. Intracellular activity of azithromycin against bacterial enteric pathogens. Antimicrob Agents Chemother
37. Mason EO, Jr, Wald ER, Bradley JS, et al. Macrolide resistance among middle ear isolates of Streptococcus pneumoniae
observed at eight United States pediatric centers: prevalence of M and MLSB phenotypes. Pediatr Infect Dis J
38. Chintu C, Bhat GJ, Walker AS, et al. Co-trimoxazole as prophylaxis against opportunistic infections in HIV-infected Zambian children (CHAP): a double-blind randomized placebo-controlled trial. Lancet
Keywords:© 2005 Lippincott Williams & Wilkins, Inc.
HIV; Malawi; trimethoprim-sulfamethoxazole; prophylaxis; malaria