General aspects of visceral leishmaniasis with or without HIV co-infection in Northeast Brazil

Background The case of visceral leishmaniasis (VL) and human immunodeficiency virus (HIV) co-infection was reported in Spain first in 1985, and coincidence of these diseases has also been confirmed in more than 35 countries. Methodology We performed a comparative study in the state’s reference hospital for infectious/parasitic diseases, which treats adults with HIV/acquired immune deficiency syndrome (AIDS), between January 2007 and July 2017. The data obtained using this protocol were analyzed using SPSS. Principal findings In total, 163 patients were evaluated in this study, including 126 patients with coincident VL/HIV and 37 patients with VL alone. Both groups consisted primarily of male patients. The most commonly affected age group was 30–39 years (p < 0.001). Fever (p < 0.001) and hair loss (p = 0.007), which were more common in patients with VL alone, were more common. On hemogram, segmented neutrophils (p < 0.0001) were found to be more in the VL/HIV group than in the VL alone group. Additionally, AST and ALT levels differed between the groups (p < 0.001). On average, HIV was diagnosed 2.6 years before VL (p < 0.001). VL relapse was observed only in the co-infection group (36.5% of cases). Fever (β = +0.17; p = 0.032) in the first VL/HIV episode was identified as a risk factor for relapse (R2 = 0.18). The death rate of co-infected patients was 11.1%. Conclusion/Significance VL/HIV was prevalent among young adults, whereas the median patient age was higher in the VL group. The classic symptomatology of VL was more common in patients not co-infected with HIV, but attention is needed regarding the presence of fever in the first episode of VL as a risk factor for relapse in co-infected patients. No cases of VL relapse occurred in patients without HIV. AUTHOR SUMMARY Visceral Leishmaniasis (VL) and HIV have maintained increasing rates of populational occurrence in the Northeast region of Brazil, with advancing in rural areas and VL advancing in urban areas. This dynamic explains the increase of co-infection from 0.7% in 2001 to 8.5% in 2012. The state of Maranhão presents a large number of cases in the Northeast Brazil, a region with the majority of patients with this co-infection. The severity presented by these patients, in the observation of frequent relapses and high lethality, is presented in this study, which contributes to epidemiological, clinical, laboratorial, and evolutionary knowledge, trying to demonstrate these in the statistical data. Therefore, it should be pointed out that young adult patients with HIV/AIDS who present with any cytopenia, with or without hepatosplenomegaly and accompanied by fever or not, must be investigated for VL.

lethality, is presented in this study, which contributes to epidemiological, clinical, 57 laboratorial, and evolutionary knowledge, trying to demonstrate these in the statistical 58 data. Therefore, it should be pointed out that young adult patients with HIV/AIDS who 59 present with any cytopenia, with or without hepatosplenomegaly and accompanied by 60 fever or not, must be investigated for VL. requiring early diagnosis to avoid further complications [2]. In Brazil, 634,051 cases of 67 human immunodeficiency virus/acquired immune deficiency syndrome (HIV/AIDS) 68 had been registered through June 2016 [3]. HIV continues to spread across Brazil, and 69 its incidence in countryside locations has been increasing. This explains its association 70 with other endemic diseases, such as VL, which has also increased in frequency in 71 urban locations, thereby accelerating the clinical evolution of both HIV and VL due to 72 cumulative immunosuppression [4,5]. On the basis of altered epidemiological profiles 73 of both HIV and VL, the risk of exposure to the diseases has grown, with the risk of an 74 HIV-infected person developing VL surging by 100-2320-fold in endemic areas. 75 Moreover, co-infected patients also have reduced responses to therapy, thus resulting in 76 increased risks of recurrence and mortality [6]. The first case of VL/HIV co-infection 77 was reported in 1985, increasing attention in Mediterranean and northern European 78 countries regarding the possibility of this association and the gravity of both diseases 79 [7]. By 2007, 35 countries had already reported cases of VL/HIV co-infection [8]. In 80 Brazil, co-infection was first reported in 1990 in a patient from the Northeast region [9].

81
In 2015, 257 cases of co-infection VL/HIV were reported in the Western hemisphere, 82 corresponding to a rate of 7.4% of co-infection between all VL cases. Among these, 83 most cases (244) were recorded in Brazil [10]. In 2017, a descriptive and exploratory 84 study that analyzed data from all 21 Brazilian states that had reported cases of VL over 85 a 10-year period identified 1301 cases of VL/HIV co-infection. The researchers noticed 86 an increase in the number of VL/HIV cases over time [11]. Of these states, 4608 cases  which was named the investigation sheet of death from VL, was used [14]. The sheet 107 was used to record the following variables: sociodemographic (sex, age, occupation,    Giemsa) [15]. Serological: The indirect fluorescent antibody test has been widely used 134 for diagnosing VL, and it is currently provided by Brazil's Unified Health System. Its 135 sensitivity ranges from 82 to 95%, and its specificity ranges from 78 to 92% [16,17]. 136 Another method used enzyme-linked immunosorbent assay (ELISA), which has 137 sensitivity of 90-100% [18].     Table 3 shows the comparative analysis of the serum profiles between groups  The distribution of variables related to treatment is shown in Table 4. The most   in which no VL case had been previously reported [26]. By observing the 244 internalization of VL, it is possible to perceive the juxtaposition that occurs between 245 these VL infections found in people living with HIV an adequate environment to 246 develop [27].

247
In line with our findings, Cota et al. described fever and hepatosplenomegaly 248 as the most frequent complications in patients with VL alone. In contrast, Hurissa et al. 249 found typical VL complications in both studied groups [24,25]. In their study, fever was from many Brazilian studies, which reported the presence of splenomegaly followed by 255 fever, weight loss, and asthenia, in 81 patients with VL/HIV co-infection in Ceará [29] 256 and another study that identified weight loss, weakness, fever, and hepatosplenomegaly 257 as the most common physical changes in 65 patients with VL/HIV co-infection [13], the 258 most common symptoms in this study were skin pallor, hepatomegaly, and levels observed in some patients with VL/HIV co-infection [33,34,35] [25,41].

314
This study had several limitations. Because of the use of secondary data, 315 mainly in the retrospective study, incomplete data were obtained in some examinations.

316
Another limitation was difficulty in analyzing some laboratory data, which precluded 317 comparative analyses between the two groups, although we observed that the co-  In summary, our study found that VL is a public health issue in northeastern 328 Brazil, especially in Maranhão, where its incidence is increasing. The severity of the 329 disease is greater due to co-infection with HIV, which modifies the epidemiology and 330 clinical presentation of VL. Thus, the classical triad of fever, paleness, and 331 hepatosplenomegaly was not expected. We also concluded that treatment remains 332 challenging, despite advances in antiretroviral treatment, so a definitive response to 333 treatment for VL cannot be predicted when the disease is associated with HIV.