A five-year prospective study of cutaneous leishmaniasis in an endemic area of Brazil revealed an annual incidence of disease of 8.1 per 1000 inhabitants and a prevalence of 14.9%. The disease fluctuated as a series of mini-epidemics. Most disease occurred in individuals who were 10-30 years of age. Mucosal disease occurred in 2.7% of patients with primary lesions and occurred a median of six years after this lesion. Disease was more common in males, in those with either large or multiple antecedent skin lesions, and in those with incomplete antimony therapy for the primary lesions. An ELISA was positive in 85% of those tested during the first two years after the primary lesion and remained positive for five to 40 years in 27% of patients. Skin testing was positive in 96% of patients with recent lesions and remained positive in 70% of patients. All patients with mucosal disease had positive serological and skin tests.

The clinicai records o fl8 2 p a tie n ts with cutaneous leishm aniasisprobably due to Leishmania braziliensis braziliensis are analysed. 68% had a single lesion which was usually an ulceron the lower anterior tibial third. M an y had short histories ofone to two months and ali age groups were represented 13% had closed lesions o f a verrucose or plaque like nature.Evolution o f these skin lesions after treatment was related to the regularity o f antimony therapy. Although healing usually occurred in three months, the time to scarring after commencing treatment was variable and related to the size o fth e lesion (p < 0.01). Usually i f sufficient antimony treatment was given the lesion closed.Seven o f the ten patients with initially negative leishmanin skin tests converted to positive after treatment. A significant decline o f indirect fluorescent antibody titres occurred in patients followed, during and after therapy.

L eishm a n ia lp a ra sites were detected in 71. 2% W e have presented evidence elsewhere4 that in this area L eish m a n ia braziliensis braziliensis (Lbb) is the parasite isolated from man in 96.7% of cases. W e will describe in these papers the clinicai presentation and initial evolution of this parasite in man. To date no animal model exists for Lbb and parasitological diagnosis in man is difficult1. Therefore in this first paper we discuss our diagnostic procedures in patients with cutaneous or mucosal disease consídered in the two subsequent papers. M A T E R IA L A N D M E T H O D SPatients on first consultation were alloted an LTB number (Leishm aniasis Três Braços) used for ali subsequent reference. A completed protocol included details of past and present residence for subsequent follow up and history of skin or mucosal lesions. Skin lesions were measured, their localisation recorded on a body map and their features described as regards morphology. A search for previous suggestive skin scars and evidence of mucosal disease was made.M ucosal lesions were examined using a good light source, nasal speculum, tongue depressor and indirect laryngoscopy. Exam ination of the post nasal space was not done.Recebido para publicação em 16/8/84.Two 4mm punch biopsies were taken from the 161

In an analysis o f5 7 p a tien ts m ucosaldisease was com m onestin m ales (7 7 % ) in the third decade o flife although the age range was wide a n d even two children were affected A li but nine patients (1 6 % ) had signs ofcutaneous leishm aniasis but in only 8(14% ) I t is the m ucosal lesions that give m ost cause for concem in human Leishm ania braziliensis braziliensis infections. The difficulty in treating this condition is notorious and our experience in Três Braços confirms this fac t It is in this group that often prolonged hospital admission has been necessary to halt the progress of multilating disease. A lso amphotericin B had to be used in some patients when antimony treatment failed. Both in the hospital and the field ali the deaths we can attribute to this disease over the years have occurred in this group. D eath is usually due to a complex of factors, secondary malnutrition due to difficulty in swallowing, secondary infection of the lungs due to aspiration of infected secretions or suffocation due to closure of the laryngeal aperture. M A T E R IA L A N D M E T H O D SW e have previously described the diagnostic methods and routine procedures adopted for the treatm ent of leishmaniasis in T rês Braços?. Mucosal lesions were recorded on a diagram of the nose throat and larynx, scars of cutaneous infections were des cribed and m easured and a previous history of cutaneous infection obtained. G ranulom a was biopsed using a cutting biopsy punch. The techniques for histology and parasite isolation are as described in the first paper in this series?. Patients were classified as to whether a single mucosal surface or multiple structures were affected.If the patient had never received adequate antimony treatm ent and someone was available to give and check injections he was initially treated in the area. Three series of ten days each to a total dose of glucantime o f 1 gram of drug per kilo body weight per series was used1. This is equivalent to a daily dose of 28 mg Sbv per kilogram body weight. If response was poor and mucosal granuloma persisted we advised the patient be adm itted to our hospital in Brasília.179

Som e years ago w e published observations on the epidem iology o f cutaneous leishmaniasis due to Leishm ania viannia brasiliensis (Lvb) from our fiel area in Três Braços, Bahia, Brazil regarding the occurrence o f human infections using techniques already described4. W e now data o f a further five years observations. The total data for 15 farms under study is set out in the Table 1