Review: HIV infection and tropical parasitic diseases – deleterious interactions in both directions?

Introduction

HIV infection was recognized in sub-Saharan Africa 16 years ago and today 32 million people are estimated to be infected with HIV in the area alone. Given the high prevalence of parasitic infections in the tropics, HIV infection and parasitic diseases should frequently overlap. Based on the immunological characteristics of HIV infection/AIDS, it may be assumed that the natural history of parasitic infections, the majority of which require a competent immune response to contain the pathogen, will be altered in an unfavourable way. Concurrent HIV infection may also have an impact on the diagnosis of parasitic diseases as well as on the response of patients to antiparasitic treatment.

While the consequences of HIV on parasitic infections have attracted much interest, the impact parasites may have on the replication of this virus and eventually on progression towards AIDS gained attention only recently (Bentwich et al. 1995). However, there is now a considerable amount of evidence, direct as well as circumstantial, indicating that chronic parasitic diseases may influence the natural history of HIV infection in a deleterious way. This article reviews current knowledge and hypotheses on the bidirectional interactions between HIV/AIDS and major parasitic diseases prevailing in a tropical environment.

Malaria

Malaria is the most common parasitic disease of the tropics. At least 500 million people are infected in sub-Saharan Africa alone each year, hence co-infection with malaria parasites and HIV has to occur frequently. Earlier investigations concluded that there was no evidence for an effect of HIV on the prevalence of falciparum malaria, the susceptibility to severe disease or the response to treatment (Nguyen-Dinh et al. 1987; Allen et al. 1991).

However, in a cohort study conducted over 8 years in rural Uganda, parasitaemia and clinical malaria were significantly more common in HIV-infected individuals. HIV-infected patients had twice as many episodes of symptomatic parasitaemia than non-infected persons. Parasite density and clinical malaria both were significantly associated with falling CD4 counts and with advancing HIV disease (Whitworth et al. 2000).

Recently, cause–effect interactions between HIV and malaria were also shown to occur in pregnant women (Taha et al. 1994; Steketee et al. 1996a, b; Chandramohan & Greenwood 1998; Verhoeff et al. 1999). A cross-sectional study of pregnant women in Malawi clearly indicated an increased incidence of malaria in HIV-infected pregnant women. The higher incidence was evident not only in primigravidae but in all parities (Verhoeff et al. 1999). Besides, HIV infection seemed to impair the development of parity-related antimalarial immunity which, in holo-endemic areas, parallels increasing gestational age and the number of pregnancies (Verhoff et al. 1999). This implies that in HIV-infected multigravidae, anaemia and low birthweight, conditions which are usually controlled by the lower susceptibility for severe malaria in this group, will also be more frequent. Consequently, postnatal mortality is higher if the mother is infected with both malaria and HIV, than in infants of mothers with just one infection (Bloland et al. 1995).

In vitro, production of interferon (IFN)-γ by intervillous blood mononuclear cells of the placenta protects against placental malaria. However, IFN-γ as well as interleukin (IL)-4 and IL-10 responses by HIV-infected intervillous mononuclear cells are reduced while production of tumour necrosis factor (TNF)-α is increased in HIV-infected women (Moore et al. 2000). Cytokine production was dramatically reduced in women with CD4 T-cell counts of <500/μl. Thus, HIV-mediated impairment of cytokines relevant for mounting an effective immune response against Plasmodium falciparum may contribute to the increased susceptibility of HIV-positive pregnant women to malaria (Moore et al. 1999, 2000).

It is conceivable that enhancement of vertical transmission of HIV in P. falciparum infected women increases infant mortality (Bloland et al. 1995). This assumption has gained experimental support recently. It has been shown that malaria antigen-induced immune activation of peripheral mononuclear cells of normal donors up-regulates HIV replication 10–100-fold (Xiao et al. 1998). This increase of HIV-1 replication is because of an increase in viral messenger RNA (mRNA) expression and the activation of the long-terminal repeat-directed viral transcription. Also, malarial antigen-dependent viral replication is mediated through an up-regulation of TNF-α production (Xiao et al. 1998). In mice, infection with various Plasmodium spp. triggers a dramatic elevation in viral p24 production which returns to baseline after 2 weeks. The p24 production stems from an interaction of malaria-specific T lymphocytes with HIV-expressing antigen-presenting cells as demonstrated by a markedly reduced HIV expression after CD4 T-cell depletion (Freitag et al. 2001).

These experimental findings fit into the epidemiological observation that persons with malaria have a higher HIV-1 RNA burden than individuals without malaria and that a significant reduction of the viral load is achieved after antimalarial therapy (Hoffman et al. 1999).

Leishmaniasis

About 2 million new cases of leishmaniasis occur each year, while 15 million people are estimated to be infected (Desjeux 1999). Leishmania may cause a spectrum of diseases from self-healing ulcers to disseminated and fatal infections, depending on the Leishmania species involved and on the host's immune response.

Leishmania/HIV co-infection has emerged as a result of the increasing overlap of areas in which HIV or leishmania occur, particularly in eastern Africa, India, Brazil and Europe, and cases have so far been reported from 38 countries (Berhe et al. 1995; WHO 1996, 1999; Alvar et al. 1997; Ermak et al. 1997; Desjeux 1999; Rosenthal et al. 2000a). Although cutaneous manifestations have been described, the predominant clinical entity occurring in co-infections is visceral leishmaniasis (VL). As parasitaemia is frequent in Leishmania/HIV co-infected patients, the efficiency of transmission through sandflies is likely to be increased which, in turn, should also put other population groups at risk.

Multiple immunological mechanisms mediate the impact HIV infection has on VL and vice versa. Both pathogens infect monocytes/macrophages, may establish a latent infection or may accelerate their intracellular multiplication (Bernier et al. 1998; Wolday et al. 1999). An effective T helper (Th)1-cell mediated immune response is required to control leishmanial infections and therefore parasites disseminate with deteriorating T-cell competence. Th1-cell competence is defective in HIV infection and provides a favourable environment for the manifestation of primary VL or for the reactivation of latent leishmanial infection acquired many years earlier. Consistently, VL usually manifests itself as an opportunistic infection at CD4 cell counts below 200/ml in HIV-infected individuals (WHO 1996, 1999).

In vitro, intracellular growth of Leishmania in macrophages is increased in macrophages initially infected with HIV (Wolday et al. 1999). Uncontrolled multiplication and dissemination of the parasites is reflected by the peripheral parasitaemia found in more than 50% of co-infected persons and by the higher frequency of unusual body compartments to which the parasites have disseminated (Rosenthal et al. 2000b).

On the other hand VL goes along with a shift towards a Th2-cell mediated response and corresponding cytokine production (Zwingenberger et al. 1990). In co-infected individuals plasma levels of Th2 cytokines such as IL-4, IL-6 and IL-10 are significantly higher than in individuals with HIV or leishmanial infection alone (Cacopardo et al. 1996). These cytokines impair cellular cytotoxicity and hamper the macrophage's capability to kill parasites.

The major surface protein of Leishmania, a lipophosphoglycan, up-regulates HIV replication in monocytic cells as well as in CD4+ T cells (Bernier et al. 1998). Furthermore cytokines, particularly TNF-α and IL-6, which are generated via the typical polyclonal B-cell activation accompanying VL, are thought to play an important role in the induction of HIV expression (Wolday et al. 1999).

Clinical diagnosis of VL in leishmania/HIV co-infected persons is particularly difficult (WHO 1996, 1999; Alvar et al. 1997; Campino et al. 1997; Laguna et al. 1997; Agostini et al. 1998; Lopez-Velez et al. 1998; Rosenthal et al. 2000a; Pintado et al. 2001). Only half of the HIV-infected patients – as opposed to 95% of non-HIV infected – exhibit the characteristic clinical pattern, namely fever, splenomegaly and hepatomegaly (WHO 1996, 1999; Alvar et al. 1997; Desjeux 1999; Pintado et al. 2001). With ongoing immunosuppression ectopic localizations of parasite multiplication become common (Rosenthal et al. 2000a, b). Gastrointestinal (Rosenthal et al. 1988; Datry et al. 1990; Bailey et al. 1994; Cortes et al. 1997; Perez-Molina et al. 1997), laryngeal (Gonzales-Anglada et al. 1994; Grant et al. 1994), pulmonal and peritoneal involvement have been reported (Munoz et al. 1997). Single and multiple cutaneous forms (Cnudde et al. 1994; Dauden et al. 1996; Mattos et al. 1998; Souza et al. 1998; Colebunders et al. 1999) and/or mucosal and mucocutaneous lesions (Miralles et al. 1994; Grasa et al. 2000) have also been described in AIDS patients worldwide.

About 50% of the patients do not show detectable levels of antibodies (WHO 1996, 1999; Medrano et al. 1998; Pintado et al. 2001). However, the peripheral parasitaemia displayed by many HIV co-infected individuals allows the detection of parasites from the blood in about 50% of the cases. Cultures and polymerase chain reaction (PCR) of a buffy-coat preparation are positive in 70% and up to 100%, respectively (WHO 1996, 1999; Dereure et al. 1998).

Between 50% and 100% of co-infected patients initially respond to treatment with pentavalent antimonials (Sb^v) or Amphotericin B in its standard preparation or in its liposomal form (AmBisome^R) (Gradoni et al. 1995; WHO 1996, 1999; Laguna et al. 1999; Pintado et al. 2001). Side-effects are more frequent and pronounced (WHO 1996; Delgado et al. 1999). Relapses occur irrespective of the drug used in 25–80% of co-infected individuals (Davidson et al. 1994, 1996; Davidson & Russo 1994; Russo et al. 1996; Lopez-Velez et al. 1998; Laguna et al. 1999). The mortality rate is significantly higher and the mean survival time (13 months) is significantly shorter in co-infected than in immunocompetent persons. (WHO 1999).

The pre-treatment HIV viral load correlates with the response to anti-leishmanial therapy (Berhe et al. 1999). Significant reductions of the incidence of VL were observed in Italy, France and Spain after the introduction of antiretroviral therapy (Tumbarello et al. 2000; Pintado et al. 2001; Rosenthal et al. 2001). However, antiretroviral therapy alone, assumed to stabilize and eventually to increase the CD4 cell count, does not prevent relapses in patients with VL (Villanueva et al. 2000; Pizzuto et al. 2001). Taken together, Leishmania and HIV mutually influence each other leading to an uncontrolled multiplication of the pathogens, profound immunosuppression and eventually to fatal disease progression.

Human African Trypanosomiasis

Sleeping sickness, caused by Trypanosoma brucei gambiense and T. b. rhodesiense and transmitted by the tsetse-fly, puts about 60 million people at risk of infection in 36 countries of sub-Saharan Africa. An estimated 300 000 cases have been reported annually in the past years. Infections occur in limited foci and are sporadic in the case of T. b. gambiense, whereas in T. b. rhodesiense sleeping sickness occurs in epidemics. These characteristics of African trypanosomiasis make it difficult to find evidence for an impact of HIV infection on the prevalence of sleeping sickness and impede the generalization of findings in one or the other species.

It is therefore not surprising that in cross-sectional studies no associations between HIV infection and T. b. gambiense infection were found (Noireau et al. 1987; Louis et al. 1991; Pepin et al. 1992; Meda et al. 1995). However, in one of the studies, HIV-1 positive patients were significantly more likely to relapse after treatment with eflornithine DFMO, indicating that HIV-positive persons may be at a higher risk for treatment failure than HIV seronegatives (Pepin et al. 1992). Moreover, of 18 patients treated with melarsoprol, the four HIV-positive patients either died or had an unfavourable outcome while the 14 HIV-negative patients recovered completely (Blum et al. 2001).

Chagas' disease

Chagas' disease, the American trypanosomiasis caused by the protozoan Trypanosoma cruzi, is a health threat to 17 million people in Latin America, considerably less than a decade ago. Control of Chagas' disease has progressed and the southern cone states, Uruguay, Argentina, Chile and large parts of Brazil are now considered free of vectorial transmission (WHO 2000). The disease is characterized by two phases, an acute phase with high parasitaemia of T. cruzi and a chronic phase usually without detectable parasitaemia. In analogy to the mechanisms in Leishmania/HIV co-infections, the ongoing immunosuppression allows parasites to multiply, particularly in the chronic stage (Perez-Ramirez et al. 1999).

While central nervous system (CNS) involvement is never observed in chronic Chagas' disease, it occurs in immunocompromised persons as a result of reactivation of dormant T. cruzi infection acquired years earlier (Ferreira et al. 1997; Pacheco et al. 1998). In HIV-infected patients CNS manifestations are acute fatal meningoencephalitis (Ferreira et al. 1991; Rosemberg et al. 1992; Lazo et al. 1998), tumoral lesions (Cohen et al. 1998) or a granulomatous encephalitis called `brain chagoma' (Di Lorenzo et al. 1996). Heart disease as the main clinical manifestation of T. cruzi reactivation with or without simultaneous CNS involvement has also been described (Sartori et al. 1998a, b).

Clinically, it may be difficult to differentiate Chagas' disease reactivation as a consequence of HIV infection and chronic chagasic disease. However, reactivation of Chagas' disease in HIV-positive individuals is always associated with high parasitaemia, while in chronic disease parasitaemia is very low and can only be detected by xenodiagnosis (Perez-Ramirez et al. 1999). Treatment of T. cruzi infection in HIV-positive individuals is recommended to be started early in the reactivation process when irreversible alterations have not yet occurred (Sartori et al. 1998b).

Schistosomiasis

Schistosomiasis, a systemic helminthic infection, is the second most prevalent tropical disease after malaria and affects approximately 200 million people in Africa, Asia, South America and in the Caribbean. Morbidity depends on the schistosome species involved, the intensity of infection, the topographic site affected by sequestered eggs and the immune responsiveness of the host.

So far, there are only indirect hints that schistosomiasis may have a negative impact on HIV infection. As shown in a study in HIV-positive individuals infected with Schistosoma mansoni in Kenya, egg excretion per worm pair was reduced with decreasing CD4 cell counts (Karanja et al. 1997). In fact, egg excretion rates were significantly correlated with CD4 levels.

It is assumed that the efficacy of treatment with praziquantel depends, among other factors, on a functioning immune system. This is mainly based on experimental studies which showed that immunodeficient mice are impervious to treatment with praziquantel (Sabah et al. 1985). Efficacy of treatment with praziquantel of S. mansoni infection in persons co-infected with HIV did not, however, differ from drug efficacy in HIV-negative persons (Karanja et al. 1998). Also, individuals with low percentages of CD4 counts did not differ with respect to treatment efficacy from those with normal counts. This can be explained by a sufficient degree of remaining T-cell function in immunosuppressed (HIV infection) as opposed to immune deprived (knockout mice) conditions. As the killing of adult worms knocked down by praziquantel is an antibody-mediated process (Brindley et al. 1989), the efficacy of treatment with praziquantel should not be affected if these antibodies were already present prior to the HIV infection.

There is recent pathophysiological, immunological and epidemiological evidence suggesting that genital schistosomiasis, a special form of S. haematobium infection, is a risk factor for the transmission of HIV and presumably also alters the natural history of HIV infection in a deleterious way (Feldmeier et al. 1994, 1995; Poggensee et al. 1999, 2000).

In women, genital schistosomiasis occurs in about 60% of individuals infected with S. haematobium. Genital manifestations involve the vulva, vagina and the cervix as well as upper genital organs and result in a pathology similar to that observed in some sexually transmitted infections (Attili et al. 1983). The cervix is the site predominantly affected, followed by the vagina and vulva (Poggensee et al. 2000). Thinning, erosion and ulceration of the epithelium is a typical clinical finding of genital schistosomiasis (Feldmeier et al. 1995). This is important, as breaks in the integrity of the mucosal barrier – e.g. due to either trauma or sexually transmitted genital ulcer disease – are associated with an increased risk of HIV transmission (Mabey 2000).

Schistosome ova evoke a complex cellular and humoral immune response. Granulomata form around ova with a radius 5–10 times larger than the size of the ovum. These granulomata are composed among others of activated lymphocytes, macrophages and epitheloid cells as well as some Lagerhans' giant cells, cell types known to express the CD4+ receptor. In the wall of the bladder a considerable proportion of these granulomata ulcerates into the lumen. Similarly, in genital schistosomiasis egg-induced ulcerative lesions could be the port of entry for HIV.

In the cervix, peri-oval granulomata are frequently formed near the basal layer of the epithelium. Within the egg granulomata, and also in adjacent areas, T cells, macrophages and Lagerhans cells abound (Helling-Giese et al. 1996). In mice, CD4 T cells amount to 8–10% of all granuloma cells (Weinstock et al. 1999). The abundance of CD4 receptor-bearing cells within the confines of the granulomata and in adjacent areas make a rapid binding of virus after penetration through the friable and eroded epithelium of the cervix very likely.

Correspondingly, genital schistosomiasis in males also points at an increased risk of HIV transmission. In male adolescents and adults in Madagascar S. haematobium caused inflammation of the prostate and of the seminal vesicles in most patients (Leutscher et al. 2000). As infection with S. haematobium in males induces a chronic inflammation in the pelvic genitals, it can be hypothesized that, in analogy to the situation in bacterial urethritis, there will be increased viral shedding in the semen in HIV co-infected individuals.

Onchocerciasis

About 18 million people are estimated to be infected with Onchocerca volvulus. The disease is limited to west Africa and a few small foci in Brazil and Venezuela. A case–control study including 1910 patients with onchocerciasis and 276 controls could not detect a significant association of onchocerciasis with HIV. Furthermore, no difference in the efficacy of treatment with ivermectin or in the occurrence of side-effects was found. Even HIV-infected patients with reduced T-cell ratios and reduced CD4 cell counts showed a typical reduction of microfilaria density after treatment (Fischer et al. 1995). This is plausible, as killing of microfilariae by the drug ivermectin is independent of immune effector mechanisms.

HIV-infected persons with onchocerciasis exhibit significantly reduced antibody responses to O. volvulus antigens and tend to lose their reactivity towards these antigens over time (Tawill et al. 1996). Also, the cellular immune response in co-infected persons is impaired (Sentongo et al. 1998). It is not known whether these findings have any bearing on re-infection rates in areas with ongoing transmission.

Lymphatic filariasis

Lymphatic filariasis is widely distributed in the tropics. It particularly occurs in areas with a high HIV prevalence such as sub-Saharan Africa and south-east Asia. There is no data available to show an impact of HIV infection on the prevalence or on the natural history of the disease. However, lymphatic filariasis is an example for a harmful interaction in the other direction. In a recent study Gopinath et al. (2000) observed that the replicative capacity of HIV is significantly enhanced in peripheral blood mononuclear cells from patients with untreated lymphatic filariasis. Consequently, untreated patients with lymphatic filariasis and co-infected with HIV could be at risk for rapid progression to AIDS once infected with HIV.

Intestinal helminths

Intestinal helminths are ubiquitous in low-income countries with prevalences of, for example 50–80% for ascariasis, trichuriasis and hookworm infections in many populations. Intestinal helminths induce immunological alterations that favour the progression from HIV seroconversion to AIDS. After HIV has spread to the systemic circulation its replication is limited by the fact that usually few activated lymphocytes and differentiated macrophages are present in the blood stream and that resting T cells and undifferentiated monocytes are not susceptible to HIV infection. However, in patients infected with intestinal helminths the number of activated T cells expressing human leucocyte antigen (HLA)-DR and HIV coreceptors is elevated (Kalinkovich et al. 1998). Secondly, HIV replicates preferentially in Th2 and Th0 type clones, and Th2 cells are usually abundant in individuals infected with helminths (Bentwich et al. 1998). Thirdly, peripheral blood mononuclear cells of patients with helminthic infection are significantly more susceptible to infection with HIV than those of uninfected controls (Shapira-Nahor et al. 1998). Finally, elevated IL-4 levels, characteristic of the Th2 type of immune response in helminthic infections, down-regulate Th1 differentiation and function (Bentwich et al. 1995).

The assumption that immune dysregulation associated with chronic helminthic infections alters the natural history of HIV infection in an unfavourable manner is sustained by results from a field study in Ethiopia: HIV viral load was significantly higher in individuals with various helminthic infections than in individuals without helminths and correlated positively with the parasite load. Furthermore, the viral load decreased after elimination of the worms by antiparasitic treatment (Bentwich et al. 2000).

Conclusions

So far, the global malaria situation has been considered to be largely unaffected by HIV. As HIV and malaria are the most frequent as well as the most important health problems in developing countries, even a small interaction would be of mundane consequence. This is highlighted by the recent finding that HIV-infected pregnant women are more susceptible to infection with P. falciparum (Steketee et al. 1996a, b; Pariseet al. 1998; Verhoeff et al. 1999, 2001) and that P. falciparum infections in HIV infected pregnant women are associated with increased plasma HIV RNA-1 levels (Hoffman et al. 1999; Taylor & Hoffman 2000). In non-pregnant adults, parasite densities are higher in HIV positives with clinical malaria and control over the density of blood-stage malaria infection seems to be diminished in HIV-1 infection. Actually, HIV-1 infection was associated with increased prevalence and intensity of P. falciparum in Ugandan adults with previously acquired anti-P. falciparum immunity (Taylor & Hoffman 2000). A substantial increase in parasitaemic episodes could contribute to malaria transmission on a population basis (Whitworth et al. 2000).

Leishmania/HIV co-infection has been denominated a disease entity since about a decade. The two infections reinforce each other and both have a deleterious impact on the immune system: whereas leishmanial infection accelerates HIV replication and the onset of full-blown AIDS, the immunosuppression caused by HIV promotes the onset of active leishmaniasis. Leishmania/HIV co-infection is therefore an example of a deadly gridlock with detrimental interactions in both directions.

While for Human African Trypanosomiasis, as well as for onchocerciasis and lymphatic filariasis there is little evidence for an interaction with HIV as yet, T. cruzi clearly behaves as an opportunistic agent in the presence of HIV infection.

In female genital schistosomiasis (FGS), the mucosal transmission of HIV is likely to be increased. The immunological characteristics of schistosomiasis in general and of genital schistosomiasis in particular seem to favour the initiation of the infection in the subepithelial tissue, to increase the initial viral burst, and to facilitate rapid local and systemic viral replication. Hence, a faster progression from asymptomatic HIV infection to full-blown AIDS becomes very likely in women with FGS.

Finally, chronic intestinal helminthiases also seem to play an important role in the progression of HIV disease.

The question arises whether there is a common immunopathogenetic basis for the detrimental interaction between HIV and pathogens biologically as different as for example, plasmodia and helminths. Chronic immune activation by parasitic infection could be one of the several causes of T-cell depletion in HIV infection and could considerably contribute to the progression of HIV disease (Bentwich et al. 1995, 1998). Infection with intestinal helminths, parasitic organisms living in a compartment aside of the systemic immune system, induces a status of chronic immune activation (Borkow et al. 2000, 2001). The prolonged and enhanced immune activation is even more prominent in systemic parasitic infections, whether protozoa or helminths. Even before HIV infection supervenes, chronic immune activation induced by parasites is associated with several of the immunological features of HIV disease. A decrease in the number of CD4+ and CD8+ T cells, an impaired natural killer activity, a marked increase in T-cell apoptosis as well as cellular anergy occur (Gastl et al. 1984; Feldmeier et al. 1985a, b; Zwingenberger et al. 1989a, b; Harmset al. 1991; Sher et al. 1992; Actor et al. 1993; Borkow et al. 2000).

The common immunological basis, supposedly, is the fact that parasites preferentially activate a Th2 type of help. Among other functions Th2 cells down-regulate the development of Th1 cells, inhibit macrophage activity and impair the cytotoxic T-lymphocyte response (Wolday et al. 1999; Bundy et al. 2000).

The early presence of IL-4 is the most potent stimulus for Th2 differentiation. The inducing effect of IL-4 dominates over other cytokines so that, if IL-4 levels reach a certain threshold, differentiation of the Th cell into the Th2 phenotype ensues (Romagnani 1997). The predominant activation of the Th2 type of help may have a harmful impact on HIV as an immune profile dominated by Th1 cytokines appears to be required for the containment of HIV infection (Fincham & Markus 1999; Wolday et al. 1999; Markus & Fincham 2000).

In contrast to HIV infection – which in the tropics is mainly a sexually transmitted disease – parasitic infections usually abound in childhood and/or adolescence. Moreover, in endemic areas sensitization towards the respective antigens already occurs in utero as mothers are likely to be infected with the parasites as well. However, exposure to antigens in utero results in generation of cytokine responses similar to those found in adults, and the ability of primed T cells to react accordingly can persist into childhood (Malhotra et al. 1999). Such a twist to an almost programmed Th2 type of help would be a considerable disadvantage when later in life the immune system of the affected individual encounters HIV. Consequently, not only is HIV infection acquired more easily but more rapid progression from asymptomatic HIV infection to full-blown AIDS may ensue.

Taken together it seems justified to conclude that people living in the tropics not only face a health threat in view of a still expanding HIV epidemic, they also have to fear that once infected with HIV this will alter the natural history of parasitic infections they are suffering from in an unfavourable way. On the other hand, the parasites they harbour impair the immune response towards HIV. This makes rapid progression from HIV infection to AIDS rather likely. As the great majority of parasitic diseases can be treated and – at least partly – also be prevented, we suggest that the control of parasitic infections should be included as a tool in the combat of HIV infection.