Visceral leishmaniasis in patients infected with human immunodeficiency virus

Journal of Infection (I99O)

2I, 261-270

V i s c e r a l l e i s h m a n i a s i s in p a t i e n t s i n f e c t e d w i t h h u m a n immunodeficiency virus C. Montalban,* J. L. Calleja,* A. Erice,* F. Laguna, t B. Clotet,~ D. Podzamczer,~ J. Cobo, lj J. Mallolas,~[ M. Yebra,** A. G a l l e g o t t and the Co-operative Group for the Study of Leishmaniasis in AIDS * The Divisions of Internal Medicine and Infectious Diseases, Hospital Ramdn y Cajal, Madrid, t Instituto de Salud Carlos HI, Madrid, :~Hospital de Badalona Germans Trias i Pujol, Barcelona, ~Hospital Princeps d'Espanya, Barcelona, [[ Hospital La Paz, Madrid, ~ Hospital Clinic i Provincial, Barcelona, ** Clinica Puerta de Hierro, Madrid, JfJfDepartment of Genetics, Universidad Complutense, Madrid, Spain (Accepted for publication 2 May I99O)

Summary We describe 40 HIV-seropositive patients who developed visceral leishmaniasis. All the patients lived in areas endemic for visceral leishmaniasis and belonged to groups at risk for AIDS. Twenty-three patients (57"2 %) had definitive AIDS before or after diagnosis of leishmaniasis and 77"5 % were classified as belonging to CDC group IV. Fever was present in 95 % patients and enlargement of the liver and/or spleen in 92.5 %. Lymphopenia was found in 78"3 %, depression of the absolute number of CD 4 lymphocytes in 9° ~o and depression of the CD4 to CD8 ratio in all evaluated cases but leishmania antibodies were found in only 35"2 %. Parasites were demonstrated in the bone marrow or liver in every case. Thirty patients (75 %) showed an initial good response to antimonial drugs, although the leishmaniasis followed a chronic or relapsing course in I7 (42"5%). HIV-related mortality was 40%. A significant correlation was found only between the relapsing course of the disease and mortality. In a multivariate linear regression model, the relapsing course was the only variable that influenced mortality. Visceral leishmaniasis is an opportunistic disease that should be suspected in HIV-infected patients. We suggest that it should be included in the CDC group IV C-I and considered as a disease indicative of AIDS.

Introduction Visceral leishmaniasis (kala-azar) has been recently recognised as an o p p o r t u n i s t i c infection in i m m u n o c o m p r o m i s e d hosts, 1,2 and isolated cases have b een o b s e r v e d in patients infected with h u m a n i m m u n o d e f i c i e n c y virus ( H I V ) in Spain. 3-14 Pr e vi ous l y, 15 we have descri bed the clinical p a t t e r n o f :~$ T h e Co-operative Group for the Study of Leishmaniasis in AIDS included the following members and Institutions : R. Martinez-Fernandez (Hospital I2 de Octubre), R. Martinez and C. Ga. Aguado (Instituto de Salud Carlos III), M. J. Perez-Elias and L. Buz6n (Hospital Ram6n y Cajal), J. Verdejo (Hospital de la Princesa) and C. Barros (Hospital de Mostoles), Madrid, J. Tor and R. Muga (Hospital de Badalona Germans Trias i Pujol), F. Bolao and F. Gudiol (Hospital Princeps d'Espanya), 1. M. Gatell and J. M. Mir6 (Hospital Clinic i Provincial), Barcelona, Spain. Address correspondence and requests for reprints to: Dr C. Montalban, Servicio de Medicine Interna, Hospital Ram6n y Cajal, Carretera de Colmenar Km 9, Madrid-28o34, Spain. oI63-4453/9o/o6o26I + Io $02.00/0

© I99O The British Society for the Study of Infection

262

C. M O N T A L B A N E T A L .

visceral leishmaniasis in HIV-infected patients whose most outstanding features were the lack of leishmania antibodies and a frequent chronicrelapsing course. We have also pointed out that visceral leishmaniasis is a common opportunistic infection in HIV-seropositive patients, so advancing the possibility that an increasing number of cases may emerge. Such an increase in Spain prompted us to study the characteristics of visceral leishmaniasis in 4o HIV-infected patients as well as the influence of the immunological and clinical features on the course of the disease. Materials and m e t h o d s

In the present collaborative study, that includes most of the cases observed in Spain, we retrospectively reviewed 4o HIV-infected patients who were diagnosed as having visceral leishmaniasis at IO hospitals in Madrid and Barcelona between January I983 and June 1989. Fifteen cases, 2-5' 15 and details of four o t h e r s , 12'13 have been published elsewhere. Antibodies against H I V were detected by E L I S A and confirmed by Western blot analysis. Diagnosis of leishmaniasis or relapses of the disease were made in each case by demonstrating the amastigotes of Leishmania donovani in bone marrow or liver. Cultures for L. donovani were performed in standard media in five cases. Leishmania antibodies were assayed by indirect immunofluorescence (IIF) (29 cases), haemagglutination (HG) (12 cases) and E L I S A (three cases). Titres greater than 4o, 64, and lOO, respectively, were considered to be positive. Counter-immunoelectrophoresis (CIEP) was performed in two cases. Total numbers of CD4 lymphocytes and CD4 and CD8 lymphocyte ratios were determined in 3I cases. T h e system for classifying H I V infection into four mutually exclusive groups 16 and the revised recommendation for case definition for AID S 17 of the Centers for Disease Control, U.S.A. were applied. Patients were followed for a mean of I2"8+_(s.E.) 2"7 months (range o'571 months). All patients were treated initially with a 3 week course of pentavalent antimony in a single daily dose of 85omg. A total of 37 patients received N-methylglucamine (Glucantime) and three others received sodium stibogluconate (Pentostam). Relapses were treated with either antimony salts or alternative therapies (pentamidine, ketoconazole, amphotericin B or clindamycin). Six cases (two cases after relapse and four other cases at diagnosis) received, in addition to pentavalent antimonial drugs, adjuvant prophylactic long-term therapy (three patients 3oo mg daily Allopurinol and three others N-methylglucamine antimoniate in a single monthly dose of 850 mg). Analysis o f data

Student's t-test was used to determine whether the means of binary variable (presence or absence of definitive A I D S , presence or absence of leishmania antibodies, and actual status, dead or alive), of multinomic variables (group I to IV of the CDC classification) and of continuous variables (absolute CD4 lymphocyte count and the CD4 to CD8 lymphocyte ratio) were significantly different in the groups of 17 relapsing and 19 non-relapsing patients. T h e data

Visceral leishmaniasis in H I V patients

263

of four patients were not analysed because of their early death within I month of diagnosis and treatment, it being impossible to assign them into one of either group. T h e correlation coefficients between each pair of the following variables: absolute CD4 lymphocyte count, CD4 to CD8 lymphocyte ratio, presence or absence of definitive A I D S , group of H I V infection, presence or absence of leishmania antibodies, relapsing or non-relapsing course and actual status of the patient (dead or alive), were calculated and tested for significance at the o'05 level. Data on all variables were available for 24 patients. A forward multivariate analysis was performed, considering mortality as a dependent variable and all the other six aforementioned variables as regressor variables. T h e forward analysis was made by means of a multiple linear regression model program contained in a computer package S P S S / P C + . In the multiple linear regression model, the effect of an independent variable may be assessed by calculating the relative weights of each regressor variable on the outcome. A prediction of the value of the dependent variable of a given patient can be calculated from the regression equation by adding to a constant value the weighted sum of the pertinent regressor variables. T o obtain this equation, the best subset of regressor variables was used. We consider that this subset is the one in which the adjusted coefficient of multiple determination is highest. Results Demographic results All patients were white, and lived in Madrid or Barcelona. T h e mean age of the patients was 29"4_(S.E. ) I'O8 years (range, 20-53 years); men outnumbered women by a ratio of 9 : z. All belonged to population at risk for A I D S : 37 (92"5 %) were intravenous drug abusers, one patient was a male homosexual, another was a polytransfused haemophiliac, and in the remainder, remunerated blood donation was the only known risk factor. Nineteen patients had criteria for definitive A I D S previously to or concomitantly with the diagnosis of visceral leishmaniasis and four more fulfilled them after leishmaniasis had been diagnosed. Seven patients (I7"5 %) were classified in group II, two (5 %) in group III and 31 (77"5 %) in group IV of the CDC classification system for H I V infection. Clinical and laboratory findings Prominent findings are shown in Table I. Fever in 38 (95 %) patients and hepatosplenomegaly in 32 (80 %) were the most common presenting features. Splenomegaly alone was found in three (7"5 %) patients, hepatomegaly alone in two (5 %), lymphadenopathy in 23 (57"5 %) and constitutional symptoms (asthenia, anorexia, loss of weight) in 29 (72"5 %). Involvement of the skin was seen in four patients (IO %), two of whom (5 %) had changes of the tongue. Pancytopenia was observed in 33 (82"5 %) patients and single-cell forms of cytopenia were seen in the remainder. Lymphopenia ( < IOOOx io6/1) was present in 29 of 37 patients evaluated (78"3%) (mean 715+68"5) and an absolute decrease of CD4 lymphocytes ( < 4oo x io6/1) was found in 28 of 31 patients (9o %) (mean 204"3 + z7"5). All 3I cases evaluated showed a decrease

264

C. M O N T A L B A N

ET AL.

of the CD4 to CD8 lymphocyte ratio (mean o'5o+o'o6). Hypergammaglobulinaemia was found in 24 of 27 patients (88.8%). Significant titres of leishmania antibodies were found in only I2 of 34 patients (35.2 %) by means of at least one of the various techniques used. T h e I I F test was positive in 1 i of 29 cases (37"9 %), the passive haemagglutination test in two of I2 cases (I6"6 %), E L I S A in one of three cases (33 %), and CIEP was negative in the two cases tested. Discordant results were found in three cases: two patients with I I F titres of i6o had negative E L I S A and haemaglutination tests and a third had positive E L I S A and a negative I F I test. Leishmania donovani amastigotes were found in the bone marrow in 35 of 37 cases (94"5 %), in the liver in seven of eight cases biopsied (87"7 %) and in the skin in four cases. In all four cases, amastigotes were also found in the bone marrow a n d / o r the liver. Leishmania promastigotes were cultured from the bone marrow in three of five cases assayed (6o %). Clinical outcome

Four (Io %) patients with definitive A I D S died immediately after diagnosis without completing a single course of treatment; six (I5 %) had no response to the initial treatment and followed a chronic-relapsing course, while 30 (75 %) showed a good response and were apparently cured after the first course of therapy. Eleven patients (36"6 %) from this last group subsequently suffered one or more relapses (mean 2 +0"38, range 2-5 relapses). In summary, I7 of 4o cases (42"5 %) showed a chronic or relapsing course. None of the patients on prophylactic treatment with either allopurinol or N-methylglucamine relapsed after a mean follow-up time of 6"3 + 1.5 months (range 2-60). Four of the six (66 %) non-responders, six of the I I (54"5 %) responders who later relapsed and three of I9 (I5 %) apparently cured patients died. Altogether, 23 patients were still alive after a mean time of I2"5 + 3"7 months (range I-6o months) while I7 (42 %) had already died. Death in all but one patient was related to H I V infection. Nine (42"8 %) of 2I patients without definitive A I D S before diagnosis of leishmaniasis followed a chronic-relapsing course and I2 (57"2%) had no apparent relapses. In the chronic-relapsing group, four (44%) patients developed definitive A I D S after the second relapse, whereas none in the nonrelapsing group progressed to A I D S after a mean follow-up period of 9"3 + 3"4 months (range 1-39 months). T h r e e of the four patients with unusual manifestations of visceral leishmaniasis (skin a n d / o r tongue) showed a relapsing course; two of them had definitive A I D S previous to leishmaniasis and the remainder developed A I D S after 31 months. Considering the outcome of all 23 patients with definitive A I D S vs the I7 without A I D S , I3 (56"5 %) patients in the A I D S group died and Io were still alive (five with relapsing disease and five apparently cured of leishmaniasis, although the follow-up period of these ' c u r e d ' patients was short at the time of submitting this paper, mean 4"6+ P28, range I-9 months), whereas four patients in the n o n - A I D S group died (23"5 %) and I3 were still alive (two with relapsing disease and i i apparently cured after a mean follow-up time of 7"9 + 3"2 months, range 1-39 months).

Visceral leishmaniasis in H I V patients

265

T a b l e I Clinical and other findings related to visceral leishmaniasis (VL) in 40

patients infected with the human immunodeficiency virus (HIV) Men/women sex ratio Mean age (years)+ S.E. Risk population Intravenous drug abusers Mate homosexual Haemophiliac Remunerated blood donor Definitive AIDS Before or coincident with VL After diagnosis of VL Group of HIV infection II III IV Fever Constitutional symptoms Hepato- and/or splenomegaly Hepatosplenomegaly Hepatomegaly alone Splenomegaly alone Lympadenopathy Skin changes Tongue changes Pancytopenia Lymphopenia (< iooo lO6/1) Hypergammaglobulinaemia Depression of the CD4/CD8 ratio (n > 1"8) Absolute decreased count of CD4 lymphocytes (< 400 IO6/1) Positive for leishmania antibodies Identification of parasites Bone marrow Liver Skin Bone marrow culture positive Initial response to antimonial treatment Early deaths Good response No response Chronic or relapsing course Deaths Deaths related to complication of HIV

36/4 29-4--+ 1"O8(range 2O-53) 37 (92"5%) I (2"5%) I (2'5 %) I (2'5 %) 23/4o (57"5%) 19/4o (47"5%) 4/4o (IO %) 7 (17'5 %) 2 (5 %) 31 (77%) 38/40 (95 %) 29/40 (72"5%) 37/40 (92"5%) 32/40 (8o %) 2/4o (5 %) 3/40 (7"5%) 23/40 (57"5%) 4/40 (io %) 2/40 (5 %) 33/40 (82"5%) 29 (78"3%) 24/27 (88"8%) 31/31 (IOO%) 28/31 (90%) 12/34 (35"2%) 40/4o (ioo %) 35/37 (94'5 %) 7/8 (87"5%) 4 3/5 (60 %) 4/40 (io %) 30/4 ° (75 %) 6/40 (15 %) 17/4o (42"5%) I7/4o (42"5%) 16/4o (4o%)

Analysis o f data T h e comparison by univariate analysis of the groups of patients who followed a chronic-relapsing course or not, showed that only the m e a n n u m b e r of dead patients was m a r k e d l y and significantly (P < o'oo3) higher in the c h r o n i c relapsing group (58 %) t h a n in the non-relapsing group (I5 %). T h e r e were no significant differences between the means o f the remaining variables at the 0"05 level o f significance.

266

C. M O N T A L B A N

ET AL.

In the correlation study, the only non-obvious significant correlation coefficient, being significant at the 5 % level, was that between the relapsing course and mortality (o'47; critical v a l u e _ o'4o). This indicates that patients with a relapsing course tend to have a higher mortality, as appeared in the previous comparison. T h e value of the correlation coefficient between results of serological tests and the relapsing course (0"39) closely approached significance, suggesting that patients with leishmania antibodies tend to follow a chronic-relapsing course. Using the multiple linear regression model, we found only one variable (relapsing course) influencing mortality. By means of a constant value (o'I 5) and the specific relative weight of this variable (o'46), the outcome for each patient may be estimated as follows: mortality = o- 15 + o'46 (relapsing course) T h e multiple correlation coefficient corresponding to this fit is o'48, and that for the full model (including the six variables) is o'52. Discussion

T h e outcome of visceral leishmaniasis depends primarily on cell-mediated responses. U n d e r normal circumstances, T-lymphocytes respond to leishmania antigens by proliferation and by production of lymphokines; these lymphokines, mainly Interferon- 7 (INF-7) and probably other yet unknown factors, can then activate the oxygen-dependent and independent microbicidal systems in the macrophages that are the essential defence mechanisms against L. d o n o v a n i . 18-2° During active leishmaniasis, the production of lymphokines decreases, and there is evidence of broader suppression of lymphocyte functions. 18'19 In most cases, lymphokine production is restored after treatment with antimonial drugs, and microbicidal mechanisms in macrophages are able to kill the parasites and cure the disease. In endemic areas and during epidemics, some patients may be unable to restore lymphokine production, thus permitting amastigotes to persist in macrophages for months or years. This latent infection may follow a chronic, subclinical or even asymptomatic course, ~1'52 and may be activated when new factors depress the immunological defences. Indeed, reactivation of latent visceral leishmaniasis has been recorded in association with chronic diseases, in patients treated with steroids and other immunosuppressor agents and in various subsets of immunocompromised hosts, such as patients with haematological neoplastic disease, autoimmune diseases or renal transplant recipients. ~,2 In the last 3 years, w e 3-5'12'13'15 a n d o t h e r s 6-11'14'23-26 have described isolated cases of visceral leishmaniasis in HIV-infected patients living or travelling to Mediterranean countries. Most cases have been described in Spain, 3-~'26 probably because our country is an endemic area for visceral leishmaniasis, with an increasing incidence of H I V infection and a growing awareness of this association after first description of these cases. T h e increasing occurrence of visceral leishmaniasis in HIV-infected patients and the finding that most recent cases of leishmaniasis in adults seen in our hospitals are H I V seropositive, 4'26 favour the hypothesis that this disease is a common opportunistic infection in HIV-infected patients in Spain.

Visceral leishmaniasis in H I V patients

267

In some of our cases and in patients from another series, 2~ episodes of visceral leishmaniasis have occurred previous to H I V infection, suggesting that the disease in some HIV-seropositive patients may result from the reactivation of latent leishmanial infection in the presence of H I V - i n d u c e d immunosuppression. Although probably biased by the fact that intravenous drug abusers form the leading risk group for H I V infection in Spain, the finding that most of our cases were in intravenous drug abusers suggests that the disease may well be transmitted intravenously by means of hypodermic needles in HIV-infected patients. This is not an unexpected finding~ since h u m a n - t o - h u m a n transmission is common in the Indian form of kala-azar and intravenous transmission may result from blood transfusion in immunocompetent persons. A reported case of vaginal leishmaniasis in an immunocompetent woman, ~7 and the even rarer case of a severe rectal lesion due to leishmania infection in a male homosexual with A I D S , 25 suggest that venereal transmission may also occasionally take place. Although it has been p r o p o s e d that visceral leishmaniasis in immunocompromised hosts shows different clinical manifestations lacking visceromegalies or fever, ~'14 95 % cases in our series had fever and 92"5 % hepatoa n d / o r splenomegaly, most of them having the clinical features of classical kala-azar. Rarer manifestations involved the skin and the tongue. Our patients showed varying degrees of depression of the CD4 to CD8 lymphocyte ratio, while leishmaniasis was seen in patients either with or without definitive A I D S and in the various groups of the CDC classification system for H I V infection. Most cases (77%) were patients classified as group IV on the basis of the presence of other diseases included in the group, indicating that visceral leishmaniasis prevails in the late stages of H I V infection. Our observations confirm the previous finding of an absence of leishmania antibodies in most cases of visceral leishmaniasis in HIV-seropositive patients, 15,26 their being found in only 35"2%. Leishmania antibodies are usually detected in visceral leishmaniasis among other subsets of immunocompromised hosts; 1'2 thus their absence in HIV-seropositive patients seems to be related to the H I V infection. This causes severe B- and T-cell dysfunction that may impair the capacity of these cells to elaborate leishmania and other antibodies. T h e fact that some of our patients had leishmania antibodies whereas others did not suggests that their presence may be related to the degree of immune suppression. In our statistical analysis, however, the presence of antibodies did not correlate with the CD4 to CD8 lymphocyte ratio, the total n u m b e r of CD4 lymphocytes, the group of H I V infection or the presence of definitive A I D S , signalling that their presence may depend on other unknown factors. T h e fact that 45"5 % of our patients followed a chronic-relapsing course, is very remarkable, since this protracted course is found in only Io°//o cases of visceral leishmaniasis in immunocompetent patients in Spain. 2s This relapsing course was seemingly favoured by H I V infection, which also suppresses lymphokine production and results in an impaired production of I N F - 7 in patients with A I D S , so correlating with the inability of monocytes from A I D S patients to phagocytose L. donovani. 29 T h e relapsing course in our patients did .10

J I N 21

268

C. M O N T A L B A N E T A L .

not seem to depend on the degree of immunosuppression, the group of H I V infection or the presence of definitive AIDS. T h e only correlation close to significance was that between serological findings and the relapsing course. Although the specific role of leishmania antibodies has not yet been established, their absence might be expected to parallel severe immunosuppression and a worsened outcome. Even so, we cannot explain the surprising finding that the illness in our patients with leishmania antibodies tended to follow a chronic-relapsing course. In contradistinction to the findings in our previous series, 15 the chronicrelapsing course in the present study was the only factor that significantly influenced mortality, and that was otherwise independent of the immune and clinical state of the patients. T h a t 44 % of the patients without definitive A I D S at the time of diagnosis of visceral leishmaniasis, and who followed a relapsing course, developed full-blown A I D S after the second relapse suggests that relapsing disease may predict the development of A I D S due to other causes. Patients who presented with atypical symptoms tended also to follow a relapsing course and to develop AIDS. Maintenance of prolonged visceral leishmaniasis seems to be a most important immunodepressant event in an HIV-seropositive patient that aggravates the already altered immune responses due to H I V and probably favours the onset of more aggressive and lethal infections, being by itself a decisive risk of mortality. Clinical relapses in all patients were confirmed by identification of the parasites but, in apparently cured patients, neither bone marrow puncture nor culture was routinely performed. We cannot therefore assess whether a real cure of visceral leishmaniasis ever takes place in HIV-infected patients. It would be very important to perform such studies after clinical remission in every case in order to confirm our impression that visceral leishmaniasis behaves as other infections in HIV-seropositive patients. Such infections, caused, for example by Mycobacterium tuberculosis, Salmonella typhi, Toxoplasma gondii, and Pneumocystis carinii usually show a good clinical response to treatment but actually persist as latent infections with a high rate of recurrence requiring prolonged or continuing treatment. Latent visceral leishmaniasis in HIV-infected patients and the dysfunction of T-lymphocytes due to H I V infection that may hinder the restoration of interleukine production after antiparasitic treatment may explain the frequent relapses and the failure of conventional therapies in controlling the disease despite an apparently good initial response. This ultimate resistance to therapy stresses the need for more successful therapies that probably should include either long-term prophylactic treatment or newer therapeutic approaches. In our study, long-term prophylactic treatment seems to have favourably influenced the outcome of the disease, since none of our patients treated with either allopurinol or monthly doses of N-methylglucamine has relapsed. Nevertheless, since the mean follow-up period of the patients was rather short, this apparent protective effect must be confirmed in a prospective study. On the other hand, I N F - 7 therapy may be another alternative. Microbicidal function and its response to lymphokines in macrophages seems to be normal in patients with A I D S , and the apparent usefulness of I N F - y as an adjunct to pentavalent antimony in an experimental model of h u m a n macrophages ~° and

Visceral leishmaniasis in H I V patients

269

in a group o f immunocompetent patients from Brazil with refractory disease ~1 suggests that it might be worth exploring this combined treatment in H I V infected patients. Visceral leishmaniasis must be considered in the diagnosis o f HIV-infected patients with fever, hepatosplenomegaly or haematological abnormalities in endemic areas. Since this disease shares many symptoms with H I V infection and its complications, and owing to the fact that leishmania antibodies may be absent, the diagnosis can be very difficult and must be ruled out only after reliable diagnostic tests such as bone marrow aspiration. Given the usefulness o f the culture, ~6 the bone marrow or any other suspected clinical specimen must be cultured in every case. We have suggested elsewhere ~ that visceral leishmaniasis should be included in the IVE subgroup of the C D C classification system for H I V infections, considering that it is a disease not included in the previous groups but related to a defect in cell-mediated immunity. Furthermore, we have demonstrated that visceral leishmaniasis is a common opportunistic disease, at least in Spain, and that relapsing or atypical disease may predict the development of A I D S due to the acquisition o f other indicator diseases and that it is very likely that chronic-relapsing visceral leishmaniasis can influence the outcome of H I V infected patients. With all these arguments in mind, we suggest that visceral leishmaniasis should be included in the C D C group I V C - I , as is the case with other protozoal infections commonly found in HIV-infected patients, as well as being considered among the diseases indicative of A I D S .

References

I. Badar6 R, Carvalho EM, Rocha H, Queiroz AC, Jones TC. Leishmania donovani: an opportunistic microbe associated with progressive disease in three immunocompromised patients. Lancet I986; i: 647-648. z. Fernandez-Guerrero ML, Aguado JM, Buz6n L et al. Visceral leishmaniasis in immunocompromised hosts. Am J Med I987; 83: IO98-I zo2. 3. Montalban C, Sevilla F, Moreno A, Nash R, Celma ML, Mufioz RF. Visceral leishmaniasis as an opportunistic infection in the acquired immunodeficiency syndrome. J Infect I987; I5 : 247-25o. 4. Martinez Fernandez R, Garcia Diaz JD, Gutierrez Sanchez J, Rioperez Carmena E, Valdivieso Varela L, Alonso Navas F. Leishmaniasis visceral en pacientes adictos a drogas pot via paraenteral, VIH positivos. Med Clin (Barc) 1987; 88: 5o9-5I I. 5- Yebra M, Segovia J, Manzano L~ Vargas JA, Bernaldo de Quir6s L, Alvar J. Disseminatedto-skin kala-azar and the acquired immunodeficiency syndrome (Letter). Ann Intern Ailed I988; Io8: 49o-49I. 6. Franco-Vicario R, Beltr~in de Heredia JM, Rojo P, Hermosa C. Leishmaniosis visceral asociada al sindrome de la inmunodeficiencia adquirida (Letter). Med Clin (Barc) I987; 88: 565-566. 7. Medrano Gonzalez F, Alem~in Lorenzo A, Beato Perez JL. Leishmaniosis visceral de curso fatal asociada a infecci6n por H T L V - I I I (Letter). Med Clin (Barc) I986; 87: 78o-78I. 8. De la Loma A, Alvar J, Martinez Galiano E, Blazquez J, Alcahi Mufioz A, Najera R. Leishmaniasis or AIDS ? (Letter). Trans R Soc Trop Med Hyg I985: 79: 421-422. 9. Guix J, Aguilar E, Gabriel F, Labios M, Escrig V. Leishmaniasis visceral, adicci6n a drogas por via paraenteral y seropositividad frente al HIV (Letter). Enf Infec Microbiol Clin I987; IO: 636-637. Io. Olivfin A, Reparaz J, Uriz J, Sohl J. Leishmaniasis e infecci6n por HIV (Letter). Enf Infec Microbiol Clin I989; 7: 60. 10-2

270

C. M O N T A L B A N E T A L .

I I. Rivero A, Santos J, Marquez M, Gavilan JC, Carralero C. Leishmaniasis visceral asociada a infeccion por el HIV (Letter). Enf Inf Microbiol Clin I989; 7: 59I2. Laguna Cuesta F, Garcia Aguado C, Puente Puente S, Mateu Paris B. Leishmaniasis visceral e infecci6n por HIV: tres nuevos casos (Letter). Rev Clin Esp I989; I84: 267-268. I3. Ribelles M, Casals A, Vails J, Clotet B. Leishmaniasis visceral de evoluci6n benigna en un paciente infectado por el virus de la inmunodeficiencia humana (Letter). Med Clin (Barc) r989; 92: 679. 14. Grau JM, Boch X, Salgado AC, Urbano-Marquez A. Human immunodeficiency virus (HIV) and aplastic anemia. Ann Intern Med r989; n o : 576-577. I5. Montalban C, Martinez Fernandez R, Calleja JL et al. Visceral leishmaniasis (kala-azar) as an opportunistic infection in patients infected with the human immunodeficiency virus in Spain. Rev Infect Dis i989; I I : 655-66o. I6. Centers for Disease Control. Classification system for human T-lymphotropic virus type III/lymphadenopathy-associated virus infection. Ann Intern Med I986; lO5: 234-237. I7. Centers for Disease Control. Human immunodeficiency virus (HIV) infection codes: official authorized addendum ICD-9-CM (Revision no. i) effective x/I/88. M M W R I987 : 36 (Suppl 7): IS-2oS. I8. Pearson RD, Wheeler DA, Harrison LH, Kay HD. The immunobiology of leishmaniasis. Rev Infect Dis I983; 5:9o7-927 • x9. Carvalho EM, Badar6 R, Reed SG, Jones TC, Jonshon WD Jr. Absence of gamma interferon and interleukin-2 production during active visceral leishmaniasis, ff Clin Invest I985 ; 76: 2o66-2069. 20. Murray HW, Rubin BY, Rothermel CD. Killing of intracellular Leishmania donovani by lymphokine-stimulated human mononuclear phagocytes : evidence that interferon-7 is the activating lymphokine, ff Clin Invest 1983 ; 72 : 15o5-I 5 xo. 2L Badar6 R, Jones T C , Carvalho E M et al. N e w perspectives on a subclinical form of visceral leishmaniasis. J Infect Dis z986; x54: ~oo3-IOI r. 22. Pampiglione S, Manson-Bahr PEC, Giungi F, Giungi G, Parenti A, Trotti GC. Studies on Mediterranean leishmaniasis. 2. Asymptomatic cases of visceral leishmaniasis. Trans R Soc Trop Med Hyg I974; 68: 447-453. 23. Senaldi G, Cadeo G, Camevale G, Di Perri G, Carosi G. Visceral leishmaniasis as an opportunistic infection (Letter). Lancet r986; i: Io94. 24. Clauvel JP, Couder LJ, Daniel MT, Rabian C, Seligmann M. Visceral leishmaniasis complicating the acquired immunodeficiency syndrome (AIDS) (Letter). Trans R Soc Trop Med Hyg r986: 8o: I o I o - I o r I . 25. Rosenthal PJ, Chaisson RE, Hadley WK, Leech JH. Rectal leishmaniasis in a patient with the acquired immunodeficiency syndrome. A m J Med I988 ; 84: 3o7"-3o9. 26. Berenguer J, Moreno S, Cercenado E, Bernaldo de Quiros JCL, Garcia de la Fuente A, Bouza E. Visceral leishmaniasis in patients infected with human immunodeficiency virus (HIV). Ann Intern Med I989; I I I : i29-I32. 27. Symmers WSC. Leishmaniasis acquired by contagion: a case of marital infection in Britain. Lancet I96o; i: r27-I32. 28. Mariscal Sistiaga F, Dominguez Moreno B, Lenguas Portero F, Martinez Forde JM, Galvfin Guijo B. Kala-azar en el Hospital de enfermedades infecciosas. Revisi6n de IO4 casos. Rev Clin Esp I98O; x59: 47-5o. 29. Murray HW, Scavuzzo D, Jacobs JL et al. In vitro and in vivo activation of human mononuclear phagocytes by interferon-y: studies with normal and AIDS monocytes. J lmmunol r987; X38: 2457-2462. 30. Murray HW, Berman JD, Wright SD. Immunochemotherapy for intracellular Leishmania donovani infection: y-interferon plus pentavalent antimony. J Infect Dis I988; *57: 973-978. 3 r. Badar6 R, FalcoffE, Badar6 FS et al. Treatment of visceral leishmaniasis with pentavalent antimony and interferon gamma. N Engl J Med r99o; 322: I6-2I.