Reconsidering empirical cotrimoxazole prophylaxis for infants exposed to HIV infection


Remise en question du traitement préventif empirique par le cotrimoxazole chez les nourrissons exposés à l'infection par le VIH


Reconsideración de la profilaxis empírica con trimetoprim y sulfametoxazol en lactantes expuestos a la infección por VIH



Christopher J. GillI, 1; Lora L. SabinI; Joseph ThamII; Davidson H. HamerIII

IAssistant Professor, Department of International Health, Center for International Health and Development, Boston University School of Public Health, Boston, MA, USA (cgill@bu.edu)
IIAdjunct Assistant Professor, Department of International Health, Center for International Health and Development, Boston University School of Public Health, Boston, MA, USA
IIIAssociate Professor of Medicine and Nutrition, Tufts University School of Medicine and Freedman School of Nutrition, Boston, MA, USA




Infants with HIV infection are vulnerable to Pneumocystis carinii pneumonia (PCP) during their first year of life. WHO and the Joint United Nations Programme on HIV/AIDS now recommend that all children of HIV-positive mothers receive prophylactic cotrimoxazole against PCP from six weeks of age and continue this therapy until exposure through breast milk ceases and the infant is confirmed to be HIV-negative (rarely before one year of age). Empirical prophylaxis invokes a trade-off between possible benefit to the infant versus the risk of resistance to antibiotics and antimalarials. From a critical analysis of the literature, we offer a conceptual model demonstrating how, under certain circumstances, a policy of mass cotrimoxazole prophylaxis may be counterproductive.

Keywords: Trimethoprim-sulfamethoxazole combination/therapeutic use/adverse effects; Pneumonia, Pneumocystis/drug therapy; HIV infections/ complications; AIDS-related opportunistic infections/drug therapy; Disease transmission, Vertical; Drug resistance, Microbial; Sulfadoxine/ pharmacology; Pyrimethamine/pharmacology; Infant; Risk assessment; Meta-analysis; Models, Theoretical; Africa (source: MeSH, NLM).


Les nourrissons atteints d'infection par le VIH sont sensibles à la pneumocystose (pneumonie à Pneumocystis carinii) jusqu'à l'âge d'un an. L'OMS et le Programme commun des Nations Unies sur le VIH/SIDA recommandent actuellement d'administrer à tous les enfants nés de mère VIH-positive un traitement préventif par le cotrimoxazole contre la pneumocystose dès l'âge de six semaines et jusqu'à arrêt de l'exposition via le lait maternel et confirmation de la séronégativité de l'enfant vis-à-vis du VIH (rarement avant l'âge d'un an). Le traitement préventif empirique représente un compromis entre les avantages potentiels pour le nourrisson et le risque de résistance aux antibiotiques et aux antipaludiques. D'après une analyse critique des observations publiées, nous présentons un modèle théorique montrant comment, dans certaines circonstances, une politique de traitement préventif de masse par le cotrimoxazole peut être contre-productive.

Mots clés: Triméthoprime-sulfaméthoxazole, Association/usage thérapeutique/effets indésirables; Pneumocystose/chimiothérapie; HIV, Infection/ complications; Infections opportunistes liées SIDA/chimiothérapie; Transmission verticale maladie; Résistance microbienne aux médicaments; Sulfadoxine/pharmacologie; Pyriméthamine/pharmacologie; Nourrisson; Evaluation risque; Méta-analyse; Modèle théorique; Afrique (source: MeSH, INSERM).


Los lactantes infectados por el VIH son vulnerables a la neumonía por Pneumocystis carinii (NPC) durante el primer año de vida. En la actualidad, la OMS y el Programa Conjunto de las Naciones Unidas sobre el VIH/SIDA recomiendan que todos los niños cuyas madres son VIH-positivas reciban profilaxis con trimetoprim y sulfametoxazol contra la NPC desde las seis semanas de vida, y que sigan con este tratamiento hasta que cese la exposición a través de la leche materna y se confirme que el niño es VIH-negativo (raramente antes del año de vida). La profilaxis empírica se fundamenta en un equilibrio entre el posible beneficio para el lactante y el riesgo de resistencia a los antibióticos y los antimaláricos. Basándonos en un análisis crítico de la literatura médica, proponemos un modelo conceptual que demuestra que, en determinadas circunstancias, una política de profilaxis masiva con trimetoprim y sulfametoxazol puede ser contraproducente.

Palabras clave: Combinación trimetoprim-sulfametoxazol/uso terapéutico/efectos adversos; Neumonía por Pneumocystis carinii/quimioterapia; Infecciones por VIH/complicaciones; Infecciones oportunistas relacionadas con el SIDA/quimioterapia; Transmisión vertical de enfermedad; Resistencia microbiana a las drogas; Sulfadoxina/farmacología; Pirimetamina/farmacología; Lactante; Medición de riesgo; Meta-análisis; Modelos teóricos; África (fuente: DeCS, BIREME).




In 2000, WHO and the Joint United Nations Programme on HIV/AIDS (UNAIDS) issued a provisional recommendation: all African infants born to mothers infected with human immunodeficiency virus (HIV) should receive empirical cotrimoxazole (trimethoprim-sulfamethoxazole) prophylaxis against Pneumocystis carinii (now called Pneumocystis jirovecia) pneumonia (PCP) from six weeks of age (1). This proposal was made for the following reasons: the high number of deaths from PCP in infants with HIV infection; cotrimoxazole's efficacy for preventing PCP in adults; and the difficulty in determining HIV infection in exposed infants in poor countries. Nevertheless, cotrimoxazole prophylaxis based solely on HIV exposure, without confirmation of HIV infection status within the first month of life, is not the way this drug is usually used in developed countries and may have unforeseen consequences.

In developed countries, cotrimoxazole prophylaxis for infants exposed to HIV is also recommended from six weeks of age but usually discontinued once HIV infection has been excluded using a direct viral detection assay - ideally within the first few months of life. Direct viral assays are expensive, however, forcing poorer nations to rely on antibody-based HIV tests. Antibody assays are useful for diagnosing HIV in adults or older children but less reliable for neonates or infants. Persistence of maternal HIV antibodies gives high numbers of false positives until at least 12 months of age (2). Moreover, in countries where safe, culturally acceptable, feasible, affordable and sustainable alternatives to breastfeeding are lacking, up to 40% of infant HIV infections occur via breastfeeding (3). Because breastfeeding is nearly universal in developing countries, even tests that directly test for the presence of HIV rather than serologic responses would not resolve this dilemma unless applied after the child has been weaned. In this context, a negative viral assay would not alter the indication of cotrimoxazole until the child is no longer at risk of acquiring HIV through breast milk. Hence, infants exposed to HIV would continue prophylaxis until both weaned and shown to be antibody negative - conditions rarely satisfied before one year of age.

Between 600 000 and 800 000 infants with HIV infection are born each year, 90% of them in sub-Saharan Africa (4). This however represents only 13-30% of infants with peripartum HIV exposure (3). Consequently, at least 2 million infants annually would need prophylaxis under WHO/UNAIDS guidelines. Even though cotrimoxazole is inexpensive on a per-dose basis, providing so many children with a year's supply poses formidable financial and logistical challenges for Africa's poorest nations. Even assuming these challenges could be overcome, it remains unproven whether the potential benefits would exceed the potential risks of mass prophylaxis.

To explore these issues, we critically analyse the natural history of HIV and PCP in African infants, the evidence for cotrimoxazole's benefit in this group, and the known and theoretical risks of prophylaxis. From this analysis, we offer a theoretical model contrasting the risks and benefits of prophylaxis as an aid to framing future public policy responses to the problem of PCP in infants with HIV infection, and for defining future research priorities.


Natural history of HIV and PCP in young children

HIV infection proceeds more rapidly to AIDS in infants than adults. AIDS-related deaths peak during the first six months of life, followed by a more gradual progression to AIDS in the survivors (5). Natural history studies from Europe and the USA before the era of highly active antiretroviral therapy (HAART) are illustrative. Twenty six per cent of infected children developed clinical AIDS before reaching 12 months and 17% died (6); median survival was 38 months (7). Mortality rates in African infants are even higher. In Rwanda, mortality was 45% and 62% by age 2 years and 5 years, respectively (8). Of Malawian infants surviving their first year, 30% died by age 2 years and another 45% died by age 3 years (9). Overall, 90% of HIV-infected African children die before age 5 years (10).

In the pre-HAART era, HIV-infected infants in developed countries frequently died from PCP, particularly during their first six months of life (11) - a trend central to current prophylaxis guidelines. In African infants, the proportion of mortality caused by PCP versus other infectious diseases is unknown but probably small compared with the estimated annual 6 million deaths from malaria, diarrhoeal disease, and bacterial pneumonia (12). Autopsies and studies of African children infected with HIV and hospitalized for severe pneumonia demonstrate the importance of PCP (Table 1) but fail to indicate the proportion of all HIV-infected African infants who succumb to PCP compared with other causes. Natural history studies would provide these data but to date none of the prospective African trials have included PCP incidence data. In the absence of these data, we are forced to extrapolate the incidence data using pre-HAART studies from the USA, where first-year PCP incidence has been reported to range between 12.8% and 25% (7, 13). Based on these data, a plausible estimate of the annual incidence of infant PCP in Africa is between 76 000 (0.128 x 600 000) and 200 000 (0.25 x 800 000) cases.


Will cotrimoxazole prevent PCP in African children?

The survival benefit of cotrimoxazole is well established for adults with HIV infection living in developed countries but less well known for Africa (14). Only three randomized trials have studied primary prophylaxis with cotrimoxazole in Africa. The first was conducted among a population of adults with HIV infection receiving treatment for TB in Abidjan, Côte d'Ivoire. In this cohort, cotrimoxazole reduced all-cause mortality by 46% and hospitalizations by 43% compared with placebo (15). Cotrimoxazole users had reduced rates of enteritis (1.1% vs 3%, P = 0.04), bacteraemia (0.5% vs 3%, P = 0.01), and mycobacterial disease (3.9% vs 6.4%, P = 0.14) (15). The second study, also from Abidjan, was not limited to HIV patients with TB. Here, fewer patients receiving cotrimoxazole were hospitalized for pneumonia (0.7% vs 7%, P < 0.001), malaria (1% vs 9.6%, P < 0.01), or isosporiasis (0.7% vs 4%, P = 0.02). Prophylaxis had no significant impact on mortality (8.4% vs 9.2%, P = 0.5), but did cause increased rates of unexplained fever (13.6% vs 10%, P < 0.01) (16). While insufficiently powered to detect a mortality difference, the Kaplan-Meier curve provides little optimism that a clinically meaningful benefit was missed. The third study, from Dakar, Senegal, found no impact on either mortality or hospitalization, but was ultimately inconclusive, having been terminated early following the publication of the Abidjan studies (17).

While sought, PCP was not detected in either Abidjan study. This finding reinforces an emerging view that PCP may be a less important opportunistic pathogen for African adults with AIDS, compared with adults in developed nations. One possible explanation is that Africans with HIV succumb to TB or other infectious diseases long before their CD4 counts drop to a level where PCP becomes a realistic threat. If so, cotrimoxazole's main benefit for African adults may lie in preventing common bacterial and parasitic diseases.

The WHO/UNAIDS guidelines - including the recommendation for empiric cotrimoxazole for infants - emerged shortly after the publication of the Abidjan studies. This is remarkable given that neither study provided any information about the optimal use or efficacy of cotrimoxazole in children. While the evidence for PCP's importance in African children with HIV infection is robust (Table 1), evidence for cotrimoxazole's efficacy in this age group is weak. To date, no randomized trials of cotrimoxazole's efficacy have been conducted. Evidence supporting the use of cotrimoxazole in children with HIV infection rests principally on observational studies from the USA (18). Studies from developing countries consist of an ecological study from Thailand that documented a rise in community cotrimoxazole use over a period when PCP hospitalizations in children declined (19), and two retrospective analyses showing a reduced proportion of PCP in African infants with HIV infection and pneumonia among those using prophylaxis before hospitalization (20, 21). A third case-control study, however, found no reduction in PCP rates due to prophylaxis (22). One concern with this study's conclusion arises from the fact that cases were defined as children with HIV infection and confirmed PCP, and therefore may have included many children who had simply failed prophylaxis. It is unclear whether or not these children are representative of all children with HIV infection. Nor is it possible to infer how many children receiving prophylaxis remained well and avoided hospitalization.

Despite these ambiguities, the protective benefit found in studies from the USA establishes a strong moral imperative to provide cotrimoxazole to African infants either known or clinically suspected to have HIV infection - in the absence of data demonstrating its ineffectiveness or harm. Accepting the likelihood that prophylaxis will reduce PCP incidence in African infants with HIV infection does not mean that the benefits of mass prophylaxis will outweigh its risks when applied empirically to a population in which children infected with HIV will be a minority.


What are the risks of cotrimoxazole prophylaxis?

Rash, fever, and anaemia are all common during cotrimoxazole prophylaxis (23), and particularly in people with HIV infection (24). Cotrimoxazole appears better tolerated in children (25). Rare life-threatening side-effects, including severe hepatitis or Stevens-Johnson syndrome, can also occur though their incidence is difficult to predict from small clinical studies.

Antimicrobial resistance is arguably the most important adverse consequence of prophylaxis. Children taking antibiotics for a variety of chronic conditions - sickle-cell anaemia, recurrent urinary tract infections, and chronic otitis media - quickly become colonized with drug-resistant bacteria (26-28). Adults with HIV infection taking prophylactic cotrimoxazole also experience drug resistance (29). The list of pathogens whose epidemiology could be altered by cotrimoxazole exposure includes Streptococcus pneumoniae, Staphylococcus aureus, Haemophilus influenzae, enteropathogens (Shigella spp., Salmonella spp., Isospora belli), and Neisseria spp.

Resistant pathogens spread readily from infants to close contacts (30, 31), thereby extending this consequence of prophylaxis to families and ultimately the community at large. Rising cotrimoxazole resistance rates could impair the effectiveness of the WHO's Integrated Management of Childhood Infections (IMCI) guidelines, which currently recommend cotrimoxazole as first-line therapy for a variety of childhood infections (32).

The genetics of drug resistance adds another layer of complexity. Transmissible genetic elements often incorporate multiple antibiotic-resistance genes. For example, prior cotrimoxazole use has been identified as a risk factor for colonization with penicillin resistant pneumococci (33). Thus, our mental calculus of cotrimoxazole's impact extends beyond cotrimoxazole resistance alone, to include also penicillins, cephalosporins, macrolides, tetracyclines, and chloramphenicol (28, 34, 35).

Widespread cotrimoxazole use could also foster sulfadoxine-pyrimethamine resistance in Plasmodium falciparum, an organism which kills approximately one million children each year - the most attributable to any single childhood pathogen (36). With the loss of chloroquine as a reliable treatment for malaria, sulfadoxine-pyrimethamine has emerged as the first-line antimalarial throughout Africa. Rising resistance rates meant it was abandoned in the Amazon region of Brazil within a decade of introduction (37), and sulfadoxine-pyrimethamine may be a temporary solution to the problem of chloroquine resistance in Africa as well. Inexpensive and non-toxic alternatives are currently lacking, making it essential to maximize sulfadoxine-pyrimethamine's useful life (38).

Cotrimoxazole and sulfadoxine-pyrimethamine are close pharmacologically and share extensive in vitro cross-resistance (39, 40). Indeed, while cotrimoxazole is not typically considered an antimalarial, comparative studies show it to be equally effective as sulfadoxine-pyrimethamine for treating falciparum malaria (41, 42). As yet, there is no convincing evidence that this in vitro cross-tolerance will translate into clinical failures, though interestingly, sulfadoxine-pyrimethamine has been linked with increased rates of cotrimoxazole-resistant pneumococci (43). But this is not reassuring given the absence of prospective studies addressing the question. With the stakes that are involved, it would be irresponsible not to question whether the efficacy of sulfadoxine-pyrimethamine, and potentially other sulfonamide-containing antimalarials, could be reduced by mass cotrimoxazole prophylaxis.

Another concern is whether cotrimoxazole prophylaxis could impede the acquisition of natural malaria immunity by infants. Because of maternal antibodies, infants in areas of high malaria incidence are often born with immunity to malaria (44). As this passive immunity wanes, natural immunity emerges after exposure to infectious mosquitoes. Cotrimoxazole reduces both clinical malaria and asymptomatic parasitaemia and, in so doing, could hypothetically attenuate the normal acquisition of immunity. This creates an ethical dilemma since HIV-exposed infants discontinuing prophylaxis, after a negative HIV test at one year of age, might be more vulnerable to severe malaria than if they had never received prophylaxis.


Do the benefits exceed the risks of empiric prophylaxis?

Table 2 summarizes some of the most important risks and benefits of prophylaxis. Though their magnitudes are unknown, conceptualizing their interactions may still allow us to derive insights into how these risks and benefits compete, and identify areas where additional data could support a formal quantitative analysis.

Fig. 1 presents a basic conceptual model of the relationship between the risks and benefits of mass empiric prophylaxis when applied to a mixed population of HIV-positive and HIV-negative children. For simplicity we consider both risks and benefits to exist on a universal scale, such that 1 risk unit is equivalent in magnitude to 1 benefit unit. Because prophylaxis chiefly aims to prevent PCP, the benefit of empiric prophylaxis increases in proportion to HIV's prevalence in the target population (if prevalence were 100%, benefits would be maximal). Setting aside the less worrisome risk of cotrimoxazole side-effects, there is no compelling reason to expect that the more important risks of prophylaxis (microbial resistance) will be influenced by HIV infection per se. Therefore, the risk line should remain constant regardless of HIV prevalence. At some degree of HIV prevalence, the benefits and risks lines intersect and balance. When HIV prevalence exceeds this point, empiric prophylaxis is warranted; below this prevalence level, mass empiric prophylaxis is unjustified since the risks exceed benefits.

Our goal is to reduce PCP - and increase longevity - in HIV-positive children. Applying our current diagnostic limitations we do not know which HIV-exposed children are born infected or which will become infected during their first year. We will assume that the mother's HIV status will be ascertained during antenatal clinic voluntary counselling and testing (though in reality, this number will be reduced by those women who choose not to learn their HIV status). Currently, our options for preventing PCP in these children are either to use 4 million doses of nevirapine (half to the 2 million HIV-exposed infants and half to their HIV-positive mothers) or 750 million doses of cotrimoxazole to treat these children for one year each (2 million children exposed to HIV x 365 daily doses). Obviously, these are not mutually exclusive, and would probably be combined in the real world. This is logically appealing: nevirapine reduces the risk of HIV infection - and indirectly also PCP - in the majority, while cotrimoxazole protects the unlucky HIV-positive minority directly from PCP.

Careful examination of Fig. 1 reveals why this combined approach could have a paradoxical adverse effect. Let us assume that, in the absence of any intervention, this population's HIV prevalence places us precisely where the lines of risk and benefit intersect. If this population of HIV-exposed children receives nevirapine at birth, HIV prevalence will decline, shifting us to the left on the x axis. Even though nevirapine reduces the number of children at risk of PCP our ignorance of which children still become infected with HIV still compels us to provide cotrimoxazole presumptively to all 2 million. The result: even though cotrimoxazole's efficacy for preventing PCP for an individual HIV-positive infant remains constant, at a population level the effectiveness of mass prophylaxis deteriorates and the strategy becomes unjustifiable because the risks now exceed the benefits.

The purpose of this subjective exercise is not to suggest that this scenario would necessarily be the result of mass prophylaxis. Indeed our current knowledge of the magnitude of the risks involved is simply inadequate to know at what HIV prevalence the risk and benefits lines intersect (or even if these theoretical lines are straight or curved). But this is precisely the point - we don't know where this critical threshold lies, so why is it acceptable to assume that we are safely above it?

Without providing a quantitative measure of these outcomes, our thought experiment still identifies two problems deriving from our ignorance of which children are actually HIV-positive. First, because the same number of cotrimoxazole doses are used to prevent fewer cases of PCP, nevirapine reduces the additional benefit of cotrimoxazole prophylaxis. Second, the prior use of nevirapine increases the ratio of risk to benefit of subsequent cotrimoxazole prophylaxis. Accordingly, we suggest that improving strategies for preventing postnatal HIV transmission and for identifying HIV-infected children earlier (so that cotrimoxazole prophylaxis can be appropriately applied to those who will benefit most) may prove a better investment of limited resources than mass empiric prophylaxis.



Ultimately, the merit of a mass empiric prophylaxis strategy rests on the degree of risk it entails. Mass prophylaxis will inevitably increase rates of drug-resistant bacteria and malaria in cotrimoxazole users - and likely their communities at large - but how quickly, and to what extent? Quantifying the magnitude of this risk is essential and we urge countries that choose to implement WHO's strategy to initiate their programmes in parallel with a systematic means for prospectively monitoring the microbiological effects of mass prophylaxis in their populations. Given their unambiguous contribution to paediatric disease, monitoring the resistance patterns of S. pneumoniae and P. falciparum longitudinally would seem a logical starting point.

In their original declaration, WHO/UNAIDS acknowledged their strategy's potential for risk, but concluded that action was preferable to further debate (1). This sentiment is quite understandable given the magnitude of the humanitarian disaster unfolding before us. However, without a safety net of surveillance, it is critical that we better understand the consequences of this strategy, rather than taking a leap of faith.



We wish to thank the following people for their thoughtful advice during the preparation of this manuscript: Jonathan Simon, Donald Thea, Christine Ayash, and Matthew Fox (Center for International Health and Development); Dr Ira Wilson (Health Institute, Tufts-New England Medical Center); Dr Stephen Graham (Malawi-Liverpool-Wellcome Trust Clinical Research Programme and Department of Pediatrics, University of Malawi); and Dr Shamim Qazi (WHO).

This paper was made possible through generous support provided by the United States Agency for International Development (Applied Research on Child Health (ARCH) grant: HRN-A-00-96-90010-00), and by a CDC/ASPH/ATDSR cooperative agreement grant (Zambia-Boston University Malaria Project: S1954-21/21).

The grants cited provided salary support to conduct applied research in developing nations, but did not otherwise have any input into the content of this manuscript.

Conflicts of interest: none declared.



1. UNAIDS. UNAIDS/WHO hail consensus on use of cotrimoxazole for prevention of HIV-related infections in Africa, 2000.          http://www.unaids. org/whatsnew/press/eng/pressarc00/geneva050400.html

2. Moodley D, Bobat RA, Coutsoudis A, Coovadia HM. Predicting perinatal human immunodeficiency virus infection by antibody patterns. Pediatric Infectious Disease 1995;14:850-2.        

3. Santmyire BR. Vertical transmission of HIV from mother to child in sub-Saharan Africa: modes of transmission and methods for prevention. Obstetrical and Gynecological Survey 2001;56:306-12.        

4. Berer M. HIV/AIDS, sexual and reproductive health at AIDS 2000, Durban. Reproductive Health Matters 2000;8:160-9.        

5. Duliege AM, Messiah A, Blanche S, Tardieu M, Griscelli C, Spira A. Natural history of human immunodeficiency virus type 1 infection in children: prognostic value of laboratory tests on the bimodal progression of the disease. Pediatric Infectious Disease 1992;11:630-5.        

6. Children born to women with HIV-1 infection: natural history and risk of transmission. European Collaborative Study. Lancet 1991;337:253-60.        

7. Scott GB, Hutto C, Makuch RW, Mastrucci MT, O'Connor T, Mitchell CD, et al. Survival in children with perinatally acquired human immunodeficiency virus type 1 infection. New England Journal of Medicine 1989;321:1791-6.        

8. Spira R, Lepage P, Msellati P, Van De Perre P, Leroy V, Simonon A, et al. Natural history of human immunodeficiency virus type 1 infection in children: a five-year prospective study in Rwanda. Mother-to-Child HIV-1 Transmission Study Group. Pediatrics 1999;104:e56.        

9. Taha TE, Kumwenda NI, Broadhead RL, Hoover DR, Graham SM, Van Der Hoven L, et al. Mortality after the first year of life among human immunodeficiency virus type 1-infected and uninfected children. Pediatric Infectious Disease 1999;18:689-94.        

10. Boerma JT, Nunn AJ, Whitworth JA. Mortality impact of the AIDS epidemic: evidence from community studies in less developed countries. AIDS 1998;12 Suppl 1:S3-14.        

11. Simonds RJ, Oxtoby MJ, Caldwell MB, Gwinn ML, Rogers MF. Pneumocystis carinii pneumonia among US children with perinatally acquired HIV infection. JAMA 1993;270:470-3.        

12. Murray CJ, Lopez AD. Evidence-based health policy - lessons from the Global Burden of Disease Study. Science 1996;274:740-3.        

13. Johnson JP, Nair P, Hines SE, Seiden SW, Alger L, Revie DR, et al. Natural history and serologic diagnosis of infants born to human immunodeficiency virus-infected women. American Journal of Diseases of Children 1989;143:1147-53.        

14. Mallolas J, Zamora L, Gatell JM, Miro JM, Vernet E, Valls ME, et al. Primary prophylaxis for Pneumocystis carinii pneumonia: a randomized trial comparing cotrimoxazole, aerosolized pentamidine and dapsone plus pyrimethamine. AIDS 1993;7:59-64.        

15. Wiktor SZ, Sassan-Morokro M, Grant AD, Abouya L, Karon JM, Maurice C, et al. Efficacy of trimethoprim-sulphamethoxazole prophylaxis to decrease morbidity and mortality in HIV-1-infected patients with tuberculosis in Abidjan, Côte d'Ivoire: a randomised controlled trial. Lancet 1999;353:1469-75.        

16. Anglaret X, Chene G, Attia A, Toure S, Lafont S, Combe P, et al. Early chemoprophylaxis with trimethoprim-sulphamethoxazole for HIV-1 infected adults in Abidjan, Côte d'Ivoire: a randomised trial. Cotrimo-CI Study Group. Lancet 1999;353:1463-8.        

17. Maynart M, Lievre L, Sow PS, Kony S, Gueye NF, Bassene E, et al. Primary prevention with cotrimoxazole for HIV-1-infected adults: results of the pilot study in Dakar, Senegal. Journal of Acquired Immune Deficiciency Syndromes 2001;26:130-6.        

18. Thea DM, Lambert G, Weedon J, Matheson PB, Abrams EJ, Bamji M, et al. Benefit of primary prophylaxis before 18 months of age in reducing the incidence of Pneumocystis carinii pneumonia and early death in a cohort of 112 human immunodeficiency virus-infected infants. New York City Perinatal HIV Transmission Collaborative Study Group. Pediatrics 1996;97:59-64.        

19. Chokephaibulkit K, Chuachoowong R, Chotpitayasunondh T, Chearskul S, Vanprapar N, Waranawat N, et al. Evaluating a new strategy for prophylaxis to prevent Pneumocystis carinii pneumonia in HIV-exposed infants in Thailand. Bangkok Collaborative Perinatal HIV Transmission Study Group. AIDS 2000;14:1563-9.        

20. Zar HJ, Hanslo D, Tannenbaum E, Klein M, Argent A, Eley B, et al. Aetiology and outcome of pneumonia in human immunodeficiency virus-infected children hospitalized in South Africa. Acta Paediatrica 2001;90:119-25.        

21. Ruffini DD, Madhi SA. The high burden of Pneumocystis carinii pneumonia in African HIV-1-infected children hospitalized for severe pneumonia. AIDS 2002;16:105-12.        

22. Madhi SA, Cutland C, Ismail K, O'Reilly C, Mancha A, Klugman KP. Ineffectiveness of trimethoprim-sulfamethoxazole prophylaxis and the importance of bacterial and viral coinfections in African children with Pneumocystis carinii pneumonia. Clinical Infectious Diseases 2002;35:1120-6.        

23. May T, Beuscart C, Reynes J, Marchou B, Leclercq P, Borsa Lebas F, et al. Trimethoprim-sulfamethoxazole versus aerosolized pentamidine for primary prophylaxis of Pneumocystis carinii pneumonia: a prospective, randomized, controlled clinical trial. LFPMI Study Group. Ligue Française de Prévention des Maladies Infectieuses. Journal of Acquired Immune Deficiency Syndromes 1994;7:457-62.        

24. Martin MA, Cox PH, Beck K, Styer CM, Beall GN. A comparison of the effectiveness of three regimens in the prevention of Pneumocystis carinii pneumonia in human immunodeficiency virus-infected patients. Archives of Internal Medicine 1992;152:523-8.        

25. Fisher RG, Nageswaran S, Valentine ME, McKinney RE, Jr. Successful prophylaxis against Pneumocystis carinii pneumonia in HIV-infected children using smaller than recommended dosages of trimethoprim- sulfamethoxazole. AIDS Patient Care and STDs 2001;15:263-9.        

26. Anglin DL, Siegel JD, Pacini DL, Smith SJ, Adams G, Buchanan GR. Effect of penicillin prophylaxis on nasopharyngeal colonization with Streptococcus pneumoniae in children with sickle cell anemia. Journal of Pediatrics 1984;104:18-22.        

27. Sandock DS, Gothe BG, Bodner DR. Trimethoprim-sulfamethoxazole prophylaxis against urinary tract infection in the chronic spinal cord injury patient. Paraplegia 1995;33:156-60.        

28. Brook I, Gober AE. Prophylaxis with amoxicillin or sulfisoxazole for otitis media: effect on the recovery of penicillin-resistant bacteria from children. Clinical Infectious Diseases 1996;22:143-5.        

29. Meynard JL, Barbut F, Blum L, Guiguet M, Chouaid C, Meyohas MC, et al. Risk factors for isolation of Streptococcus pneumoniae with decreased susceptibility to penicillin G from patients infected with human immuno- deficiency virus. Clinical Infectious Diseases 1996;22:437-40.        

30. Reichler MR, Allphin AA, Breiman RF, Schreiber JR, Arnold JE, McDougal LK, et al. The spread of multiply resistant Streptococcus pneumoniae at a day care center in Ohio. Journal of Infectious Diseases 1992;166:1346-53.        

31. Melander E, Ekdahl K, Hansson HB, Kamme C, Laurell M, Nilsson P, et al. Introduction and clonal spread of penicillin- and trimethoprim/ sulfamethoxazole-resistant Streptococcus pneumoniae, serotype 9V, in southern Sweden. Microbial Drug Resistance 1998;4:71-8.        

32. World Health Organization: Improving Child Health IMCI: the integrated approach. Geneva Switzerland: World Health Organization; 1996. WHO/ CHD/97.12 Rev. 1.        

33. Feikin DR, Davis M, Nwanyanwu OC, Kazembe PN, Barat LM, Wasas A, et al. Antibiotic resistance and serotype distribution of Streptococcus pneumoniae colonizing rural Malawian children. Pediatric Infectious Disease 2003;22:564-7.        

34. Courvalin P, Carlier C. Transposable multiple antibiotic resistance in Streptococcus pneumoniae. Molular and General Genetics 1986;205:291-7.        

35. Smith JT, Lewin CS. Mechanisms of antimicrobial resistance and implications for epidemiology. Veterinary Microbiology 1993;35:233-42.        

36. Murray CJL, Lopez A., Mathers C., Stein C. The Global Burden of Diseases 2000 Project: aims, methods and data sources. Geneva: World Health Organization; 2001. Global Programme on Evidence for Health Policy Discussion Paper number 36.        

37. Vasconcelos KF, Plowe CV, Fontes CJ, Kyle D, Wirth DF, Pereira da Silva LH, et al. Mutations in Plasmodium falciparum dihydrofolate reductase and dihydropteroate synthase of isolates from the Amazon region of Brazil. Memorias do Instituto Oswaldo Cruz 2000;95:721-8.        

38. Kublin JG, Dzinjalamala FK, Kamwendo DD, Malkin EM, Cortese JF, Martino LM, et al. Molecular markers for failure of sulfadoxine-pyrimethamine and chlorproguanil-dapsone treatment of Plasmodium falciparum malaria. Journal of Infectious Diseases 2002;185:380-8.        

39. Iyer JK, Milhous WK, Cortese JF, Kublin JG, Plowe CV. Plasmodium falciparum cross-resistance between trimethoprim and pyrimethamine. Lancet 2001;358:1066-7.        

40. Whitty CJ, Jaffar S. Plasmodium falciparum cross resistance. Lancet 2002;359:80.        

41. Omar SA, Bakari A, Owiti A, Adagu IS, Warhurst DC. Co-trimoxazole compared with sulfadoxine-pyrimethamine in the treatment of uncomplicated malaria in Kenyan children. Transactions of the Royal Society of Tropical Medicine and Hygiene 2001;95:657-60.        

42. Daramola OO, Alonso PL, O'Dempsey TJ, Twumasi P, McArdle TF, Greenwood BM. Sensitivity of Plasmodium falciparum in The Gambia to co- trimoxazole. Transactions of the Royal Society of Tropical Medicine and Hygiene 1991;85:345-8.        

43. Feikin DR, Dowell SF, Nwanyanwu OC, Klugman KP, Kazembe PN, Barat LM, et al. Increased carriage of trimethoprim/sulfamethoxazole-resistant Streptococcus pneumoniae in Malawian children after treatment for malaria with sulfadoxine/pyrimethamine. Journal of Infectious Diseases 2000;181:1501-5.        

44. Snow RW, Omumbo JA, Lowe B, Molyneux CS, Obiero JO, Palmer A, et al. Relation between severe malaria morbidity in children and level of Plasmodium falciparum transmission in Africa. Lancet 1997;349:1650-4.        

45. Lucas SB, Peacock CS, Hounnou A, Brattegaard K, Koffi K, Honde M, et al. Disease in children infected with HIV in Abidjan, Côte d'Ivoire. BMJ 1996;312:335-8.        

46. Ikeogu MO, Wolf B, Mathe S. Pulmonary manifestations in HIV seropositivity and malnutrition in Zimbabwe. Archives of Disease in Childhood 1997;76:124-8.        

47. Jeena PM, Coovadia HM, Chrystal V. Pneumocystis carinii and cytomegalovirus infections in severely ill, HIV-infected African infants. Annals of Tropical Paediatrics 1996;16:361-8.        

48. Nathoo KJ, Gondo M, Gwanzura L, Mhlanga BR, Mavetera T, Mason PR. Fatal Pneumocystis carinii pneumonia in HIV-seropositive infants in Harare, Zimbabwe. Transactions of the Royal Society of Tropical Medicine and Hygiene 2001;95:37-9.        

49. Chintu C, Mudenda V, Lucas S, Nunn A, Lishimpi K, Maswahu D, et al. Lung diseases at necropsy in African children dying from respiratory illnesses: a descriptive necropsy study. Lancet 2002;360:985-90.        

50. Ansari NA, Kombe AH, Kenyon TA, Mazhani L, Binkin N, Tappero JW, et al. Pathology and causes of death in a series of human immunodeficiency virus-positive and -negative pediatric referral hospital admissions in Botswana. Pediatric Infectious Disease 2003;22:43-7.        

51. Kamiya Y, Mtitimila E, Graham SM, Broadhead RL, Brabin B, Hart CA. Pneumocystis carinii pneumonia in Malawian children. Annals of Tropical Paediatrics 1997;17:121-6.        

52. Graham SM, Mtitimila EI, Kamanga HS, Walsh AL, Hart CA, Molyneux ME. Clinical presentation and outcome of Pneumocystis carinii pneumonia in Malawian children. Lancet 2000;355:369-73.        



Submitted: 26 August 03
Final revised version received: 7 November 03
Accepted: 8 December 03



1 Correspondence should be sent to this author.
a Stringer JR, Beard CB, Miller RF, Wakefield AE. A New Name (Pneumocystis jiroveci) for Pneumocystis from Humans. Emerging Infectious Diseases 2002; 8:891-6.

World Health Organization Genebra - Genebra - Switzerland
E-mail: bulletin@who.int