Visceral leishmanisis in paediatrics: a study of 367 cases in southwest Iran

Introduction

Visceral leishmaniasis (VL), or Kala-azar, is most commonly caused by L. donovani, L. infantum and L. chagasi, but on occasion other leishmania spp. such as L. amazonensis are isolated from patients with typical VL. ¹ A viscerotropic syndrome caused by L. tropica was identified among a small number of American military personnel who served in the Middle East. ² The first case of VL in Iran was reported from the north of the country in 1949. Thereafter, an increasing number of cases have been diagnosed from other parts of the country. ³ The major foci of the disease are located in Meshkin–Shahr in the northwest and Farash–band in Fars Province. Fars Province is located in the southwest of Iran and consists of two geographic regions (a plain region and a mountainous region). The epidemiological form of diseases in Iran is the Mediterranean type, with a canine (dog, fox and jackal) reservoir, and the aetiologic agent is L. infantum which is transmitted by the bite of sand flies. ⁴ We performed a retrospective study of epidemiological, clinical and therapeutic features and the laboratory findings of VL in children.

Methods

This study was conducted at hospitals affiliated to the Shiraz University of Medical Sciences in southwest Iran. The medical records of all children younger than 15 years with a final diagnosis of VL were reviewed from 1996 to 2006. Inclusion criteria were: a positive serological test result (indirect immunofluorescence antibody [IFA] test >1/128) and/or the presence of leishmania in the bone marrow aspiration samples. Data recorded included age, sex, nationality, history of travels, fever, weight loss, anorexia, diarrhoea, cough, abdominal distention, respiratory symptoms, temperature at presentation, lymphadenopathy, hepatosplenomegaly and skin changes. Laboratory records were reviewed for complete blood count, white blood count (WBC) with differential, erythrocyte sedimentation rate, reticulocyte, liver function tests, Widal test for typhoid and brucella, blood film for malaria, blood cultures for bacteria, urinalysis and urine culture, stool for ova and parasites, IFA and chest X-ray. Bone marrow aspirates were taken for L. donovani bodies. Patients were treated with sodium glucantim 20 mg antimony/kg daily for 21 days. Responses were assessed by defervescence, improvement of general condition, weight and anaemia and regression of organomegaly. Drug side-effects also were reviewed.

Results

Of the patients, 62% were boys and 38% girls aged from 3.5 months to 10 years. The disease was most common in those younger than 2 years of age. The majority (40.2%) presented in late winter (March and April), spring (23.5%) and few were seen in summer (18%) and autumn (11.3%). About 80% were from rural areas or were nomads. The majority (98.4%) presented with a long history of fever (especially the nocturnal type), usually lasting for one month or more. Although more patients were febrile on admission, a few (3.8%) were afebrile. Other symptoms included: loss of appetite (64.3%); abdominal distention (50.1%); weight loss (40.6%); cough (38.7%); diarrhoea (37.5%); abdominal pain (13.8%); jaundice (9.6%); and vomiting (5%). Joints pain, haemoptysis and haematemesis were less common (<0.5%). On examination, 96% had pallor. In the majority (91.2%) the spleen was enlarged and in 78% of them it was felt >6 cm below the costal margin. The liver was enlarged in 86.6%; 10.1% had lymphadenopathies which were mostly in the cervical area.

Fever subsided within the first week and hepatosplenomegaly regressed gradually between the second and third weeks. Relapse was observed in (8.2%, 30). These patients received a second course of glucantim plus allapurinol or amphotricin B. The mortality in this series was 7.3%. Twenty patients died during therapy (after receiving between two and 12 doses of glucantim). They all had jaundice and markedly elevated liver enzymes (SGOT > 1000). The cause of death in these patients was secondary bacterial infection (9) and severe bleeding (14). Four of them had positive blood cultures and received appropriate antibiotics in addition to the antileishmania therapy.

Discussion

The results of the present study indicated that VL is a relatively common disease in southwest Iran. VL in southern Iran affects predominantly infants and young children. Several studies in countries where the infantile form of VL is common (Saudi Arabia, Yemen) have shown similar results. ⁵ In Africa and India, the disease affects older children and adults. ⁶ About 62% of the patients were boys and most of the patients were from rural areas or were nomads. Because of age distribution, the disease in Fars Province was thought to be similar to the Mediterranean variety; namely, a zoonosis for which canines are the main reservoir. ⁷ This was also isolated in Yemen and eastern Ethiopia. The clinical manifestations of VL exhibited by the children in this sample were similar to those in previously published reports. ⁸

Seasonal variation was observed, peaking in late winter and spring and then declining sharply in summer which is similar to the variation seen in the Mediterranean. The predominant clinical features seen among our patients were fever, weight loss, anorexia, abdominal distention and hepatosplenomegaly. Again, this is similar to the Saudi Arabia and Indian experiences. ⁹ Lymphadenopathy was less common – it is more common in the African strain of the disease. ⁷

The main haematological findings were anaemia, leukopenia and thrombocytopenia. Anaemia was probably due to a combination of iron deficiency, haemolysis and bone marrow suppression (anaemia of chronic disease). Neutropenia and thrombocytopenia are likely to be due to hypersplenism as they are absent in patients who have had splenectomy. ⁷ Although the liver was enlarged in most patients, liver enzymes and bilirubin showed minimal derangement. Serum albumin was low in the majority, probably as a result of chronic infection and malnutrition. IFA tests were positive in all patients. Cross-reactions occur with leprosy, malaria, schistosomiasis and cutaneous leishmaniasis.

A definite diagnosis of VL depends on the demonstration of amastigotes in tissues. Glucantim was the first-choice treatment and the side-effects were very low. The response to glucantim in our series was excellent which conforms with those shown in another Iranian study and a Saudi Arabia study. ⁹ The Indian variety of the disease also responds well to therapy. VL in Africa seems to be less responsive. Relapse rates after treatment in this study were 8.2%. But in reports from Kenya they ranged from 5% to 36.7%. Patients who do not respond to the initial course of antimony may respond to a second course.

The mortality in this series was 6.2% – similar to those in previously published reports. ⁹ Jaundice and markedly deranged liver functions were found to be bad prognostic indicators. As death occurred early in the course of therapy, it is unlikely to be related to glucantim toxicity.

In summary, in many ways VL in Fars Province bears many clinical similarities to the Indian and Mediterranean strain. All are known to have good response to therapy.