Should we care about Plasmodium vivax and HIV co-infection? A systematic review and a cases series from the Brazilian Amazon

Malaria and human immunodeficiency virus (HIV) infections are two major public health concerns [1]. Given the considerable epidemiological overlap between malaria and HIV, a substantial number of co-infections may occur [1]. In 2019, malaria accounted for 228 million cases and resulted in 405,000 deaths, globally [2]. Of the 5 Plasmodium species that can infect humans, Plasmodium vivax is the most geographically widespread, and is responsible for 75% of malaria-related cases in the Latin American region [2‐4]. Currently, 37.9 million people are living with HIV worldwide, with 1.7 million newly diagnosed in 2019. An estimated 690,000 deaths were due to AIDS-related diseases in 2019 [5]. Globally, these infections jointly claim the lives of about 2 million people each year [6].

Malaria and HIV infections can interact in a bidirectional and synergistic manner, which may lead to an exponential rise in their deleterious effects [7]. HIV can impair immune responses to malaria parasites, and lead to an inability to control parasite clearance, thus resulting in high parasitic loads, which in turn, can increase malaria transmission rates [1, 8, 9]. Clinically, HIV has been shown to contribute to a higher incidence of falciparum malaria [10], including its severe form, which is characterized by anaemia, cerebral malaria and increased risk of congenital infections [11‐13]. The impact of HIV on the severity of malaria appears to be restricted to patients with CD4 + T cell counts < 350 cells/μL [10].

Meanwhile, malaria infection is associated with strong CD4 + T cell activation and increased levels of pro-inflammatory cytokines, which provides an ideal micro-environment for HIV viral replication. This potentially worsens the clinical picture, increasing HIV progression to AIDS and maintaining the viral cycle [1, 14‐18]. The immunosuppression caused by HIV infection can reduce the control of the Plasmodium infection. Moreover, HIV therapy can impair malaria treatment, with a significant increase in adverse events, as well as potential selection of treatment-resistant parasites [19‐23]. Plasmodium co-infection has also been shown to increase HIV viral load and transiently decrease CD4 + T cell count [24‐27]. However, these interactions are mostly described for Plasmodium falciparum.

Both malaria and HIV are highly prevalent in tropical and sub-tropical regions of the world, which may result in an increased prevalence of such co-infection [40]. Several studies have reported on HIV and P. falciparum co-infection, mainly in Africa [40‐42], but only a few studies have described cases of HIV/PvCo patients. Indeed, from the initial search, only 2% of studies dealt with HIV/PvCo. This could be due to several factors, such as low co-infection rates, low prevalence of severe cases and therefore lack of reporting, and most importantly, lack of systematic HIV screening in vivax malaria-positive patients.

HIV and malaria affect millions of people in overlapping geographical areas. Traditional risk factors for HIV and vivax malaria may apparently be dissociated, which could possibly explain the HIV/PvCo low prevalence reported here compared to similar studies from other vivax endemic regions [43‐53]. Local characteristics may account for the predisposition of HIV/PvCo. About 90% of all malaria episodes in Brazil is caused by P. vivax and these are concentrated in the Amazon region [54]. The clinical spectrum varies from asymptomatic cases to mild clinical symptomatology, and complications may occur. Male subjects present a higher incidence of malaria with younger individuals predisposed to a higher risk of clinical complications [55]. Malaria transmission in the Amazon region occurs mainly in peri-urban and rural areas, with an increase in the number of cases occurring in the dry season and after public holidays and regional festivities [56]. On the other hand, HIV infection, which is prevalent in 0.4% of the general population in Brazil, occurs mainly in key groups, such as in men who have sex with men, sex workers and transgender individuals [57]. For instance, in 2019, the AIDS detection rate was 29.1 cases/100,000 habitants in the state of Amazonas, while in the capital, Manaus, 46.9 cases/100,000 habitants, which is significantly higher when compared to the rest of the state, and in comparison with the Brazilian national rate (17.8 cases/100,000 habitants) [57]. HIV transmission occurs mainly in urban environments [58]. In the state of Amazonas, Manaus, Parintins and Tabatinga cities are some of the major HIV hotspots [59]. In most municipalities in the interior of the state of Amazonas, there is not a clear delimitation between urban, peri-urban and rural regions. Constant population displacement between such regions and between municipalities is very common. This may partially explain the exposure of people living with HIV to malaria. In addition, it is uncertain to what extent cultural and social characteristics significantly overlap between both diseases to produce a low prevalence as seen in this study. However, since the systematic screening of HIV infection is not done in acute malaria cases in Brazil, the real HIV/PvCo burden is unknown. Furthermore, it is important to mention that HIV-positive cases have been increasing in recent years in the northern region of Brazil [57].

Separately, HIV, malaria and TB are considered to be the most common and severe infectious diseases in the world [60]. Interestingly, a triple infection with these three diseases was more prevalent in this study when compared with other studies from Africa, although this may be attributed to screening, as previously mentioned [42, 61].

Two patients presented with G6PDd. The lower the G6PD activity, the lower the individual’s ability to tolerate oxidative stress and, when faced with oxidative stressors, such as certain foods, e.g., fava beans and drugs, such as primaquine or sulfonamides, G6PDd individuals may develop acute life-threatening haemolysis [62]. The prevalence of G6PDd applies to both HIV and malaria. HIV infection and antiretroviral therapy (ART) are both associated, separately and together, with increased oxidative stress. The impact of G6PDd on the oxidative stress of people living with HIV (PLHIV) on ART is controversial [63‐65]. PLHIV have significantly lower levels of antioxidants, haematological parameters and CD4 + T cells compared to healthy subjects. Nonetheless, PLHIV on ART have presented a higher level of antioxidants compared to ART naïve subjects [67]. Also, antioxidant status has been shown to be significantly higher in those with CD4 + T cells ≥ 200/cu mm [66].

The prevalence of G6PDd varies across Latin American and Caribbean countries, with the African variant present in a wide range in this region [67]. Primaquine is a strong oxidative drug and may cause severe acute haemolysis in G6PDd individuals who take primaquine for malaria treatment [68]. G6PDd testing is currently recommended by WHO prior to starting primaquine in the radical cure of P. vivax and Plasmodium ovale malaria [30, 69]. Systematic screening for G6PDd is however not a requirement when starting HIV treatment [29]. Therefore, it was not possible to determine whether G6PD deficiency in HIV/PvCo influenced clinical outcomes.

Separately, malaria and HIV cause significant laboratory abnormalities; a co-infection scenario may intensify such alterations [27, 40]. Anaemia, thrombocytopenia, and leukopaenia, for both malaria and HIV, and in malaria/HIV co-infected patients, have been reported to be strong and independent predictors of morbidity and mortality [52]. The prevalence of anaemia in this case series was high, with most patients presenting mild to moderate anaemia, similar to other studies of HIV/PfCo [26, 70, 71]. Nonetheless, this is higher when compared to P. vivax mono-infected adults from the Amazonas [72].

A high prevalence of thrombocytopenia (85.7%) was also observed in this study. Two studies conducted in patients with P. vivax showed a similar prevalence (62.9 and 72%, respectively) [73]. In a systematic review [74], severe and fatal thrombocytopenia was observed in 10.1% of patients with vivax mono-infection malaria, while severe thrombocytopenia was more prevalent in this study. Anaemia and thrombocytopenia in HIV-malaria co-infections have a multifactorial origin and are a frequent complication that may become clinically important in HIV infection [40, 52, 75].

The impact of HIV on the clinical severity of falciparum malaria seems to be primarily motivated by the inability of the immune system to control the parasitic burden [41, 76, 77]. Severe malaria was observed in approximately 30% of adults with falciparum malaria and HIV in the urban area of Burkina Faso [78]. Studies in areas of low malaria transmission in South Africa and India showed an association between severe malaria and HIV [76, 79‐81]. For severe vivax malaria, the current study showed a higher prevalence compared to another study with severe vivax malaria in children and adult patients (23.8 vs 12.6%) [82]. Despite the low number of cases, HIV co-infection seems to exacerbate clinical worsening of vivax malaria, as it has a higher prevalence than that found in vivax malaria mono-infection patients treated at FMT-HVD (~ 14%) [55]. Some studies have shown that the risk of malaria severity increases in HIV patients with a CD4 + T cell count < 200 × 10⁶ cells/L or < 350 cells/μL [14, 80]. In this study, 42.1% of patients with malaria infection had CD4 + T cell counts of less than 200 cells/μL, and one of them had severe malaria.

This study had several limitations. Regarding the systematic review, prevalence studies and those exploring severe clinical outcomes may underestimate the HIV/malaria co-infection, since it is rarely screened in vivax malaria-endemic regions. The comparison of clinical disease dynamics and important outcomes was not possible due to the absence of control groups, e.g., P. vivax mono-infection and HIV mono-infection to address associations of laboratory and clinical outcomes with HIV/PvCo, which was mainly due to the lack of systematic screening. Moreover, there is a low prevalence of severe cases of P. vivax, especially when opportunely diagnosed and treated, or in the absence of co-morbidities. Despite the analysis of the present results, it is not possible to assume that malaria increases anaemia and thrombocytopenia in PLHIV, or vice-versa. Finally, an accurate prevalence of HIV/PvCo, and roughly all other co-infections, is significantly hampered by the absence of a systematic screening, which in low- and middle-income countries is performed at the discretion of a clinician upon clinical suspicion.

Malaria from P. vivax infection appears to have a low prevalence in HIV-infected individuals, with only a few studies describing clinical and laboratory outcomes. This study showed a low prevalence of HIV/PvCo, despite the important local prevalence of vivax malaria and HIV diseases separately. Even though relatively small, thus far this is the largest case series to describe HIV/PvCo. The findings described here shows the need for further research on the interaction between HIV and vivax malaria diseases, due to the potential worsening of both disease conditions in a co-infection scenario, the impact of G6PDd, and the possible epidemiological effects. At present, with a potential increase of severe cases due to the increase of vivax malaria cases and new HIV diagnoses, prospective studies are needed to elucidate aspects related to the pathogenesis of this co-infection, concomitant treatment and drug interactions, severity outcomes, and the possible increased frequency of P. vivax relapses secondary to HIV co-infection. In addition, future studies should address how vivax malaria chronically influences CD4 cells and how this is associated with HIV viral load dynamics.