Visceral Leishmaniasis in Human Immunodeficiency Virus (HIV)-Infected and Non-HIV-Infected Patients
2001; Wolters Kluwer; Volume: 80; Issue: 1 Linguagem: Inglês
10.1097/00005792-200101000-00006
ISSN1536-5964
AutoresVicente Pintado, Pablo Martı́n-Rabadán, María Luisa Fernández Rivera, Santiago Moreno, Emilio Bouza,
Tópico(s)Eosinophilic Disorders and Syndromes
ResumoIntroduction Visceral leishmaniasis (VL) is a severe disease associated with considerable morbidity, characterized by fever, hepatosplenomegaly, pancytopenia, and hypergammaglobulinemia, caused by intracellular protozoa of the genus Leishmania(27, 32, 122). VL is usually observed in immunocompetent hosts in endemic areas, but it occasionally occurs as a rare complication of immunodeficiency states and autoimmune diseases (4, 16, 22, 31, 60, 74, 96). Since the mid-1980s, VL has been recognized as a common complication of patients infected with the human immunodeficiency virus (HIV) in southern Europe (23, 35, 85, 93, 105, 113, 114, 138, 141, 144), where between 1.5% and 9% of patients with acquired immunodeficiency syndrome (AIDS) suffer from newly acquired or reactivated VL (70, 163). All reports regarding VL in HIV-infected patients concur in pointing out that this infection occurs in profoundly immunodepressed patients and may pre-sent with typical and atypical features (8, 11, 23, 93, 124, 128, 129, 163). Until recently, no comparative studies about the clinical presentation and outcome of VL in HIV-infected and immunocompetent patients have been reported (137). Regarding diagnostic procedures, a constant finding in HIV-infected individuals with VL has been the low sensitivity of the serologic studies; more than 50% of these patients show no detectable levels of antileishmanial antibodies (11, 64, 104, 164). However, there is still debate regarding the different yield of visualization techniques, which has led some authors to recommend the routine performance of bone marrow culture for Leishmania in this population (23, 93, 163). The best drug, the optimal doses, and the duration of treatment of AIDS-associated VL are still unknown, and much controversy persists regarding the selection of antimonial agents or amphotericin B. A clinical response is achieved in 38%–81% of HIV-infected patients (87, 93, 113, 141), but relapses typically occur in 25%–60% of cases (11, 124). Different maintenance regimens have been used to prevent relapses, with variable results (125, 139). In this paper we report the results of a comparative study on the epidemiology, clinical presentation, diagnostic procedures, response to treatment, relapses, and outcome of VL in a series of 80 HIV-infected and 40 non-HIV-infected patients. We describe the largest series of patients with HIV-Leishmania coinfection studied at a single institution, with a historic perspective of the major changes of the epidemiology and outcome of the disease from the beginning of the epidemic until the recent introduction of the new antiretroviral therapies. Methods Hospital setting and population of study This retrospective study was performed from January 1974 to December 1997 in a 1,750-bed general, teaching and referral hospital serving a population of 650,000 inhabitants in Madrid, Spain. The hospital has medical, surgical, pediatric, and obstetric units and, since 1991, the hospital has had a very active HIV program. Diagnostic procedures and microbiologic techniques Antileishmanial antibodies were assayed by indirect immunofluorescence (IIF) (Leishmania Indirect Immunofluorescence Antibody Test, Mardx Diagnostics, Scotch Plains, NJ). Titers ≥1:80 were considered significant. In a few patients, antibodies were measured by hemagglutination (Boehringerwerke AG, Marburg, Germany) and titers ≥1:8 were considered significant. Testing for HIV antibodies was performed using an enzyme-linked immunosorbent assay (ELISA) and was confirmed by Western blot. Leishmania cultures were made by inoculating the tissue specimens or the buffy coat layer of peripheral blood onto Schneider Drosophila Medium (Gibco, Paisley, Scotland, UK) supplemented with 30% inactivated fetal bovine serum (Flow Laboratories, Irvine, Scotland, UK), penicillin (16.25 μg/mL), and streptomycin (0.026 μg/mL) or the Novy-McNeal-Nicolle Medium. All cultures were incubated at room temperature (22 °C to 25 °C) and examined every other week for 4 weeks. Tissue samples were stained with May-Grünwald-Giemsa or Periodic Acid-Schiff staining. Blood cell counts and biochemical studies were performed by conventional automated techniques. CD4+ and CD8+ lymphocyte subpopulations were measured by flow cytometry. Case definition and diagnosis of visceral leishmaniasis Diagnosis of VL was confirmed by demonstration of Leishmania amastigotes in tissue samples or by isolation of promastigotes in culture. In a few patients with no positive cultures, the diagnosis was made in the presence of suggestive clinical features (fever, visceromegaly, and hematologic cytopenia), significant serologic titers, and response to specific treatment, in spite of a negative bone marrow aspirate and after exclusion of other opportunistic infections. Relapse was defined as a new episode of VL, diagnosed by means of the criteria mentioned, occurring after a successful response to the treatment of the previous episode. VL was considered HIV-associated if the patient was seropositive for HIV at the time of the first episode of leishmaniasis. The stages of HIV infection and the diagnosis of AIDS were defined in accordance with the 1993 criteria of the Centers for Disease Control and Prevention (Atlanta, GA) (42). Pulmonary tuberculosis, repeated pneumonia, and invasive carcinoma of the uterine cervix have been included officially as confirmatory of AIDS in Spain since January 1994. Stages A3 and B3 are not classified as confirmatory of AIDS in Europe (13), and these European classification criteria were used in the study. Clinical study An active search for Leishmania was conducted in all patients admitted to the hospital with prolonged fever and/or hepato-splenomegaly and/or hematologic cytopenias. Records of patients with VL diagnosed between 1974 and 1997 were retrospectively reviewed. The following data were obtained routinely by reviewing the clinical charts of the patients: demographic information (age, gender, underlying diseases), stage of and risk factor for HIV infection, other previous or concomitant conditions, clinical presentation, laboratory test results, methods of diagnosis of VL (parasitologic and serologic data), antileishmanial treatment, adverse events, response to treatment, antiretroviral therapy, antileishmanial prophylaxis, and outcome. Anemia was defined as a hemoglobin level lower than 12 g/dL; leukopenia, lymphocytopenia, and thrombocytopenia were defined as a leukocyte, lymphocyte, and platelet count lower than 4,500, 1,000, and 150,000/mm3, respectively. The final outcome and the survival time were evaluated at the end of the follow-up period (December 1998) for all cases included in the study between 1974 and 1997. The register of the AIDS cases and the general population served by our hospital was used as the basis for the calculation of incidence rates. Treatment, response criteria, and adverse events The following 3 treatment regimens were employed. The first was meglumine antimoniate (MA) that was combined in some cases with oral allopurinol (300–1,200 mg/day orally), or interferon-γ (100 μg/m2 subcutaneously). The second was amphotericin B (AmB) deoxycholate or liposomal amphotericin B (AmBisome; Nexstar, San Diego, CA). The third was pentamidine isethionate (4 mg/kg per day intravenously for 4 weeks). Some patients were treated with a combination of allopurinol (600–1,200 mg/day orally) with either ketoconazole (400 mg/day orally) or itraconazole (200–800 mg/day orally) for 4–6 weeks. The clinical response was defined as clinical cure if after a complete course of treatment the patient was afebrile with improvement of the pancytopenia and hepatosplenomegaly, and was asymptomatic 1 month after the end of treatment. Clinical failure was defined as the persistence of clinical symptoms 1 month after completion of therapy, and the response was considered not assessable if the patient did not complete treatment, died, or was lost for follow-up in the first month after treatment. A parasitologic control study (staining and/or parasitic culture of bone marrow) was performed on some patients; this study was made in the first month after the completion of treatment in most cases. Parasitologic cure was defined as the negative result of smears and/or culture of bone marrow aspirate. Parasitologic failure was defined by the presence of parasites on staining and/or culture of bone marrow aspirate, blood, or other specimens. The final response was evaluated through the clinical and parasitologic responses. Final response was considered successful if there was a clinical and parasitologic cure; unsuccessful if there was a clinical and/or parasitologic failure; and not assessable if the clinical response was not assessable. In those patients in whom the parasitologic response was not evaluated, the final response was considered successful if there was clinical cure. Evaluation of adverse events was carried out by reviewing the clinical charts of the patients. The following were considered serious adverse events: anemia (defined as a reduction in hematocrit that required blood transfusion), renal toxicity (a twofold increase in the normal level of serum creatinine), and hyperamylasemia (a twofold increase in normal serum amylase values). Secondary prophylaxis was offered to some HIV(+) patients who had completed a full course of therapy if their VL-related symptoms had resolved. Secondary prophylaxis with liposomal AmB (200–350 mg monthly), MA (850–1,275 mg monthly), pentamidine (300 mg monthly), and allopurinol (300–900 mg daily), was given on an outpatient basis. Statistical analysis Quantitative variable analysis was performed with the Student-t test if such variables presented a normal distribution; otherwise the Mann-Whitney test was applied. The study of the qualitative variables was performed with the chi-square test with the Yates correction or the Fischer exact test (2-tailed) when necessary. Survival curves were estimated using the Kaplan-Meier method. Differences between survival curves were evaluated using the log-rank test. To identify independently significant factors that could influence survival or relapses, a Cox stepwise proportional hazard model was applied. All p values were 2-sided and values of 0.05 or less were considered statistically significant. The SAS 6.11 software package was used for the database and all statistical procedures. Results Epidemiologic and demographic data During the 24-year study period (1974–1997), 133 newly diagnosed cases of VL were observed in the population served by our institution. During the first 12 years (1974–1985), 17 cases were detected and only 1 case, observed in 1985, was associated with HIV infection. By contrast, during the second half of this period (1986–1997), we recorded 116 new cases of VL, 81 of which (69.8%) were diagnosed in patients with HIV infection. The distribution of cases by year of diagnosis and patient's HIV status is shown in Figure 1.FIG. 1: Distribution of cases of visceral leishmaniasis in HIV-infected and non-HIV-infected individuals by years (1974–1997).We noted an increase in the incidence of VL in our setting, from 0.22 cases per 100,000 inhabitants/year in the first half of the study period (1974–1985) to 1.55 cases per 100,000 inhabitants/year in second half. As the incidence of VL in HIV(–) patients has remained fairly constant during the study period, HIV-associated cases may explain the rising incidence of VL observed in the last period. In the second half of the study period (1986–1997), 7,438 individuals with HIV infection (3,510 of whom had been diagnosed with AIDS) were recorded at our institution. These data indicate that 2% of the HIVinfected patients had suffered from VL. In the last period, for each new case of VL diagnosed in an HIV(–) patient, approximately 2 new cases were diagnosed in HIV(+) patients. In our area, the incidence of VL among AIDS patients is 418-fold higher than among HIV(–) individuals. Adequate clinical information was available for 120 of the 133 patients with VL. Underlying diseases associated with VL were HIV infection in 80 patients (66.6%), other immunodeficiency states in 2 (1 case of renal transplant and 1 case of lymphoma) and other diseases in 4 (liver cirrhosis, diabetes mellitus, chronic pulmonary disease, and systemic vasculitis, 1 case each). Three HIV(+) patients had had an episode of leishmaniasis (cutaneous in 2 and visceral in 1) before the current diagnosis of VL. Five representative case reports with different clinical aspects of HIV-associated VL are presented in the Appendix. Most patients (75%) were male and all were Caucasian. The male-to-female ratio was higher among HIV(+) patients (80%) than in HIV(–) patients (65%), but this difference was not statistically significant. Mean age was significantly higher in the HIV(+) than in the HIV(–) patients (33.2 ± SD of 8.2 yr versus 23.2 ± 26.9 yr; p = 0.002). Age distribution of VL cases is shown in the Figure 2.FIG. 2: Distribution of 120 cases of visceral leishmaniasis in HIV-infected and non-HIV-infected individuals by decades of life.Table 1 summarizes the mean demographic, clinical and immunologic data of HIV(+) patients. Risk factors for HIV infection included intravenous drug abuse in 63, homosexual contact in 6, heterosexual contact in 6, and other risk factors in 5. The percentage of intravenous drug abuse in patients with VL was similar to that of all notified AIDS cases at our institution (78.7% versus 77.3%; p = NS).TABLE 1: Demographic features in 80 patients with visceral leishmaniasis and HIV infectionThe European clinical criteria for AIDS were fulfilled before or at the time of diagnosis of VL in 43 (53.7%) of the HIV(+) patients. AIDS-defining diagnoses in these patients were as follows: pulmonary or extrapulmonary tuberculosis (21 patients), esophageal candidiasis (14), Pneumocystis carinii pneumonia (9), cytomegalovirus retinitis (4), toxoplasmic encephalitis (4), cryptosporidiosis (4), cryptococcosis (3), disseminated Mycobacterium avium infection (3), HIV-related encephalopathy (3), recurrent pneumonia (3), chronic herpes simplex infection (2), Kaposi sarcoma (1), and cervical carcinoma (1). In addition, the CD4+ lymphocyte count was < 200/mm3 in 87.1% (61/70) of the HIV(+) patients. Clinical data Complete clinical information about the first episode of VL was available for 112 patients (Table 2). Most individuals had the typical clinical features of VL: fever (91.9%), splenomegaly (86.6%), and hepatomegaly (79.4%) were the most common. VL presented as fever of unknown origin in 44.6% of cases. The clinical presentation of VL was similar in HIV(+) and HIV(–) individuals. However, HIV(+) individuals had a statistically significant lower incidence of splen-omegaly than HIV(–) patients (80.8% versus 97.4%; p = 0.02). Leishmanial involvement of the skin was seen in 6 (8.2%) HIV-infected patients, but in none of the HIV(–) patients (p = NS).TABLE 2: Clinical features of the first episode of visceral leishmaniasis in 112 patientsAnalytical data Relevant analytical data are summarized in Table 3. Pancytopenia was observed in 72.6% and 46.1% of the HIV(+) and HIV(–) patients, respectively. Leukopenia, lymphocytopenia, and thrombocytopenia were statistically more frequent (p < 0.05) in HIV(+) than in HIV(–) individuals. In addition, the absolute values of leukocytes, lymphocytes, and platelets were significantly lower (p < 0.05) in HIV(+) than in HIV(–) patients (data not shown). The frequency of hypergammaglobulinemia (gammaglobulin level ≥2.5 g/dL) was similar in both groups. Among HIV(+) patients, the CD4+ lymphocyte count was <200/mm3 in 87.1% (61/70) and <50/mm3 in 41.4% (29/70) of cases; the mean CD4+ count was 90 cells/mm3 (range, 3–470 cells/mm3) and the mean CD8+ count was 480 cells/mm3 (range, 50–3,300 cells/mm3). A decreased CD4+ to CD8+ ratio (<1) was observed in 94.9% of cases.TABLE 3: Analytical data of the first episode of visceral leishmaniasis in 112 patientsDiagnostic procedures Leishmaniasis was parasitologically confirmed in 89.2% (100/112) of cases. The diagnosis was made by direct examination of bone marrow in 83 cases (32 also had a positive bone marrow culture), by bone marrow culture alone in 15 cases, by observation of spleen aspirate in 1 case, and by blood culture in 1 case. Table 4 shows the yield of the different diagnostic methods. Bone marrow study was the procedure most often used, and was performed on all patients. Blood culture of the buffy coat was carried out in 22 patients.TABLE 4: Diagnostic methods of Leishmania infection in 112 patients *The most sensitive diagnostic method was the direct examination of the bone marrow aspirate (positive in 71.4% of cases), followed by bone marrow culture (positive in 63.5% of cases) and serology (positive in 59.8%). The sensitivity of serologic studies was significantly lower in HIV(+) than in HIV(–) patients (50% versus 80%; p < 0.001). There was no significant difference in the sensitivities of the other diagnostic tests between both groups. The diagnostic yield of bone marrow observation, including aspirate and biopsy, was 74.1% (83/112). Among the 47 cases with positive bone marrow culture, the direct observation (aspirate and/or biopsy) had already been positive in 32. It is noteworthy that bone marrow culture was the only positive test in the remaining 15 cases, accounting for 15.3% of the diagnoses made by bone marrow study. The diagnostic yield of bone marrow study, including direct staining and culture, was 87.5%. Treatment and response Treatment was given to 110 patients; 2 other patients refused to be treated. The therapeutic regimens used and the response to therapy are listed in Table 5. Most patients were treated with MA or AmB, given to 88 (78.6%) and 17 (15.2%) patients, respectively. Antimonials were given at dose of 20 mg of pentavalent antimony (SbV)/kg/day (with a maximum daily dose of 850 mg in adults) for 3–4 weeks in most patients. Allopurinol was also given in 4 cases and interferon-γ in another 4. AmB was employed at a dose of 0.5–1 mg/day for 3–4 weeks in most patients; 8 patients received liposomal AmB (see Table 5).TABLE 5: Response to treatment of the first episode of visceral leishmaniasis in 112 patientsSerious adverse events occurred in 33 (30%) of the 110 therapeutic courses. This complication was observed in 27 (30.6%) of the 88 patients receiving MA and in 7 (41.1%) of the 17 patients receiving AmB. The incidence of serious reactions was as follows. In those receiving MA, hyperamylasemia was observed in 15.9% (14/88) and anemia was seen in 14.7% (13/88). Anemia (29.4%) and renal failure (17.6%) were the most frequent events observed in patients treated with AmB. Adverse events secondary to MA were more frequent in HIV(+) than in HIV(–) individuals; MA was stopped due to an adverse event in 18.5% and 2.5% of the HIV(+) and HIV(–) individuals, respectively (p = 0.02). Among HIV(+) individuals, the frequency of adverse events requiring discontinuation of therapy was slightly higher with MA than with AmB (18.5% versus 7.1%; p = NS). Clinical cure was achieved in 67.8% of patients. HIV(+) individuals had a lower rate of response than HIV(–) patients (56.2% versus 89.7%; p = 0.004). HIV(+) patients have a clinical failure more frequently (30.1%), as is shown in Table 5. A parasitologic control study was carried out in 34 (31%) of the 110 treated patients; this study was performed during the first month after the end of therapy in 73.5% of cases. A parasitologic cure was achieved in 19 (63.3%) of the 30 HIV(+) patients and in the 4 HIV(–) patients (100%) in whom the study was performed. This difference was not statistically significant. Overall, the final response was considered successful in 67% (75/112) of the patients. HIV(+) patients had a lower rate of successful response than HIV(–) individuals (54.8% versus 89.7%; p = 0.001). In 20 (17.8%) patients it was impossible to evaluate the final response because of premature death, incomplete treatment, or loss for follow-up. No difference was observed in the successful response rate of HIV(+) patients receiving AmB (61.1%) or MA (58%). After the failure of the initial therapy, 15 HIVinfected individuals were retreated with additional courses of MA, AmB, or other drugs. After the second and third courses of therapy, a successful response was achieved in 10 and 3 patients, respectively, and a successful response was finally achieved in 53 (72.6%) of the 73 HIV(+) individuals. Antiretroviral therapy was given to 60 HIV(+) patients; this treatment was started after the first episode of VL in 41 cases. Monotherapy or double therapy with reverse transcriptase inhibitors was used in 35 and 4 cases, respectively. Patients studied in the last 3 years had the possibility of receiving highly active antiretroviral therapy (HAART) with triple drug regimens including protease inhibitors. HAART was given to 21 patients for a mean period of 15.9 months (range, 6–26 mo). Data about antiretroviral therapy were not available for 6 patients. Relapses After successful treatment of the initial episode, VL relapse was observed in 40 individuals. The relapse rate was significantly higher in HIV(+) than in HIV(–) patients (46.2% versus 7.5%; p < 0.001). The relapse rate among HIV(+) patients who successfully responded to therapy and who were followed for at least 3 months was 56.6% (30/53). Among HIV(+) individuals, 24 had 1 relapse, 11 had 2 relapses, 1 had 3, and 1 had 4. The mean period before occurrence of the first relapse was 13.8 months (range, 3–44 mo) and from the first to the second relapse was 11.5 months (range, 4–27 mo). Only 3 HIV(–) individuals had a relapse, 3–9 months after the initial episode. A comparison of the clinical features, diagnostic methods, and response to therapy of the first episode and the first relapse of HIV-related VL is shown in Table 6. Clinical presentation was similar to that of the first episode, but the mean duration of symptoms was shorter in relapses (p < 0.001). A lower proportion of relapses presented as fever of unknown origin (11.4%) when comparing with the first episode (47.9%) (p < 0.001). No significant differences were observed in the analytical data or in the yield of the different diagnostic procedures. AmB and combinations of allopurinol and azole compounds were given more frequently in the relapses, but the response rate was comparable to that of the first episode.TABLE 6: Clinical features, diagnostic procedures, and response to treatment of the first episode (73 cases) and the first relapse (35 cases) of visceral leishmaniasis in HIV-infected patientsTwenty patients received secondary prophylaxis for a mean period of 10.9 months (range, 2–20 mo). Liposomal AmB was used in 9 cases, antimonials in 7, allopurinol in 3, and pentamidine in 1. Thirty-three patients with successful response to treatment and with at least 3 months of follow-up did not receive prophylaxis. During the follow-up period, 7 (35%) patients of the group receiving prophylaxis and 22 (66.6%) patients of the group without prophylaxis experienced VL relapse. Multivariate analysis showed that the only variable that had a protective effect on VL relapse was the use of secondary prophylaxis with either liposomal AmB or MA (odds ratio = 0.19, 95% confidence intervals [CI] = 0.05–0.72, p = 0.01). Neither the clinical stage of HIV infection at diagnosis, the blood cell and CD4+ lymphocyte counts, the presence of antileishmanial antibodies, nor the antiretroviral therapy were related to recurrence. Among the 21 patients receiving HAART, there were 6 relapses (28.5%) and 2 new cases of VL after the introduction of this treatment. Outcome and prognostic factors In December 1998, after at least 1 year of follow-up completed by all the patients, 38 were alive, 46 were dead, and 36 were lost for follow-up (Table 7). Most HIV(–) individuals (50%) were lost for follow-up after achieving a successful response. In contrast, a high proportion of HIV(+) patients died (53.7%) during the follow-up period, as a consequence of AIDS progression. There was a significant difference in the mortality rate between HIV(+) and HIV(–) individuals (53.7% versus 7.5%; p < 0.001). Death occurred during the first episode of VL in 11 (13.7%) of the HIV(+) patients and during the treatment of a relapse in another 11 patients. Although active VL could have contributed to death in these 22 patients, leishmaniasis was the main cause of death in only 2 of them. Mortality was due to other AIDS-related illnesses or clinical conditions in most cases. More than three-quarters of HIV(+) individuals had other severe infections that were the main cause of death. At the end of the study, 62 (77.5%) patients met clinical criteria for AIDS. The mean survival time was shorter in HIV(+) than in HIV(–) patients (25 versus >160 mo; p < 0.001), as illustrated in Figure 3, which shows the survival curves of both groups.TABLE 7: Outcome of visceral leishmaniasis in 120 patientsFIG. 3: Long-term survival of 80 HIV-infected and 40 non-HIV-infected patients with visceral leishmaniasis.Multivariate analysis showed that the only variables that had a significant effect on the survival of HIV(+) patients were a low CD4+ lymphocyte count (risk ratio = 0.989, 95% CI = 0.98–0.99, p < 0.001), and the use of secondary prophylaxis for VL with either MA or AmB (risk ratio = 4.8, 95% CI = 1.1–20.1, p = 0.03). The median survival, estimated by the Kaplan-Meier method, was significantly shorter in patients with CD4+ lymphocyte count < 50/mm3 than in patients with > 50/mm3 (14 versus 36 mo; p < 0.001), as illustrated in Figure 4. The median survival in patients without secondary prophylaxis for VL was slightly shorter than in patients receiving prophylaxis (20 versus 25 mo; p < 0.02, Figure 5). Antiretroviral therapy also had a significant impact on survival. The median survival was significantly shorter in patients receiving monotherapy, double therapy, or no treatment than in patients receiving HAART (16 versus > 66 mo; p < 0.001, Figure 6). Because there were no deaths in the group of patients receiving HAART, this variable was not included in the final model of multivariate analysis. Several other factors, including hematologic parameters, total lymphocyte count, previous AIDS diagnosis, or VL relapse, did not influence the final outcome.FIG. 4: Long-term survival of 70 HIV-infected patients with visceral leishmaniasis with a CD4+ lymphocyte count higher and lower than 50/mm3.FIG. 5: Long-term survival of 80 HIV-infected patients with visceral leishmaniasis receiving and not receiving secondary prophylaxis for visceral leishmaniasis.FIG. 6: Long-term survival of 80 HIV-infected patients with visceral leishmaniasis treated and not treated with highly active antiretroviral therapy (HAART).Discussion Epidemiologic and demographic data Leishmaniasis is a zoonotic disease due to infection with parasites of the genus Leishmania. VL is endemic in the Mediterranean basin and usually affects immunocompetent individuals, predominantly children (30, 40, 53, 54, 67, 164). However, in recent years, VL also has been recognized as an uncommon opportunistic infection associated with some immunodeficiency states such as neoplasms and organ transplantation (4, 16, 22, 31, 60, 74, 96). Since the mid-1980s, VL has been reported increasingly as a complication of AIDS, mainly from countries of southern Europe such as Spain, Italy, and France (11, 52, 54, 70, 93, 113, 141). Coinfection has also been reported from other areas of the world (11, 24, 161, 163, 164), and the expected increase of cases in developing countries, owing to the simultaneous spread of both diseases and their increasingly overlapping geographical distribution, is a matter of concern (54). Since VL is not included among the AIDS-defining diseases, little information on its actual prevalence is available through AIDS case-notification systems. It has been estimated that between 1.5% and 9% of all AIDS patients in southern Europe will develop VL (54, 70, 163), but a slightly slower rate (0.9%) was found in our study. We have observed that the incidence of VL has increased progressively in the last years. As the incidence of VL among the general population has remained fairly constant, AIDS-associated VL seems to explain this situation. HIV-related VL usually appears in patients with advanced immunosuppression. Previous studies have reported that between 42% and 77% of patients have AIDS-defining criteria during the first episode of VL (93, 105, 113, 128, 138, 141) and that the CD4+ lymphocyte count is less than 200/mm3 in 77%-90% of patients (11, 163). In our population, VL was the first severe infection diagnosed in 46.4% of the HIV(+) individuals, a circumstance reported in 13%-22% of patients in other series (105, 113, 141). The age of presentation of VL is 1 of the main differences among HIV(+) and HIV(−) individuals. While HIV-associated VL usually appears in young adults (11, 128, 163), the disease is more frequent in infants and young children among the immunocompetent population (30, 67, 98). We have not observed significant differences in the sex distribution between the 2 groups, but males predominate in both populations. Pathogenesis of HIV/Leishmania coinfection It is well known that the cell-mediated immune response largely determines the outcome of Leishmania infection, and, therefore, only a minority of infected individuals develop clinically apparent visceral disease (18, 19). Two mechanisms have been proposed to explain the strong association between VL and HIV infection. Immunosuppression can lead to clinical leishmaniasis, allowing the reactivation of a latent infection (11, 83, 113) and/or facilitating the development of the active disease after primary Leishmania infection (6
Referência(s)