Bacterial and Mycobacterial Infections and Leishmaniasis

Bacterial and Mycobacterial Infections and Leishmaniasis

Toxin Involvement in Staphylococcal Scalded Skin Syndrome

Introduction.—Strains of Staphylococcus aureus isolated from children with staphylococcal scalded skin syndrome (SSSS) were examined for the production of staphylococcal exfoliative toxin A and toxin B, toxic shock syndrome toxin-1, and enterotoxins A-E. The designation SSSS has included generalized exfoliative syndrome, bullous impetigo, and staphylococcal scarlet fever, but the last of these diseases has similarities with staphylococcal toxic shock syndrome.

Bacterial and Mycobacterial Infections and LeishmaniasisMethods.—Sixty cases fulfilling the definition of SSSS were identified from records of the French National Reference Center for Staphylococci for the years 1985 through 1996. Fifteen children were given a diagnosis of generalized exfoliative syndrome, 28 of bullous impetigo, and 17 of staphylococcal scarlet fever. The mean ages of the patients in the 3 groups did not differ significantly (1.5, 2.1, and 2.6 years, respectively). Eighty-five S. aureus isolates were obtained from the 60 patients.

Results.—All 60 isolates were positive for the production of 1 or more staphylococcal exotoxins. All strains isolated from children with generalized exfoliative syndrome or bullous impetigo produced staphylococcal exfoliative toxin A and/or B. When injected subcutaneously into newborn mice, all strains from these 2 groups caused a Nikolsky’s sign, with gentle pressure causing slippage of the upper epidermal layer. Only 1 of the 17 cases of staphylococcal scarlet fever, however, was found to have exfoliative toxins, and only 1 was associated with a positive Nikolsky’s sign.

Sixteen S. aureus strains from staphylococcus scarlet fever produced toxic shock syndrome toxin-1 and/or an enterotoxin.

Conclusion.—All S. aureus strains isolated in association with a generalized exfoliative syndrome or bullous impetigo produced exfoliative toxins exclusively, and all cases tested exhibited a positive Nikolsky’s sign. Staphylococcal scarlet fever has been considered a milder manifestation of SSSS, but only 1 of the 17 cases in this series had exfoliative toxin production and a positive Nikolsky’s sign. These findings suggest that staphylococcal scarlet fever may be an abortive form of staphylococcal toxic shock syndrome or a separate disorder.

 These authors present a convincing argument in support of a new classification of staphylococcal diseases. Traditionally, staphylococcal skin disease has been subdivided into the generalized exfoliative syndrome (SSSS), staphylococcal scarlet fever, and bullous impetigo, all thought to be caused by the exfoliative toxins A and B. In this study, however, S. aureus associated with staphylococcal scarlet fever almost always produced toxic shock syndrome toxin-1 and/or enterotoxin, rather than the exfoliative toxins A and B. This supports the conclusion that staphylococcal scarlet fever is more closely related to toxic shock syndrome—in which the etiologic toxin is toxic shock syndrome toxin-1 or enterotoxin—than to SSSS or bullous impetigo. This new classification is consistent with clinical findings, e.g., both strawberry tongue and palmar/plantar desquamation are found in staphylococcal scarlet fever and toxic shock syndrome but not in SSSS or bullous impetigo.

Comparison of Oral Cephalexin, Topical Mupirocin and Topical Bacitracin for Treatment of Impetigo

Background.—In a previous trial, the authors found that Staphylococcus aureus was the most frequent cause of impetigo in their study populations. Cephalexin was the best treatment, although erythromycin was nearly as good; penicillin V was ineffective. As erythromycin-resistant S. aureus becomes an increasing problem worldwide, there is concern that erythromycin may not be the optimal treatment for impetigo. Neither mupirocin nor bacitracin has ever been compared with oral cephalexin for the treatment of impetigo. These 3 treatments were compared in a prospective, double-blind, placebo-controlled trial.

Methods.—Participants included 26 children with impetigo. They were randomly assigned to receive cephalexin monohydrate suspension, 50 mg/kg/day in 3 divided doses, plus a placebo topical ointment; mupirocin 2% ointment, applied to the affected areas 3 times daily, plus an oral liquid placebo; or bacitracin 500 units/g ointment, applied 3 times daily, plus an oral liquid placebo. All treatments were given for 10 days. Evaluations were performed in blinded fashion at 3-5 and 8-10 days.

Results.—The 3 groups were similar in terms of sex, age, and lesion size, type, and distribution. There were no treatment failures with mupirocin or cephalexin, whereas most patients failed treatment with bacitracin. Cephalexin and muciprocin were equally effective, and both were more effective than bacitracin. Of 22 patients who underwent lesion culture, all had S. aureus. Fourteen percent had group A -hemolytic Streptococcus in addition. Susceptibility testing in 21 S. aureus isolates showed penicillin resistance in 95%, erythromycin resistance in 19%, and cephalexin resistance in none.

Conclusions.—A high rate of treatment failure occurred in patients with impetigo treated with bacitracin. Thus bacitracin should no longer be used for the treatment of impetigo. Topical mupirocin and oral cephalexin are equivalent in their effectiveness. Although this study is small, the difference in treatment results is large.

 I generally have been of the opinion that the purported benefits of mupirocin over currently available over-the-counter antibiotics such as bacitracin and neosporin were more perception than reality. Specifically, I assumed that the manufacturer of mupirocin was able to claim equivalence with oral antibiotics simply because that company did the study to compare the 2 modalities, whereas the manufacturers of bacitracin and neosporin had not taken the time and effort to assess their effectiveness vs. oral therapy. This study, which shows the obvious superiority of mupirocin over bacitracin, has changed my thinking.

Role of Benzathine Penicillin G in Prophylaxis for Recurrent Streptococcal Cellulitis of the Lower Legs

Purpose.—Cellulitis of the lower leg most commonly results from infection with streptococci. Staphylococcus aureus and other gram-negative rods also may be causative organisms. Although the cellulitis responds to treatment with antibiotics, recurrences are common. The slow-releasing agent benzathine penicillin G is commonly used to prevent recurrent rheumatic fever. This agent was tested for its ability to prevent recurrences of cellulitis.

Methods.—The study included 115 patients with confirmed or presumptive streptococcal cellulitis of the lower legs. All were advised to have monthly intramuscular injections of benzathine penicillin G 1.2 megaunits per month for 1 year for prevention of recurrent episodes of cellulitis. Thirty-one patients consented and completed 1 year of prophylaxis; the remaining 84 patients, who declined or received incomplete prophylaxis, served as controls.

Results.—The recurrence rate was 13% in patients who received prophylaxis and 19% in those who did not. Of the total sample, 50% had factors predisposing to cellulitis, mainly reflecting impairment of the local circulation. The recurrence rate with benzathine penicillin G prophylaxis was zero among patients free of predisposing factors but 20% in those with predisposing factors.

Conclusions.—In patients who lack predisposing factors for cellulitis, penicillin G prophylaxis can prevent recurrent attacks of cellulitis. There is no such benefit for patients with predisposing factors. A shorter prophylactic interval or increased dosage might improve prophylaxis among patients with predisposing factors.

 Predisposing factors for recurrent streptococcal cellulitis include diabetes mellitus; impaired venous drainage; lower leg edema; a previous tibial fracture; and, in the patient population studied, filariasis.

Culture-confirmed Infection and Reinfection With Borrelia burgdorferi

Introduction.—Caused by infection with the spirochete Borrelia burgdorferi, Lyme disease is the most common vector-borne disease in the United States. No microbiological evidence has been found to substantiate reinfection, although it has been described clinically. The occurrence of reinfection was documented by isolating B. burgdorferi during 2 different clinical episodes of Lyme disease.

Case Report.—Woman, 60, with erythema migrans lesions on her upper right thigh underwent a biopsy that showed the presence of B. burgdorferi. After 14 days of oral amoxicillin therapy, her rash resolved. Two years later, the patient had an erythema migrans lesion on the right popliteal fossa and a skin biopsy specimen was negative for B. burgdorferi. After 10 days of oral therapy with doxycycline, the rash resolved. One year later, the patient had an erythema migrans lesion on the left axilla. A biopsy culture revealed B. burgdorferi, and the rash resolved after 10 days of oral doxycycline therapy.

Results.—Polymerase chain reaction of the 1991 and 1994 isolates showed that the 2 isolates were different strains of B. burgdorferi. This suggested reinfection and not reactivation of a previous infection.

Conclusion.—This is the first reported case in which B. burgdorferi was isolated during separate episodes of erythema migrans. Two distinct molecular types of B. burgdorferi were involved, according to polymerase chain reaction-restriction fragment length polymorphism analysis.

 Unfortunately, infection with B. burgdorferi does not seem to protect against reinfection. Patients traveling through endemic areas must take appropriate precautions to prevent tick bites. These findings may hinder the development of an effective Lyme disease vaccine.

Extraordinary Case Report: Cutaneous Anthrax

Introduction.—Anthrax is rarely observed in the United Kingdom. Cutaneous anthrax is seen in about 95% of patients with this infection. A patient was seen with cutaneous anthrax, which was initially misdiagnosed.

Case Report.—Man, 57, was seen in the emergency department after a 4-hour history of feverishness and collapsing upon standing. He reported he was bitten by an insect on the left upper chest 1 day previously when working as an electrician on a large ventilation system. A necrotic lesion was observed on the chest that was surrounded by an area of erythema and edema. He had a fever of 38.2oC and was hypotensive. History was negative for travel abroad, medications, or any significant medical history. He was given the diagnosis of streptococcal cellulitis and septicemia and IV benzyl penicillin, floxacillin, and IV fluids were administered immediately. His antistreptolysin 0 titer was slightly elevated. His condition worsened and renal failure and jaundice developed. Increasing erythema was seen in the necrotic area. The chest lesion was excised 36 hours after his arrival at the emergency department because of concern about necrotizing faciitis. Blood cultures taken upon admission revealed Bacillus anthracis. He improved and was discharged 2 weeks after admission. Further history revealed that he had been working shirtless for a leather company and was exposed to untreated leather 1 week before symptom onset. This was the likely source of his infection. He had not been advised about vaccination or safety because he was a casual laborer.

Histopathology.—Massive edema of the upper dermis with attenuation of the overlying epidermis was observed in the skin lesion. The deep dermis and subcutaneous fat revealed hemorrhage, massive fibrin deposition, and a dense infiltrate of lymphocytes, histiocytes, and abundant neurophils with abscess formation. Vascular thromboses were seen. The initial gramstain was negative, but repeat staining confirmed the presence of large gram-positive rods in the papillary dermis.

Conclusion.—The rarity of cutaneous anthrax makes it easy to not even consider the diagnosis. Because patients do contract this infection sporadically, clinicians and pathologists need to be aware of its possible occurrence.

 A high index of suspicion is necessary to diagnosis a rare disease. As always, a correct diagnosis depends as much on the physician’s skill in taking a good history as in performing a good physical examination.

A Man Who Pricked His Finger and Smelled Putrid for 5 Years

Introduction.—When clostridial infections such as gas gangrene occur in humans, they are usually severe. Although clostridia are part of the normal gastrointestinal flora in human beings, they are not generally found as skin commensals. A patient with an untreatable clostridial infection of the hand complicated by a foul odor was seen.

Case Report.—Man, 29, came to the hospital with an erythematous finger with a distinct odor after having pricked the linger with a chicken bone at work. The infection did not resolve with flucloxacillin, ciprofloxacin, erythromycin, and metronidazole. There was no foreign body, pus, or soft-tissue damage. Skin biopsy culture grew a Clostridium novyi type B-like organism that was highly sensitive to antibiotics in vitro. However, the infection could not be cured by prolonged antibiotic treatment or hyperbaric oxygen therapy. Three different clostridia were cultured in skin biopsy specimens taken all the way up the arm to the chest. The patient had no apparent immunosuppression, but did have a precipitating IgM antibody to the infecting organism in serum. Many different treatments were tried, including isotretinoin, psoralen ultraviolet light therapy, colpermin, probanthene, chlorophyll, and antibiotic withdrawal. Even when the clinical condition of the hand improved, the foul odor remained. The putrid smell was almost intolerable in close quarters. Although similar, the initially isolated organism did not have the carbohydrate fermentation reactions characteristic of C. novyi type B. The patient was still carrying the same 3 clostridial species in his skin 5 years after the initial injury.The odor has caused the patient chronic disability and social isolation. Studies using air extraction techniques and gas liquid chromotography have confirmed that the odor is caused by substances produced by bacterial metabolism.

Discussion.—A unique clostridial infection of the hand that does not respond to treatment is presented. The authors speculate that this patient has some immunologic “blind spot” that permits colonization with spore- forming organisms.

 The authors conclude their short report with a plea for assistance from colleagues who may have encountered a similar case. Any suggestions?

The Histopathologic Spectrum of Erythema Induratum of Bazin

Introduction.—The worldwide increase in the annual incidence of tuberculosis has revived interest in cutaneous tuberculosis. One type of cutaneous tuberculosis, erythema induratum of Bazin (EIB), occurs mainly on the calves of women with tuberculin hypersensitivity. The recent detection of Mycobacterium tuberculosis DNA in skin lesions of EIB proved that mycobacterial components are involved in the pathogenesis of the disease. Skin biopsy specimens from 20 patients with well-documented EIB were examined to record the histopathologic spectrum of EIB.

Methods.—The patients, all women, had a mean age of 29 years. None had overt, active tuberculosis or systemic disease. Characteristic skin lesions were present in all cases, and skin biopsy specimens exhibited at least 3 of the following histopathologic features: septolobular or lobular panniculitis, granulomatous inflammation, primary vasculitis, and significant subcutaneous necrosis. Each case was classified as either focal panniculitis or diffuse panniculitis.

Results.—The 6 cases classified as group I exhibited focal septolobular panniculitis in close association with a single muscular artery or small vessel with primary neutrophilic vasculitis. Diffuse septolobular panniculitis was observed in the 13 group II cases, together with primary neutrophilic vasculitis of either large or smaller vessels. Classification was impossible in the remaining case because of an inadequate specimen. Both groups exhibited varying combinations and degrees of acute and chronic inflammation, coagulative and caseation-like necrosis, and granulomatous inflammation. Granulomas were generally poorly developed. Inflammation and necrosis were more extensive in group. Immunostaining, performed in 5 cases, showed S100+ antigen-presenting cells. Macrophages and T-lymphocytes were present in moderate to large numbers in all 5 cases, but B-lymphocytes were rare.

Conclusion.—There is evidence for a direct causal relationship between tuberculosis and EIB. Skin lesions in all of these patients cleared on antituberculous drugs. Primary vasculitis is almost certainly the initial event in the disease. The most prominent histologic findings in a specific skin lesion biopsy specimen will determine the histopathologic differential diagnosis of EIB.

 Schneider and Jordaan review the primary histologic features of 20 patients with well-documented erythema induratum of Bazin, an eruption that is associated with tuberculosis (i.e., a tuberculid). Two major histopathologic groups were identified. One group demonstrated focal septolobular panniculitis associated with focal neutrophilic vasculitis in a single muscular artery or small vessel. The second group demonstrated more diffuse septolobular panniculitis, again with leucocytoclastic vasculitis. Granuloma formation, which included tuberculoid, sarcoidal, and palisaded patterns, was common.

Detection of Mycobacterium tuberculosis DNA in Lobular Granulomatous Panniculitis (Erythema Induratum-Nodular Vasculitis)

Introduction.—Chronic and recurrent, tender, subcutaneous, and sometimes ulcerated nodules of the lower leg are characteristic of erythema induratum of Bazin. There is controversy over the tuberculous origin of this condition, particularly because attempts to cultivate mycobacteria from the lesions have been unsuccessful. The controversy was renewed with the recent detection of mycobacterial DNA in lesions of erythema induratum of Bazin. In this study, cases of lobular granulomatous panniculitis associated with Mycobacterium tuberculosis were identified using polymerase chain reaction (PCR) amplification.

Methods.—There were 65 patients in an 8-year period who had 72 skin biopsies that produced a histologic diagnosis of erythema induratum or nodular vasculitis. The age of the patients ranged from 19 to 80 years. All patients had variably tender, erythematous to violaceous, subcutaneous nodules or plaques. The outcome measure was detection of a 123-base pair fragment of the IS6110 insertion sequence specific for M. tuberculosis complex.

Results.—In 77% of the skin biopsy specimens, the results of PCR amplification were positive for M. tuberculosis complex DNA. With respect to the age of the patients, reactivity to purified protein derivative, ulceration of the nodules, personal and family history of tuberculosis, abnormal results of a chest x-ray examination, and PCR results, no significant difference could be detected. In the PCR-positive group, the presence and degree of necrosis on histologic examination were significantly higher. The presence of vasculitis, number of giant cells, degree of granulomatous infiltrates, and presence of well-organized granulomas were not associated with PCR results.

Conclusion.—A considerable number of skin biopsy specimens of lobular granulomatous panniculitis can detect the DNA of M. tuberculosis. The results of PCR amplification could not be accurately predicted by any of the clinical and histologic variables evaluated.

 Although M. tuberculosis is clearly implicated in at least some cases of lobular granulomatous panniculitis, it has not been determined whether the detection of mycobacterial DNA predicts a response to antituberculous treatment.

Development of Resistance to Clarithromycin After Treatment of Cutaneous Mycobacterium chelonae Infection

Background.—Mycobacterium chelonae is an atypical mycobacterium that is resistant to most first-line antituberculous drugs and many oral antimicrobials. This report describes a patient with cutaneous M. chelonae infection who developed resistance to single-drug therapy with clarithromycin.

Case Report.—Woman, 66, with long-term pemphigus vulgaris, had multiple erythematous, slightly tender nodules develop on her left leg. The pemphigus vulgaris was being treated successfully with prednisone and azathioprine. She began a course of minocycline until the culture results were known. Once culture revealed M. chelonae infection, therapy was switched to clarithromycin, 500 mg twice daily. Improvement was rapid, and by 7 weeks only erythematous macules remained and she discontinued taking clarithromycin. However, 2 months later the nodules recurred, and a subsequent 3-month course of erythromycin ethylsuccinate provided only mild improvement. Culturing at that time showed an isolate that was resistant to, among other drugs, clarithromycin. The only drug with any potential activity was ciprofloxacin, which she began taking in conjunction with azithromycin. However, skin lesions have not significantly improved at the time of this writing.

Conclusions.—Despite an initial good response, this patient rapidly developed resistance to monotherapy with clarithromycin. Thus to treat M. chelonae infection, clarithromycin should be used in conjunction with another oral antimicrobial selected on the basis of susceptibility testing.

 Most authorities now recommend combination therapy for cutaneous atypical mycobacterial infections, as monotherapy often leads to drug resistance. Although typically these cutaneous infections remain localized and do not disseminate to other organs, several months of therapy is necessary to definitively eradicate the organism.

Pregnancy in Patients With Hansen Disease

Purpose.—Despite the introduction of multidrug therapy, the incidence of Hansen disease (leprosy) has not fallen to the extent anticipated. Some patients with this disease are of childbearing potential. There have been few reports of pregnancy in Hansen disease during the last 2 decades. Three cases of pregnancy in patients with Hansen disease are reported.

Patients.—The 3 pregnant patients were identified from a group of 40 patients with Hansen disease over a 6-year period. One patient had initial symptoms of Hansen disease appear during pregnancy. She began treatment with dapsone and rifampin, and 18 months later had no signs of cutaneous disease and no detectable nerve dysfunction. The second patient was treated for lepromatous Hansen disease diagnosed shortly after childbirth. Two years later, when pregnant again, symptoms returned; the clinical impression was type 2 erythema nodosum leprosum reactional state, though the pathologic interpretation was disease reactivation. The patient responded to dapsone monotherapy. In the third case, a patient had been receiving multidrug therapy for approximately 1 year and was clinically clear when she became pregnant. She received dapsone monotherapy for the rest of her pregnancy. She had no disease relapse or reaction, and gave birth to a healthy infant at term.

Discussion.—The decrease in cellular immunity during pregnancy allows proliferation of Mycobacterium leprae, which can lead to precipitation or worsening of Hansen disease. With careful management of reactional states and single-agent treatment with dapsone, permanent nerve damage can be avoided. The infant is usually at lower risk than the mother. However, maternal antibacterial treatment must be selected with an eye toward avoidance of teratogenicity and in utero adverse effects.

 Dr. Lyde reports the outcome of 3 pregnancies in women with Hansen disease. Although the average clinician is unlikely to see these 2 conditions together, there are important factors of the disease and its treatment that need to be considered for the safety of both the mother and the child. The author reviews these nicely in this article, which should be studied by anyone caring for patients with Hansen disease.

A 2-Year Study of Liquid Nitrogen Therapy in Cutaneous Leishmaniasis

Introduction.—In Jordan, cutaneous leishmaniasis is caused by Leishmania major and less commonly by Leishmania tropica. The disease has a 2-month incubation period and lesions can be active from 2 to 12 months. The lesions are anthroponotic (dry type), or zoonotic (wet type). Those in poor rural communities are at the most risk, and 90% are between 18 and 30 years of age. Antimonial and sulfonamide therapy are the standard forms of treatment, and liquid nitrogen is a newer treatment.

Methods.—A total of 468 patients with cutaneous leishmaniasis received 1 of 3 therapies: 293 were treated with liquid nitrogen cryosurgery; 146 were treated with 10 mg/kg per day of pentavalent antimonials for 2 weeks; and 27 were treated with sulfamethoxazole 400 mg and trimethoprim 80 mg, 2 tablets twice a day for 2 weeks.

Results.—There was a significant and effective clinical response among the 215 patients who were followed up after liquid nitrogen therapy. No systemic adverse effects occurred. There were infrequent mild cutaneous complications, including scar formation, hypopigmentation, and hyper-pigmentation.

Conclusions.—For cutaneous leishmaniasis, liquid nitrogen therapy may be considered an effective treatment. It can be used in an outpatient setting and has few systemic adverse effects. The mainstay of cutaneous leishmaniasis therapy continues to be pentavalent antimony, despite its adverse effects including headache, syncope, myalgias, arthralgias, and seizures.

 This brief article from Jordan describes successful cryotherapy of cutaneous leishmaniasis in a large number of patients. The authors report a success rate of almost 100%; in addition, they claim a low incidence of side effects. Unfortunately, there are several key omissions in this study. First, the outcome of other patients treated with antimoniais and sulfonamides is not reported. Second, the period of follow-up is also not reported, and thus the relapse rate is unknown. Nonetheless, the results are encouraging because cryotherapy may be a safe and cost-effective way to treat what is a relatively common condition in some parts of the world. Certainly, large-scale, controlled studies are indicated based on these preliminary results.

Treatment of Cutaneous Leishmaniasis With Allopurinol and Stibogluconate

Purpose.—The primary treatment for leishmaniasis is pentavalent antimonial compounds, despite their high toxicity and expense. Alternative medications such as amphotericin B and pentamidine have similar problems, and oral agents such as ketoconazole have not proven effective. There is increasing evidence that allopurinol may be a useful treatment for leishmaniasis. However, the biochemical basis of its activity is unknown. This randomized, controlled trial compared allopurinol plus stibogluconate with stibogluconate alone for the treatment of cutaneous leishmaniasis.

Methods.—The study, performed in southern Colombia, included 100 patients with cutaneous leishmaniasis. They were randomized to receive stibogluconate, 20 mg/kg/day, by injection for 15 days, either alone or with allopurinol, 20 mg/kg/day, orally in 4 divided doses for 15 days. The study definition of cure was lesion healing after a 3-month course of treatment and persistence of healing through 1 year of follow-up.

Results.—The cure rate was 71% for patients receiving stibogluconate plus allopurinol vs. 39% for those receiving stibogluconate alone. Although the relapse rate was similar for the 2 groups, the failure rate was 12% for stibogluconate plus allopurinol vs. 43% for stibogluconate alone. Although eosinophilia and rash were more frequent in the combination therapy group, clinically important side effects occurred only in the patients receiving stibogluconate alone.

Conclusion.—As in previous studies, adding allopurinol to pentavalent antimony improves the cure rate of cutaneous leishmaniasis. Generic allopurinol is inexpensive—less than $2 for the total dosage used in this study. Allopurinol warrants further study as an adjunctive therapy for cutaneous and visceral leishmaniasis.

 The World Health Organization has cited leishmaniasis as 1 of the 5 most important diseases worldwide. In some cases, allopurinol alone may be effective; more often, it must be used in conjunction with antimony or another therapeutic agent. Intralesional injection of pentavalent antimony has also been advocated.

Efficacy of Sodium Stibogluconate Alone and in Combination with Allopurinol for Treatment of Mucocutaneous Leishmaniasis

Introduction.—A devastating late consequence of cutaneous leishmaniasis, mucocutaneous leishmaniasis (MCL), can lead to suffocation when the trachea and bronchi are severely involved. Pentavalent antimonial agents are usually effective when the disease is restricted to the nose, but the cure rate is only 7% when the larynx and vocal cords are involved. A randomized, open, controlled clinical trial examined the efficacy, tolerance, and safety of sodium stibogluconate plus allopurinol and sodium stibogluconate alone in patients with moderate or severe MCL.

Methods.—Eligible patients were ages 15-60 and had a documented history of MCL with proven presence of parasites by culture or polymerase chain reaction analysis. Phase 1 of the study admitted patients with moderate and severe disease, but because both treatment regimens yielded high failure rates in the first 22 patients (all with severe disease), it was not acceptable to include more patients with severe disease. Phase 2 included only patients with moderate disease. Patients were admitted to the hospital for administration of therapy and were followed for 12 months.

Results.—Eighty-one patients entered the 2 phases of the study and were randomized to sodium stibogluconate alone or with allopurinol. Treatment groups were comparable in demographic, epidemiological, and clinical characteristics. Lesions of MCL improved in phase 2 patients, but only 2 had clinical cures (both had received sodium stibogluconate alone). Cure rates in the phase 2 patients with moderate disease were 75% for sodium stibogluconate alone and 63.6% for sodium stibogluconate plus allopurinol. The 2 groups had similar rates of clinical adverse events. Two of 22 patients in phase 1 and 9 of 59 in phase 2 withdrew from the study, mainly because of treatment-related toxicity. Severe thrombocytopenia led 8 patients to withdraw, 2 in the single agent group and 6 in the combined therapy group.

Conclusion.—In vitro and tissue culture studies showed a synergistic action of allopurinol and antimonial compounds against Leishmania, but the combination of sodium stibogluconate plus allopurinol had no therapeutic advantage over sodium stibogluconate alone in the treatment of patients with moderate or severe MCL.

 These results differ markedly from those of the study by Martinez etal. reviewed above. It is possible that the failure of allopurinol to enhance the effect of sodium stibogluconate in this study is a reflection of the severity of disease in these patients. The published literature is quite confusing. As suggested by Bryceson, “important limitations in published reports on leishmaniasis therapy have been identified (inadequate designs, unclear diagnostic criteria, unclear definition of evaluated outcomes, different dosages of the drug and duration of treatment, etc.).”