Search

Study Objectives: Pleural tuberculosis (TB) is a diagnostic challenge because of its nonspecific clinical presentation and paucibacillary nature. The inefficiency of conventional laboratory methods and the reliance on pleural biopsy have motivated the evaluation of alternative diagnostic strategies. We have evaluated polymerase chain reaction (PCR) directed to the IS6110 sequence of Mycobacterium tuberculosis, the determination of adenosine deaminase (ADA) activity, and measurement of interferon (IFN)-gamma levels in pleural fluid in the diagnosis of pleural TB.Patients: ADA activity, IFN-gamma levels, and PCR were evaluated in 140 cases of pleural effusion, 42 with confirmed pleural TB, 19 with probable pleural TB, 70 with a nontuberculous etiology, and 9 having an undetermined etiology.Results: ADA activity, IFN-gamma levels, and PCR were 88%, 85.7%, and 73.8% sensitive, respectively, and 85.7%, 97.1%, and 90% specific, respectively, for pleural TB that had been confirmed by either culture or pleural biopsy specimens. The combination of PCR, IFN-gamma measurement, and ADA activity determination allowed the selective increase of sensitivity and specificity for probable and confirmed cases compared to individual methods. Positive and negative predictive values for these individual or combined methods were maintained over a wide range of prevalence of pleural TB in the patient population presenting with pleural effusions. Fever and younger age were associated with tuberculous pleural effusion (p < 0. 0001), while blood in sputum and older age were associated with malignant etiology (p < 0.008).Conclusions: These clinical variables together with the use of ADA activity determination, PCR, and measurement of IFN-gamma levels provide the basis for the rapid and efficient diagnosis of pleural TB in different clinical settings. [ABSTRACT FROM AUTHOR]

Paraflagellar rod proteins (PFR) are a potent immunogen against experimental Trypanosoma cruzi infection. PFR are highly conserved among kinetoplastid parasites. We therefore evaluated the immunogenicity of the Leishmania mexicana pfr-2 gene and protein product in the hamster model of American cutaneous leishmaniasis. Immunization with pfr-2 DNA-induced specific antibody, confirming immunogenicity. Subsequent challenge with 10,000 and 500 stationary phase L. mexicana promastigotes respectively, resulted in delayed appearance of lesions, and significant reduction in lesions post infection in male hamsters, yet exacerbated lesions in female hamsters. Immunization with recombinant PFR-2 protein (rPFR-2) prevented lesion development in female hamsters challenged with L. panamensis, but was ineffective against L. mexicana. Nevertheless, prime boost immunization of female hamsters with rPFR and pfr-2 DNA significantly reduced lesion size following challenge with 500 L. mexicana promastigotes, supporting the relevance of PFR-2 as a potential vaccine constituent.

Intraspecific heterogeneity was demonstrated in the mini-exon gene localization from Leishmania (Viannia) panamensis and L. (Viannia) guyanensis. Different karyotypes were detected in human isolates circulating in endemic areas of Colombia. The presence of mini-exon gene sequences on chromosomes of different sizes, ranging from 370 to 800 kb in L. (V.) panamensis and from 500 to 800 kb in L. (V.) guyanensis, was observed and was neither strain nor species specific. In some cases, hybridization with 2 chromosomes in the same strain was observed. The variability of chromosomal localization of mini-exon gene sequences of these 2 species highlights the genetic variability of the Viannia subgenus and the potential utility of the mini-exon gene as a molecular epidemiologic marker.

Metastatic disease is a major concern of dermal leishmaniasis caused by Leishmania of the Viannia subgenus. The golden hamster provides an experimental model of systemic dissemination and cutaneous metastasis of Leishmania Viannia. We have exploited this model to examine the expression of parasite virulence in cloned populations derived from a strain of L. guyanensis previously shown to be highly metastatic in the hamster. Metastatic capacity manifested as dissemination throughout the lymphoid organs; cachexia and secondary cutaneous lesions were found to differ among clones, yielding a spectrum of virulence. The metastatic phenotype of clonal populations was stable over 5 sequential passages in hamsters. In addition, the low or high propensity to disseminate and produce cutaneous metastatic lesions was reproduced. Capacity to disseminate from the inoculation site was conserved following subcloning of metastatic clones that had been passaged in culture for several generations; clinical manifestations, cachexia, and cutaneous metastatic lesions were variably expressed. Dissemination of parasites and cachexia were significantly related (P = 0.004). Overall, cachexia was an earlier manifestation of dissemination than cutaneous metastases (P < 0.001). The reproducible expression of virulence phenotypes by discrete populations of Leishmania in the golden hamster provides an experimental model for clinically relevant expression of virulence in human leishmaniasis.

In vitro sensitivity to pentavalent antimony (SbV) as meglumine antimoniate (Glucantime) of Leishmania of the Viannia subgenus isolated prior to treatment from patients with uncomplicated cutaneous leishmaniasis was evaluated for intracellular amastigotes in the U-937 human monocytic cell line and log phase promastigotes. The 50% effective dose (ED50) of pharmaceutical and additive-free formulations of Glucantime were determined based on the kinetics of the response of Leishmania Viannia to SbV in vitro. ED50 to SbV was inversely related to time of exposure to drug. The potency of the additive-free formulation of Glucantime was significantly greater than that of the pharmaceutical formulation, irrespective of the parasite form. In vitro sensitivity to SbV ranged from < 5.3 micrograms/ml to > 170.0 micrograms/ml. Under the conditions used, 11 (39%) of 28 strains were sensitive to clinically achievable serum concentrations of SbV. No correlation was observed between the total amount of SbV required for healing of lesions and the in vitro response to the pharmaceutical formulation of Glucantime. In contrast, a significant correlation (P = 0.001) was observed between clinical response and the in vitro sensitivity of promastigotes to the additive-free formulation of Glucantime. The greater potency of the additive-free formulation of Glucantime, the correlation of in vitro sensitivity of promastigotes to this formulation and the clinical response to treatment, and the effect of time of exposure to SbV demonstrate the importance of assay conditions on the outcome and interpretation of in vitro evaluation of drug sensitivity.

During natural infections, Leishmania is in contact with a variety of mononuclear phagocytic cells in different tissues, including resident macrophages and monocytes mobilized to the site of infection from the bone marrow and blood circulation. Because the functional capabilities of fully differentiated macrophages and blood monocytes differ, the outcome of infection by Leishmania may depend upon the stage of differentiation of the host cells. To address this question, we evaluated Leishmania panamensis infection of (1) the human promonocytic/histiocytic cell line U-937 before and after induction of differentiation by phorbol myristate acetate; (2) fresh human peripheral blood monocytes; and (3) macrophages derived from monocytes by differentiation in vitro. Based on the percentage of cells infected and the number of parasites per cell, macrophages derived from monocytes or by induction of differentiation of U-937 cells were significantly more permissive to infection by stationary-phase L. (Viannia) panamensis promastigotes than monocytes. Increasing time and maturation in culture prior to exposure to infective promastigotes was associated with the increased permissiveness of differentiated macrophages to infection (P<0.05). The percentage of cells infected and number of amastigotes per cell increased with time postinfection for both monocytes and macrophages but remained significantly greater for macrophages. The increased expression of CD68, CD16, and lysozyme, and decreased expression of peroxidase by macrophages cultured for 5 days in vitro compared with fresh monocytes, whether adherent or in suspension, supported the distinct maturation status of these cells.

The emergence of Leishmania less sensitive to pentavalent antimonial agents (SbVs), the report of inhibition of purified topoisomerase I of Leishmania donovani by sodium stibogluconate (Pentostam), and the uncertain mechanism of action of antimonial drugs prompted an evaluation of SbVs in the stabilization of cleavable complexes in promastigotes of Leishmania (Viannia). The effect of camptothecin, an inhibitor of topoisomerase, and additive-free meglumine antimoniate (Glucantime) on the stabilization of cleavable DNA-protein complexes associated with the inhibition of topoisomerase was assessed in the human promonocytic cell line U-937, promastigotes of L. (Viannia) panamensis selected for SbV resistance in vitro, and the corresponding wild-type strain. The stabilization of cleavable complexes and the 50% effective dose (ED50) of SbVs for parasites isolated from patients with relapses were also evaluated. The median ED50 for the wild-type strain was 16. 7 microg of SbV/ml, while that of the line selected for resistance was 209.5 microg of SbV/ml. Treatment with both meglumine antimoniate and sodium stibogluconate (20 to 200 microg of SbV/ml) stabilized DNA-protein complexes in the wild-type strain but not the resistant line. The ED50s of the SbVs for Leishmania strains from patients with relapses was comparable to those for the line selected for in vitro resistance, and DNA-protein complexes were not stabilized by exposure to meglumine antimoniate. Cleavable complexes were observed in all Leishmania strains treated with camptothecin. Camptothecin stabilized cleavable complexes in U-937 cells; SbVs did not. The selective effect of the SbVs on the stabilization of DNA-protein complexes in Leishmania and the loss of this effect in naturally resistant or experimentally derived SbV-resistant Leishmania suggest that topoisomerase may be a target of antimonial drugs.

Both host and parasite determinants influence the outcome of Leishmania infections. Human host responses in cutaneous leishmaniasis of limited duration caused by a single species of the Viannia (V) subgenus were studied in skin biopsies obtained from lesions caused by Leishmania (V) panamensis in 31 male patients from the Colombian Pacific Coast. Dermal infiltrates and histopathologic changes were characterized using monoclonal antibodies and an indirect immunoperoxidase method. Dermal distribution of T-cell subpopulations and B-lymphocytes was nonrandom: CD4+ and CD8+ T cells were most frequent in the upper dermis, and B cells were most abundant in the lower dermis. Parasites, macrophages, and neutrophils were localized predominantly in the middermis. Multiple regression analyses to establish associations between lesion type (ulcer, nodule, or papule), immune response data (Montenegro skin test, indirect fluorescence antibody test titers, lymphocyte blastogenesis), and particular cell populations demonstrated statistically significant correlations between CD4+ lymphocytes and macrophages (p < 0.05). CD8+ lymphocytes were associated with plasma cells (p < 0.001), as was the presence of amastigotes (p < 0.05). These associations and the in situ divergence of CD4 and CD8 ratios suggest that prognostic indicators for disease evolution could be identified by prospective analysis of cellular relationships and response to therapy.

The existence of an intracellular stage of Trypanosoma rangeli in the vertebrate host was evaluated by experimental infection of the U937 histiocytic cell line with the San Agustín strain and the Ub66-5b clone. The identity of the parasites at the beginning and end of the experiments was confirmed through biological behavior in the vector and mammal hosts, isoenzymes, polymerase chain reaction (PCR), and monoclonal antibodies. Infectivity to U937 cells of T. rangeli obtained from culture and salivary glands was evaluated under different experimental conditions. These included 34 C vs. 37 C, opsonized vs. nonopsonized parasites, and 2, 4, 6, 24, 48, and 72 hr of cell-parasite contact. Trypanosoma rangeli adopted a characteristic nondividing amastigote-like form within U937 cells, which was different in size (P = 0.001) from Trypanosoma cruzi amastigotes. Culture forms of T. rangeli were more infective than parasites from salivary glands (P = 0.049) but were less infective than T. cruzi (P = 0.0001). Variations in temperature (34-37 C) and complement opsonization did not affect infectivity. Viability of intracellular forms was confirmed by feeding Rhodnius prolixus with T. rangeli-infected cells. Resistance of T. rangeli to the intracellular milieu could be an important mechanism in producing chronic infections in mammals and in the infection of triatomines.

Molecular karyotype and kDNA restriction analyses were utilized to examine the genetic heterogeneity and plasticity of the Leishmania (Viannia) guyanensis strain WHI/BR/78/M5313, composed of metastatic and non-metastatic populations. Cloning revealed that the strain was constituted by multiple closely related populations that were distinguishable by restriction fragment polymorphisms in kDNA. Size polymorphisms in molecular karyotype were not detected. Passage of clones in hamsters and recovery of parasites from cutaneous metastatic lesions yielded evidence of further genetic heterogeneity among some of the progeny populations. Overall, six kDNA minicircle restriction patterns or schizodemes were observed among clones, subclones and progeny. Although the possibility that population heterogeneity was not resolved by cloning cannot be ruled out, subcloning and kDNA restriction analysis to determine whether the putative clones consisted of homogeneous populations showed the schizodeme of subclones of 3 out of 4 clones to be identical to the clone of origin, while a subclone of the fourth had a co-efficient of similarity of 0.95. Metastasis did not segregate with a particular schizodeme: all six restriction profiles were represented among populations isolated from metastatic lesions and some clones with the same restriction profile did not produce metastatic lesions. The strain from which the clones, subclones and progeny were derived had a kDNA restriction pattern identical to the most prevalent schizodeme (38%) among these subpopulations. This finding together with the reappearance of the repertoire of schizodemes found among clones in the populations recovered from metastatic lesions in hamsters inoculated with a single clone, suggest that sequence polymorphisms in kDNA can emerge during infection.

Leishmania parasites were isolated from peripheral blood leucocytes of 2 patients with mucosal disease among a total of 23 parasitologically confirmed cases of leishmaniasis. One had had mucosal leishmaniasis for 4 years and active pulmonary tuberculosis was also diagnosed. The other patient presented a cutaneous lesion on his right leg of 3 months duration and asymptomatic mucosal involvement. He had received intravenous antimonials before isolation of parasites. Both patients had positive indirect fluorescent antibody and Montenegro skin tests. L. (Viannia) braziliensis was isolated from both patients. This culture of parasites from leucocytes provided direct evidence for metastatic spread of Leishmania via the blood.

The pattern and kinetics of internal dissemination and frequency of cutaneous metastatic lesions resulting from experimental infection of golden hamsters with Leishmania (Viannia) panamensis and Leishmania (Viannia) guyanensis were examined. Nineteen strains were evaluated: 16 L. (V.) panamensis isolated from patients and 3 L. (V.) guyanensis, 2 isolated from human cases and 1 WHO reference strain originating from a sandfly vector. Lymphatic dissemination occurred within 3 mo and was observed for 16 of 16 (100%) of L. (V.) panamensis and 3 of 3 (100%) of L. (V.) guyanensis. Parasites were cultured infrequently from liver and spleen: 3 of 125 (2%) L. (V.) panamensis and 1 of 22 (5%) L. (V.) guyanensis. Decreased frequency of isolation from the inoculation site and draining lymph nodes over time was accompanied by increased frequency of isolation from distant lymph nodes. Dilution of triturated tissue samples resulted in an increased efficiency of parasite culture. Both primary lesions and secondary cutaneous metastatic lesions were more severe in hamsters infected with L. (V.) guyanensis than with L. (V.) panamensis. Cutaneous metastatic lesions were produced more frequently by L. (V.) guyanensis, 24 of 46 hamsters (52%), than by L. (V.) panamensis, 28 of 252 hamsters (11%). Individual Leishmania strains displayed distinctive propensities to produce cutaneous metastases, manifested as a reproducible phenotype. Metastatic pathogenicity was independent of the inoculum dose, supporting the dissociation of infectivity and pathogenicity.

Skin biopsies from 221 parasitologically confirmed cases of tegumentary leishmaniasis caused by Leishmania braziliensis spp. were evaluated with respect to histopathology, the qualitative and quantitative nature of the cellular infiltrate, and the presence of Leishmania amastigotes. These variables were cross correlated with the Leishmania-specific immune response, clinical presentation, and response to treatment. Physical evidence of prior leishmanial lesions was associated with the absence of amastigotes (P less than or equal to 0.001) and the presence of giant (P = 0.03) and epitheloid cells (P = 0.03) in the biopsy of the active lesion. The presence of amastigotes was inversely related to the duration of the lesion (P less than or equal to 0.001) and the presence of eosinophils (P less than or equal to 0.01), whereas the presence of adenopathy (P = 0.01), necrosis (P = 0.001), histiocytes (P = 0.001), and increased serum antibody titer (P = 0.02) were directly associated with the presence of amastigotes. The lymphocyte transformation response was correlated with the presence of granulomas (P = 0.001), but showed no correlation with cutaneous delayed type hypersensitivity. The presence of epithelioid (P = 0.04) and giant cells (P = 0.03) was associated with less drug being required to achieve healing. In contrast, necrosis was associated with a greater amount of drug to achieve healing (P = 0.05). The observed correlations between tissue responses and immune and clinical parameters provide further evidence for the role of antibody and other soluble mediators of the cellular immune response in the evolution of disease.(ABSTRACT TRUNCATED AT 250 WORDS)

Leishmanin skin test (LST) antigens prepared from Leishmania braziliensis panamensis were compared with respect to sensitivity, specificity, and side effects. Within the dose range 0.5-3.0 x 10(5) promastigotes of L. b. panamensis and 10 x 10(5) promastigotes of combined L. amazonensis and L. b. panamensis, specificity in healthy controls was nearly 100% for all antigens. Sensitivity increased minimally with increasing dose. Lot-to-lot differences were small. Side effects, such as vesiculation and ulceration at the site of LST application increased with antigen dose. Storage under harsh conditions decreased LST potency but not sensitivity while storage at 2-8 degrees C affected neither potency nor sensitivity. Eighty-five percent of parasitologically diagnosed, LST-positive cases of leishmaniasis remained LST-positive when retested six months to three years later. The LST did not sensitive 19 healthy controls who were skin tested twice or thrice.

Strains of Leishmania braziliensis subspecies isolated from initial and recurrent lesions in 24 patients from the Pacific coast of Colombia were examined for distinguishing polymorphisms by enzyme electrophoresis, restriction endonuclease analysis of kDNA, and molecular karyotyping of nuclear DNA. Recurrent strains from 12 patients (50%) were identical to the initially infecting strain by all methods of characterisation. Phenotypic and genotypic identity, together with clinical data, support endogenous reactivation as the mechanisms of recurrent disease in these 12 patients. 5 of the 24 (22%) recurrent strains differed from the initial strain by all methods. The remaining 7 strain pairs, not separated by enzyme polymorphisms, showed differing schizodeme and/or karyotype profiles. Patients whose recurrent lesions were caused by strains different from those causing the initial lesions had a significantly longer disease-free interval than patients whose lesions were caused by identical strains. Recurrent lesions occurred further from initial lesions in the former than in the latter group. Exogenous reinfection is the most plausible explanation for recurrences due to disparate organisms. These findings have important implications for both treatment evaluation and vaccination strategies for American tegumentary leishmaniasis.

Three Leishmania strains were isolated from cutaneous and mucosal lesions of a Colombian male. These strains, shown to be phenotypically identical based on isoenzyme polymorphisms and monoclonal antibody reactivity, were identified as Leishmania braziliensis guyanensis. Six clones of the mucosal strain were phenotypically identical to the 3 strains. The clinical presentation of the cutaneous lesions and the geographic origin of the infection were consistent with infection by L. b. guyanensis.

Leishmania-specific immunoglobulin subclass response was evaluated in 133 patients infected with Leishmania braziliensis. The indirect immunofluorescent antibody test (IFAT) was employed with amastigotes of L. mexicana amazonensis as antigen. Among the 133 sera obtained at consultation for diagnosis of active lesions, IgM was detected in 54 following absorption with Staphylococcus aureus Cowan strain I, and in 5 sera prior to absorption. IgM reactive with Leishmania antigen was only found in sera from patients whose lesions had evolved over the past two months or less. Leishmania-specific IgG was detected in all sera prior to absorption. Sera obtained at the time of recurrence or after complete healing of lesions presented only specific IgG. The combined use of the Montenegro skin test and specific IgM increased the sensitivity of immunodiagnostic methods in patients with lesions of less than 2 months duration. Normal control volunteers were negative for specific IgM and unreactive to Montenegro skin testing. Among 16 patients with non-leishmanial lesions, 3 with sporotrichosis showed IgG reactive with Leishmania; none, including 4 with lesions of less than two months duration, showed specific IgM. We conclude that in patients infected with L. braziliensis the presence of specific IgG and IgM is associated with the time of lesion evolution and the primary or recurrent nature of the lesions. In addition, the combined use of IgM titer and Montenegro reactivity is of potential utility in the diagnosis of early lesions.

Dermal reactions and IgE levels were compared in 51 asthmatic Colombian children identified on the basis of anthropometric measurements as nutritionally normal (25) or mildly (16) or moderately (10) undernourished. Twenty-five nonatopic children served as controls. Total serum IgE concentrations were significantly elevated in the asthmatic group as a whole. Moderately malnourished (grade II) asthmatic children had more than twice as much serum IgE as normal or mildly malnourished (grade I) asthmatic subjects and seven times more than nonatopic children. Intestinal parasitism did not appear to contribute to these differences in IgE levels. Serum levels of IgA and IgD were similarly elevated in grade II asthmatics. Concentrations of serum IgG, IgM, and C3 and C4 complement were unaffected by nutritional or allergic status. Eosinophilia in nasal mucus was significantly reduced in grade I and grade II malnourished asthmatic children. Among asthmatics, the most frequent dermal reactions were to mite antigens (96%), house dust (67%), and grass pollens (35%). Significant levels of specific IgE were detected by the RAST to two species of mites in nearly all atopic children. There was no apparent influence of nutritional status on the distribution of reactivity or specific IgE assay. The clinical significance of hyperimmunoglobulin E in atopic, moderately malnourished children remains to be elucidated.

Peripheral white blood cells have been shown to bind serum immunoglobins when prepared in low ionic strength sucrose solutions. The presence and distribution of Ig on neutrophils, monocytes and lymphocytes was described using an immunoferritin probe and the electron microscope. It appears that this approach reveals immunoglobulins associated with peripheral white blood cells that are not observed by the usual methods of examining these cell surfaces after preparation at physiological ionic strength.

Seven methods of diagnosing leishmaniasis were compared in 177 patients presenting with lesions of the skin (165) or mucosa (12) in Tumaco and Cali, Colombia. The three methods of visualizing amastigotes in tissue samples (histological staining of tissue sections, impression smears of punch biopsies, and smears of dermal scraping from slits in the lesion margins) were less sensitive than the four Leishmania isolation methods (aspiration of lesion border cultured in biphasic media, aspirate inoculated into hamster nasal tissue, culture of punch biopsy macerate, and hamster inoculation of macerate). The aspirate-culture and biopsy-hamster methods employed in this study proved most sensitive of the four methods for the recovery of parasites. The combined overall sensitivity of the 7 methods was 67% for all enrolled patients and 75% for Montenegro skin test-positive patients. The individual sensitivities for the methods for all patients and Montenegro-positive positive, patients, respectively, were: histopathology 14% and 16%, impression smear 19% and 21%, dermal scraping 22% and 26%, aspirate-culture 58% and 64%, aspirate-hamster 38% and 41%, biopsy-culture 50% and 55%, and biopsy-hamster 52% and 57%. All methods were less sensitive in lesions of greater than 6 months duration than in lesions of more recent onset. Mucosal lesions were best diagnosed by the culture or hamster inoculation of a macerated mucosal biopsy. The diagnosis by inoculation of hamsters was achieved within 2 to 12 weeks, a mean of 34.5 days. Promastigotes were seen on Senekjie's medium within 3-8 days.(ABSTRACT TRUNCATED AT 250 WORDS)

Tegumentary leishmaniasis is highly prevalent in the Pacific coast region of Colombia. We have identified 90 foci of transmission in this region based on 179 parasitologically diagnosed patients. Human transmission occurred in mangrove forests, secondary growth and intervened tropical rain forest. A parasitological diagnosis, that is, either isolation or visualization of Leishmania was made in 68.6% of suspected cases. Three phenotypically distinguishable groups of L. braziliensis were encountered based on isoenzymes: L. b. panamensis variants (82%), variants of L. b. braziliensis (14.5%), and stocks intermediate between L. b. panamensis and L. b. guyanensis reference strains (3.5%). The L. b. braziliensis variants produced cutaneous disease alone relatively infrequently (12% of classified cutaneous stocks) but were more frequently (38% of all mucosal stocks) isolated from mucosal lesions. Leishmania infection of the mucous membranes caused a wide spectrum of disease, severity being closely related to time of evolution. Both contiguous and metastatic spread to the mucous membranes was supported by the clinical course of 19 mucosal cases.

Reproducibility of Leishmania braziliensis s.l. metastatic behavior in hamsters was studied for 9 isolates of L.b. panamensis and 2 of L.b. guyanensis with a previous record of metastasis. Also, the influence of corticosteroids and trauma was evaluated. In the corticosteroid-treated group, metastases appeared earlier than in the nontreated group, and localization at the site of trauma was more frequent (4/9) than in the nontreated hamsters (1/5). Nine of the 11 strains (82%) were capable of reproducing metastatic behavior. Studies on dissemination of L. b. panamensis showed that the regional lymph node is invaded as soon as 5 days postinfection, with further nonhematic dissemination to other tissues and organs in less than 4 wk.

It is generally held that with rare exception Leishmania braziliensis braziliensis is the parasite responsible for the metastatic development of mucocutaneous leishmaniasis in the New World. Yet the infrequency of mucocutaneous disease compared with cutaneous manifestations, and the difficulty of isolating parasites from mucocutaneous lesions have restricted the study of the organisms involved. We here report the biologic, isoenzymatic, and monoclonal antibody specificity characteristic of eight Leishmania isolates obtained from the mucosal lesions of the same number of patients. Individually and collectively, the identifying criteria implicate at least two L. braziliensis subspecies as etiologic agents of mucocutaneous leishmaniasis in Colombia and suggest that a spectrum of intrinsically distinguishable organisms may be involved in this disease form.

The indirect immunoperoxidase method was evaluated in 265 biopsies with the purpose of increasing the sensitivity of the diagnostic histopathology of tegumentary lesions caused by subspecies of the Leishmania braziliensis complex. A diagnosis of leishmaniasis was established by parasitological methods (181) or clinical criteria (12) in 193 patients (72.8%). In the latter group of confirmed cases standard histochemistry and immunoperoxidase were compared with direct examination of tissue scraping and culture of lesion aspirates. The detection and localization of amastigotes was more efficient using the immunoperoxidase method (61.3%) than conventional histopathology with hematoxilin and eosin (34.6%) or direct examination of tissue scraping (43.9%). However, culture of lesion aspirates was the most sensitive procedure (89.8%). The efficiency of the immunoperoxidase method was greater in recent lesions, being positive in 75% of cases with less than 3 months evolution, while 55.6%, 37.5%, and 21.1% of cases with lesion evolution of 3-5.9, 6-11, and 12 months or greater, respectively, were positive. The combined use of the direct examination of lesion scraping and immunoperoxidase applied to histological sections of the biopsy from the lesion border allowed an etiologic diagnosis of 72% of confirmed cases. Cross-reactivity was observed with Paracoccidioides braziliensis but not with Mycobacterium leprae, Sporothrix schenckii, or Histoplasma capsulatum.