From the early 1900s, visceral leishmaniasis (VL; kala-azar) has been among the most important health problems in Sudan, particularly in the main endemic area in the eastern and central regions. Several major epidemics have occurred, the most recent--in Western Upper Nile province in southern Sudan, detected in 1988--claiming over 100,000 lives. The disease spread to other areas that were previously not known to be endemic for VL. A major upsurge in the number of cases was noted in the endemic area. These events triggered renewed interest in the disease. Epidemiological and entomological studies confirmed Phlebotomus orientalis as the vector in several parts of the country, typically associated with Acacia seyal and Balanites aegyptiaca vegetation. Infection rates with Leishmania were high, but subject to seasonal variation, as were the numbers of sand flies. Parasites isolated from humans and sand flies belonged to three zymodemes (MON-18, MON-30 and MON-82), which all belong to the L. donovani sensu lato cluster. Transmission dynamics have not been elucidated fully; heavy transmission in relatively scarcely populated areas such as Dinder national park suggested zoonotic transmission whereas the large numbers of patients with post kala-azar dermal leishmaniasis (PKDL) in heavily affected villages may indicate a human reservoir and anthroponotic transmission. Clinical presentation in adults and in children did not differ significantly, except that children were more anaemic. Fever, weight loss, hepato-splenomegaly and lymphadenopathy were the most common findings. PKDL was much more common than expected (56% of patients with VL developed PKDL), but other post-VL manifestations were also found affecting the eyes (uveitis, conjunctivitis, blepharitis), nasal and/or oral mucosa. Evaluation of diagnostic methods showed that parasitological diagnosis should still be the mainstay in diagnosis, with sensitivities for lymph node, bone marrow and spleen aspirates of 58%, 70% and 96%, respectively. Simple, cheap serological tests are needed. The direct agglutination test (DAT) had a sensitivity of 72%, specificity of 94%, positive predictive value of 78% and negative predictive value of 92%. As with other serological tests, the DAT cannot distinguish between active disease, subclinical infection or past infection. The introduction of freeze-dried antigen and control sera greatly improved the practicality and accuracy of the DAT in the field. An enzyme-linked immunosorbent assay using recombinant K39 antigen had higher sensitivity than DAT (93%). The polymerase chain reaction using peripheral blood gave a sensitivity of 70-93% and was more sensitive than microscopy of lymph node or bone marrow aspirates in patients with suspected VL. The leishmanin skin test (LST) was typically negative during active VL and converted to positive in c. 80% of patients 6 months after treatment. Immunological studies showed that both Th1 and Th2 cell responses could be demonstrated in lymph nodes from VL patients as evidenced by the presence of messenger ribonucleic acid for interleukin (IL)-10, interferon gamma and IL-2. Treatment of peripheral blood mononuclear cells from VL patients with IL-12 was found to drive the immune response toward a Th1 type response with the production of interferon gamma, indicating a potential therapeutic role for IL-12. VL responded well to treatment with sodium stibogluconate, which is still the first line drug at a dose of 20 mg/kg intravenously or intramuscularly per day for 15-30 d. Side effects and resistance were rare. Liposomal amphotericin B was effective, with few side effects. Control measures have not been implemented. Based on observations that VL does not occur in individuals who have a positive LST, probably because of previous cutaneous leishmaniasis, a vaccine containing heat-killed L. major promastigotes is currently undergoing a phase III trial.