MAJOR ARTICLE

Efcacy of Thermotherapy to Treat Cutaneous Leishmaniasis Caused by Leishmania tropica in Kabul, Afghanistan: A Randomized, Controlled Trial
R. Reithinger,1,4 M. Mohsen,1 M. Wahid,2 M. Bismullah,2 R. J. Quinnell,3 C. R. Davies,4 J. Kolaczinski,1,4 and J. R. David5
1

Malaria and Leishmaniasis Control Program, HealthNet International, University Town, Peshawar, Pakistan; 2Afghan Ministry of Health, Kabul, Afghanistan; 3School of Biology, University of Leeds, Leeds, and 4Department of Infectious and Tropical Diseases, London School of Hygiene & Tropical Medicine, London, United Kingdom; and 5Department of Immunology and Infectious Diseases, Harvard School of Public Health, Boston, Massachusetts

(See the editorial commentary by Wortmann on pages 11568)

Background. Pentavalent antimony is the agent recommended for treatment of cutaneous leishmaniasis (CL). Its use is problematic, because it is expensive and because of the potential for drug-associated adverse effects during a lengthy and painful treatment course. Methods. We tested the efcacy of thermotherapy for the treatment of CL due to Leishmania tropica in a randomized, controlled trial in Kabul, Afghanistan. We enrolled 401 patients with a single CL lesion and administered thermotherapy using radio-frequency waves (1 treatment of 1 consecutive application at 50 C for 30 s) or sodium stibogluconate (SSG), administered either intralesionally (a total of 5 injections of 25 mL every 57 days, depending on lesion size) or intramuscularly (20 mg/kg daily for 21 days). Results. Cure, dened as complete reepithelialization at 100 days after treatment initiation, was observed in 75 (69.4%) of 108 patients who received thermotherapy, 70 (75.3%) of 93 patients who received intralesional SSG, and 26 (44.8%) of 58 patients who received intramuscular SSG. The OR for cure with thermotherapy was 2.80 (95% condence interval [CI], 1.455.41), compared with intramuscular SSG treatment (P p .002 ). No statistically signicant difference was observed in the odds of cure in comparison of intralesional SSG and thermotherapy treatments. The OR for cure with intralesional SSG treatment was 3.75 (95% CI, 1.867.54), compared with intramuscular SSG treatment (P ! .001 ). The time to cure was signicantly shorter in the thermotherapy group (median, 53 days) than in the intralesional SSG or intramuscularly SSG group (median, 75 days and 1100 days, respectively; P p .003). Conclusions. Thermotherapy is an effective, comparatively well-tolerated, and rapid treatment for CL, and it should be considered as an alternative to antimony treatment. World-wide, the largest focus of cutaneous leishmaniasis (CL) is in Kabul, Afghanistan, where Leishmania tropica is anthroponotically transmitted by the sand y Phlebotomus sergenti and where the estimated annual case load has been 67,500200,000 patients during the past decade [13]. Because of limited resources available during and after the Afghan civil war, the treatment of patients with CL by the Afghan Ministry of Health and by nongovernmental organizations is the only strategy to control the epidemic. Pentavalent antimonialsnamely, sodium stibogluconate (SSG) and meglumine antimoniateare the mainstay of anti-leishmanial therapy [46]. A leishmaniasis vaccine does not exist [7]. In countries where leishmaniasis is endemic, antimonials are typically administered intramuscularly (at a dosage of 20 mg/kg/ day for 2028 days) or intralesionally (at a dosage dependent on the lesion size) [4, 5]. However, antimonials can have serious (usually reversible) adverse effects (e.g., pancreatitis, hepatotoxicity, and cardiotoxicity) when given intramuscularly [4,5], and they are expensive [8]. Moreover, the invasiveness of the stan-

Received 11 November 2004; accepted 7 December 2004; electronically published 16 March 2005. Presented in part: Annual Meeting of the American Society of Tropical Medicine and Hygiene (Philadelphia), 711 December 2003. Reprints or correspondence: Dr. R. Reithinger, 807 S. Overlook Dr., Alexandria VA 22305 ([email protected]). Clinical Infectious Diseases 2005; 40:114855 2005 by the Infectious Diseases Society of America. All rights reserved. 1058-4838/2005/4008-0012$15.00

1148 CID 2005:40 (15 April) Reithinger et al.

dard procedurea lengthy course of painful inoculations often leads patients to default on their full treatment course. Low compliance appears to be the principal reason behind the emergence of drug-resistant parasite strains, especially in areas where there is anthroponotic leishmaniasis; for example, in India, Sudan and Nepal [9]. Hence, research is focusing on the development of alternative treatments that use different dosage schedules, drugs, or methods of treatment [6]. Surprisingly, despite its regional importance, there is a dearth of clinical data on CL caused by L. tropica, and there are only limited data on natural or treatment-induced cure rates [10]. We evaluated the efcacy of thermotherapy as treatment for CL due to L. tropica in Kabul and compared it with standard intramuscular and intralesional SSG treatment in a controlled, randomized trial. METHODS Study location and participants. The study was carried out at the HealthNet International Khair Khana clinic in Kabul, Afghanistan. This clinic has been operational since 1995 and is the main leishmaniasis treatment center in Kabul [2, 3], with 4751 new and 25,783 follow-up patients treated in 2003. Eligible patients were those who attended the clinic for leishmaniasis treatment and who had only 1 suspected CL lesion. Inclusion criteria were age of 15 years; the presence of a single, parasitologically conrmed CL lesion; and no prior history of disease and/or antimonial treatment. Exclusion criteria were the presence of a CL lesion located on or immediately adjacent to the nose, lips, or eyes; pregnancy; breast-feeding; major surgery in the previous 3 months; presence of any uncontrolled medical condition; and anticipated unavailability for follow-up. Most patients were current Kabul residents. Study design and procedures. The study was a randomized, controlled trial. There was no placebo group, because this would have been unethical due to the severity of CL that L. tropica causes and the social stigma associated with the disease [2]. The London School of Hygiene and Tropical Medicine Ethics Committee and the Afghan Ministry of Health approved the study protocol and the consent form. According to HealthNet International policy, all medical services provided during the study were free of charge. To detect a 20% difference in the cure rate between the SSG and thermotherapy groups, assuming an 80% cure rate in the SSG groups [46], with a 90% power and a 5% 2-sided type I error, 98 subjects were needed in each group. To compensate for anticipated loss to follow-up, 40% more patients were enrolled in each group. Eligible patients coming to the clinic for treatment were briefed about the study, its aims, and the protocol. Patients were enrolled in the study after written consent had been given. Patients then proceeded to pick 1 of 3 identical cardboard pieces out of a hat (the cardboard had been labeled

with different treatment codes on one of its sides, the codes being nonvisible to the patient). After patients were randomly assigned to receive a treatment, the cardboard piece picked was returned to the hat. The assigned treatments were as follows: (1) intralesional administration of generic SSG (Albert David Ltd., Calcutta, India), 5 injections of 25 mL (depending on lesion size) every 57 days for a total of up to 29 days; (2) daily intramuscular administration of SSG 20 mg/kg (up to a maximum daily dose of 850 mg) for 21 days; and (3) a single thermotherapy treatment ( 1 consecutive application of 50 C for 30 s, depending on lesion size). Both SSG regimens are standard World Health Organization recommended treatment for CL in Afghanistan [11]. For intralesional treatment, SSG was inltrated around the lesion until complete blanching of the lesion and its margin was obtained [11]. For thermotherapy, the lesion and a 1520-mm border of healthy skin around the lesion were cleaned with stabilized 0.1% chlorine dioxide solution, anesthetized with 1% lidocaine HCl, and moistened with sterile saline solution; then heat was applied locally with a portable, battery-operated, localized current eld radio-frequency generator (ThermoMed 1.8; Thermosurgery Technologies), according to the manufacturers instructions. The generator has received 501K clearance by the US Food and Drug Administration for CL treatment. It produces a 6.78-mHz frequency, applied with a handset that includes an applicator gauge with 2 electrodes that are placed onto the diseased skin. The area between the electrodes covers 4973 mm2, depending on the applicator gauge size used; therefore, several thermotherapy applications may be required to cover a lesion. Once treatment begins, the temperature is measured by a thermistor embedded in one of the electrodes, ensuring that the applied temperature remains constant. The applied radio frequencies excite the tissue molecules, producing heat that evenly penetrates the upper dermis, exposing Leishmania amastigotes to high temperature without injuring the healthy underlying tissue. After all treatments, a chlorine dioxide gel was applied to lesions, and lesions were covered with gauze to prevent secondary infections. Before the rst treatment, all patients received a full physical examination. The location and duration (prior to treatment) of the lesion were recorded; its diameter was measured with a caliper. The status of each lesion was evaluated during 4 patient follow-up visits; the trial end point was 100 days after start of therapy. Lesions with a secondary bacterial infection before, during, or after treatment were treated with topical antibiotics. If systemic treatment was required, patients received treatment with an antibiotic that has no activity against Leishmania (e.g., erythromycin). The occurrence of adverse effects was evaluated blindly by means of patient interviews and physical examinations during follow-up visits. Treatment efcacy was measured by the percentage of pa-

Thermotherapy for Cutaneous Leishmaniasis CID 2005:40 (15 April) 1149

Figure 1. Flow chart summarizing the enrollment and randomization of patients in a trial of thermotherapy for cutaneous leishmaniasis and listing reasons for exclusion or withdrawal from the study. All values are no. of patients.

tients cured at 100 days after treatment initiation and by time to cure. Cure was dened as the complete reepithelialization of the CL lesion, with no evidence of papules, inammation, or induration. Patients that did not experience cure after 100 days were offered intralesional or intramuscular SSG, as appropriate. Parasitological studies. Parasitological conrmation of CL was by microscopic examination in Kabul and parasite identication by PCR at Leeds University (Leeds, United Kingdom). Microscopic examination was performed blindly. Scrapings from the lesion edge were smeared onto a slide, and the slide was dried, xed with methanol, Giemsa-stained, and examined under the microscope at 100 magnication for presence of Leishmania amastigotes. For PCR, lesion scrapings were preserved in ethanol at 20 C prior to DNA extraction using a QIAamp DNA mini kit (Qiagen). Samples were amplied in a nested PCR with Leishmania-specic kinetoplast minicircle primers, under conditions published elsewhere [12]. This protocol differentiates between the major Leishmania species in Central Asia, namely L. tropica, Leishmania major, and Leishmania infantum. Each PCR included appropriate negative and positive controls. To evaluate sample degradation or PCR inhibition, sample DNA was also amplied for a 740-bp fragment of the human TNFB gene [13].

Statistical analysis. All patient data were entered into Excel software (Microsoft). A x2 test was used to test for signicance (P ! .05) between proportions (e.g., sex and loss to follow-up). All other analyses were done with Stata software, version 6.0 (Stata). A Kruskal-Wallis test was used to test for signicance between pretrial characteristics (e.g., age, body weight, lesion size, and lesion duration) of treatment groups. The effects of the treatments on the proportion of patients cured by 100 days were tested by logistic regression. The analyses incorporated the effect of explanatory variables (discussed above) and treatment type. The signicance of each variable was tested by backwards deletion; that is, by observation of whether these variables explained a signicant (P ! .05) proportion of the deviance remaining after removal from the model. Variables were removed from the models in order of least signicance until only signicant variables (those with P ! .05 ) were retained in the minimum adequate model. The Kaplan-Meier method was used for the time-to-healing analysis; to compare the healing curves for the 3 treatment types, the log-rank test was used. RESULTS A total of 431 patients were enrolled in the study between January and September 2003. After random treatment alloca-

1150 CID 2005:40 (15 April) Reithinger et al.

Table 1.

Demographic and clinical characteristics of patients before treatment for cutaneous leishmaniasis.
Intralesional SSG group 146 67/79 13 (8.2522) 36 (2250) 12.75 (720) 6 (37) Intramuscular SSG group 117 63/54 13 (1022) 38.5 (2651.25) 13.75 (822.5) 5.5 (37) Thermotherapy group 138 70/68 14 (1020) 40 (2451) 10.25 (720) 6 (3.758) .75 .70 .11 .19

Variable No. of patients Sex, no. male/female Age, median years (IQR) Body weight, median kg (IQR) Lesion diameter, median mm (IQR) Lesion duration, median months (IQR)
NOTE
a

All patients 401 200/201 13 (1020) 39 (2451) 12 (720) 6 (37)

IQR, interquartile range; SSG, sodium stibogluconate.

A Kruskal-Wallis rank test was used to compare difference in pretreatment patient characteristics between treatment groups.

tion, 30 patients decided to withdraw from the study (gure 1); 146 patients received intralesional SSG, 117 received intramuscular SSG, and 138 received thermotherapy treatment. Baseline patient characteristics. All patients had lesions parasitologically conrmed by microscopy. For a subset of 39 patients, lesion scrapings were obtained for PCR-based parasite identication. All samples yielded amplication products for the human TNFB gene fragment. Of these 39 samples, 27 (69%) were PCR-positive for Leishmania DNA; for all 27, L. tropica was identied. Demographic and clinical characteristics of patients are presented in table 1. No statistically signicant differences were observed between treatment groups with respect to sex, age, body weight, lesion size, or lesion duration. The lesions were primarily located on the face (43.4% of patients), as well as on the hands (38.2%), legs (15.9%), and arms (2.4%). The median times of follow-up visits for patients who completed the trial were day 13, day 21, day 49, and day 85, for the rst, second, third, and fourth visit, respectively. Efcacy. A total of 259 patients (63.8%) completed treatment and completed the 4 visits and 100 days of follow-up. Of these 259 patients, 108 were treated with thermotherapy, 93 were treated with intralesional SSG, and 58 were treated with intramuscular SSG (gure 1). In the intramuscular SSG group, 27 patients (47%) had their SSG dose limited by the protocol ceiling of 850 mg/day. Thermotherapy-treated patients were shown to be least likely to be lost to follow-up either during or after the end of treatment (table 2). Complete cure by 100 days was observed in 69.4% of patients treated with thermotherapy, 75.3% of patients treated with intralesional SSG, and 44.8% of patients treated with intramuscular SSG (table 3). None of the patients with complete healing had relapse during the 100 days of the study. No statistically signicant association was shown between age, sex, body weight, lesion size, lesion location or lesion duration and trial outcome. The OR for cure with thermotherapy was 2.80 (95% CI, 1.455.41), compared with intramuscular SSG treatment (P p .002). No statistically signicant difference was observed between the odds of cure with intralesional SSG or thermo-

therapy treatment (OR, 1.34; 95% CI, 0.722.50; P p .359). The OR for cure with intralesional SSG treatment was 3.75 (95% CI, 1.867.54), compared with intramuscular SSG treatment (P ! .001). An intention-to-treat analysis of the data (i.e., including the patients lost to follow-up, who were considered to have had treatment failure) yielded similar results for the comparison of the odds of cure for the different treatments (table 3). According to the Kaplan-Meier survival analysis (which analyzes all available data, including those for patients who dropped out of the study during treatment), the time to cure was signicantly shorter for patients treated with thermotherapy (median, 53 days) than for those who were treated with intralesional or intramuscular SSG (medians, 75 days and 1100 days, respectively; P p .003, by the log-rank test) (gure 2). Secondary infections were noted in 8 patients treated with thermotherapy (2 before and 6 after treatment), in 5 patients treated with intralesionally SSG (2 before and 3 after start of treatment), and in 2 patients treated with intramuscular SSG

Table 2.

Rates of loss to follow-up of patients in the trial.

Period, treatment group During treatment Intramuscular SSG Intralesional SSG After the end of treatment Intramuscular SSG Intralesional SSG Throughout the study Intramuscular SSG Intralesional SSG

Thermotherapy group 0.93 (0.441.97) 0.70 (0.351.39) 3.48 (1.946.24)b 1.97 (1.133.47)
2

Intralesional SSG group 1.89 (1.063.38)a 1.34 (0.593.05) 1.78 (1.053.03)a

NOTE. ORs were estimated using the x test using data shown in gure 1. Excluded from the analyses are patients who withdrew from the study because of the adverse effects of treatment. SSG, sodium stibogluconate.
a b

Not signicant; P ! .05. Not signicant; P ! .001.

Thermotherapy for Cutaneous Leishmaniasis CID 2005:40 (15 April) 1151

Table 3.

Cure rates and odds of cure under different assumptions for the 3 treatment groups at the end of the trial.
Proportion (%) of patients with cure at 100 days

OR (95% CI) TH group IL SSG vs. TH 0.77 (0.471.27) 2.15 (1.283.63)

Analytical assumptiona Per protocol ITT ITT var

IL SSG group 70/93 (75.3) 70/146 (47.9) 105/146 (71.9)

IM SSG group 26/58 (44.8) 26/117 (22.2) 70/117 (59.8)

IL SSG vs. IM SSG 3.22 (1.815.77)b 1.72 (0.992.98)

IM SSG vs. TH 0.24 (0.130.43) 1.25 (0.742.13)

75/108 (69.4) 75/138 (54.3) 75/138 (54.3)

NOTE Differences between per-protocol cure rates were analyzed using logistic regression (see Methods). IL, intralesional; IM, intramuscular; ITT, intention to treat; SSG, sodium stibogluconate; TH, thermotherapy. Per protocol: analysis excluding patients that were lost to follow-up. ITT: analysis including patients lost to follow-up throughout the study, who were considered to have experienced treatment failure. ITT var: analysis including patients lost to follow-up during treatment, who were considered to have experienced cure, and including patients lost to follow-up after treatment completion, who were considered to have experienced treatment failure. b P ! .001; pairwise comparison of ITT cure rates was done with a x2 test of proportions, with ORs given in the table. c P ! .01. d P ! .05.
a

(1 before and 1 after start of treatment), but the differences were not statistically signicant. All secondary infections resolved with antibiotic treatment. Excluded from the study after treatment initiation were 2 patients treated with intralesional SSG, because of bradycardia and an undened local reaction to the treatment; and 3 patients treated with intramuscular SSG, because of bradycardia, tachycardia, and palpitation. In the thermotherapy-treated group, the original CL ulcer often increased in size immediately after and up to 2 weeks after treat-

ment, with patients experiencing supercial second-degree burns where the electrodes were applied; thereafter, the lesion closed rapidly (gure 2). This observation may explain the low rate of follow-up among thermotherapy-treated patients in the rst 2 weeks after treatment; 25 (83%) of the 30 patients lost to follow-up in the thermotherapy group were lost in the rst 2 weeks. Patients may have become disheartened and unwilling to seek further medical advice when lesions became bigger. Though evaluation of scarring was not performed blindly, nei-

Figure 2. Survival analysis of time to healing of cutaneous leishmaniasis lesions, with data on the no. of patients enrolled in the trial at baseline and 4 other time points. IL, intralesional; IM, intramuscular; SSG, sodium stibogluconate; TH, thermotherapy.

1152 CID 2005:40 (15 April) Reithinger et al.

ther the examining clinician nor patients noted a visible difference in the scarring between patient groups after successful treatment. DISCUSSION We demonstrate that a single treatment with accurately measured localized heat is as effective as the administration of intralesional SSG and more effective than the administration of intramuscular SSG for the treatment of CL due to L. tropica. The time to cure was shown to be shorter with thermotherapy than with SSG regimens in a Kaplan-Meier analysis of data for all study patients. We observed no signicant effect of patient characteristics on the cure rate. Because of the postconict situation in Afghanistan, enrollment of patients was continuous throughout the study period. The reason the intramuscular SSG treatment group was smaller was that 27 patients who had been randomized to receive intramuscular SSG treatment refused to give consent (gure 1). Due to the social stigma associated with the disease, Kabul residents are very knowledgeable about leishmaniasis. They know that intramuscular SSG injections are painful and are usually given to patients with multiple lesions or lesions on sites where intralesional SSG administration will be difcult. These patients requested intralesional SSG or thermotherapy treatment and were excluded from the trial and given the treatment they requested. Further proof of patients low acceptance of intramuscular SSG treatment was that the number of patients lost to follow-up during treatment was signicantly greater in the intramuscular SSG group than in the intralesional SSG group (P ! .05), but it was not signicantly different after the end of treatment (P p .58 ) (gure 1 and table 2). Though the number of patients lost to follow-up after the end of treatment was not signicantly different between treatment groups (P p .54) (gure 1), one caveat of our study is that we cannot exclude the possibility that patients dropped out during treatment because lesions were healing and, knowing that they had to face the remaining SSG injections, they did not want to attend further treatment visits. If one assumes that patients lost to follow-up during treatment experienced cure, the cure rate among patients treated with intralesional SSG would be signicantly larger than that for patients treated with intramuscular SSG or with thermotherapy; the cure rates in the latter 2 groups would not differ signicantly (table 3). Laboratory studies showed that Leishmania parasites do not readily multiply in macrophages at temperatures 139 C in vitro [14, 15]. These observations led to studies investigating the efcacy of thermotherapy treatment of CL with hot-water baths [16], infrared light [17], direct-current electrical stimulation [18], ultrasound [19], and laser light [2023]. Specically, 3 studies suggested that thermotherapy with radio-frequency

waves could be effective for CL treatment. In a placebo-controlled trial, thermotherapy (3 treatments of 50 C for 30 s at 7-day intervals) was as effective as antimony therapy (meglumine antimoniate, 850 mg/day for 15 days) in treating L. braziliensis and L. mexicana infection [24]. Cure rates were identical73% (16 of 22 patients), in each treatment group, compared with 27% (6 of 22 patients) in the placebo control group13 weeks after treatment initiation [24]. In a second, uncontrolled study of thermotherapy (a single treatment of 50 C for 30 s) in L. mexicanainfected patients, cure rates of 95% (116 of 122 patients) and 90% (172 of 191 patients) were observed 4 and 8 weeks after treatment, respectively [25]. In a case report, a Sudanese patient with multiple L. tropica lesions was cured 6 months after receiving thermotherapy (a single treatment of 50 C for 30 s) [26]. However, it is difcult to draw conclusions from these studies. First, they are case studies or include small numbers of patients that preclude in-depth statistical analyses [16, 17, 19, 24, 26]. Second, some followed an undened study protocol (e.g., they lacked either placebo or control treatment groups) [19, 25, 26]; one study had a short follow-up period that could have excluded relapses [25] and another had a long follow-up period that could have included cases of self-cure [26]. Third, one study included patient groups that were treated with suboptimal durations of antimony treatment (i.e., !2028 treatment days) [24]. Fourth, one study included patients infected with different or unknown Leishmania species [24]. There are many reports of successful administration of intralesional antimony to cure Old World CL [4, 5, 27], but only one study in Saudi Arabia compared intramuscular versus intralesional administration of antimony [28]; cure rates were 68% and 73%, respectively, at 30 days after treatment (the difference was not statistically signicant). Therefore, to our knowledge, our study is the rst that shows that the intralesional route of SSG administration is more effective than the intramuscular route for treatment of Old World CL. This has practical relevance in terms of drug management because, on average, 10 times less drug is used when treating patients intralesionally instead of intramuscularly (R.R., unpublished data). Also, as observed here, patient compliance with intralesional treatment is better, because fewer clinic visits and injections are required. Surprisingly, comparable reported data on cure rates for the treatment of conrmed L. tropica infection with antimony are scarce: there is a single study reporting a 76% cure rate 10 weeks after intralesional meglumine antimoniate administration [29]. Localized heat could be an alternative to antimony for the treatment of CL and, in particular, would be very cost-effective in those areas of endemicity where the number of cases of Leishmania infection is high and focal (e.g., areas with anthro-

Thermotherapy for Cutaneous Leishmaniasis CID 2005:40 (15 April) 1153

ponotic foci of CL). Reliable data on the cost of leishmaniasis treatment are difcult to obtain and depend on several factors (e.g., whether treatment is given on an inpatient or an outpatient basis and whether cheaper, generic SSG is used); the cost per patient treated range from US$20 in Afghanistan (R.R. and P. G. Coleman, unpublished data) to US$280 in Guatemala [30] to 1US$5500 in the United States (N. Aronson, personal communication). The retail price of the thermotherapy device used in the present study is US$23,450. There are 2 main advantages to the tested thermotherapy protocol, compared with antimony treatment: (1) patient compliance rates are improved because of the lack of potentially serious adverse effects of treatment, because treatment is administered nonparenterally, and because the treatment schedule is shorter (i.e., 1 day, compared with 521 injection-days for antimony); and (2) the shorter administration schedule also increases the patient turnover rate, a prerequisite for controlling the patient case load and, hence, disease transmission. The tested thermotherapy method uses a handheld device and limited additional medical equipment, making it suitable for eld conditions in areas with rudimentary medical infrastructure; the device, however, needs a power source to recharge the battery. Though we tested the efcacy of thermotherapy on patients with single CL lesions only, patients with multiple lesions could be treated in the same way. Patients with CL lesions adjacent to the eyes and lips will still have to be treated intramuscularly with SSG. Also, L. tropica is one of the more temperature-resistant Leishmania species [14, 15]. One would expect the thermotherapy method we tested to be more effective against less temperature-resistant Leishmania species; this awaits conrmation, for example for L. major. Ultimately, the decision to use thermotherapy will depend on clinical factors (e.g., the location, size, and number of lesions, and the patients responsiveness to antimony therapy) and patient management factors (e.g., patients availability for follow-up and the total treatment time per patient). In conclusion, thermotherapy with the tested device proved to be effective, safe, and relatively noninvasive for treating patients with CL in the current postconict context of Kabul, Afghanistan.

Potential conicts of interest. Since March 2004, R. Reithinger has been a part-time employee of Thermosurgery Technologies. All other authors: no conicts.

References
1. World Health Organization. Cutaneous leishmaniasis, Afghanistan. Weekly Epidemiological Record 2002; 77:246. 2. Reithinger R, Mohsen M, Aadil K, Sidiqi M, Erasmus P, Coleman PG. The burden of anthroponotic cutaneous leishmaniasis in Kabul, Afghanistan. Emerg Infect Dis 2003; 9:7279. 3. Reyburn H, Rowland M, Mohsen M, Khan B, Davies CR. The prolonged epidemic of anthroponotic cutaneous leishmaniasis in Kabul, Afghanistan: bringing down the neighbourhood. Trans R Soc Trop Med Hyg 2003; 97:1706. 4. Berman JD. Human leishmaniasis: clinical, diagnostic and chemotherapeutic developments in the last ten years. Clin Infect Dis 1997; 24:684703. 5. Croft SL, Yardley V. Chemotherapy of leishmaniasis. Curr Pharm Design 2002; 8:273301. 6. Croft SL, Coombs GH. Leishmaniasiscurrent chemotherapy and recent advances in the search for novel drugs. Trends Parasitol 2003; 19: 5028. 7. Handmann E. Leishmaniasis: current status of vaccine development. Clin Microbiol Rev 2001; 14:22943. 8. Boelaert M, Le Ray D, Van Der Stuyft P. How better drugs could change kala-azar control: lessons from a cost-effectiveness analysis. Trop Med Int Health 2002; 7:9559. 9. Bryceson A. A policy for leishmaniasis with respect to the prevention and control of drug resistance. Trop Med Int Health 2001; 6:92834. 10. Dowlati Y. Cutaneous leishmaniasis: clinical aspect. Clin Dermatol 1996; 14:42531. 11. Reyburn, H. A Guide to the treatment of cutaneous leishmaniasis. Kabul, Afghanistan: World Health Organization, 2000. 12. Noyes H, Reyburn H, Bailey JW, Smith D. A nested-PCRbased schizodeme method for identifying Leishmania kinetoplast minicircle classes directly from clinical samples and its application to the study of the epidemiology of L. tropica in Pakistan. J Clin Microbiol 1998; 36: 287781. 13. Messer G, Spengler U, Jung MC, et al. Polymorphic structure of the tumor necrosis factor (TNF) locusan NcoI polymorphism in the rst intron of the human TNF-beta gene correlates with a variant amino acid in position 26 and a reduced level of TNF-beta production. J Exp Med 1991; 173:20919. 14. Berman JD, Neva FA. Effect of temperature on multiplication of Leishmania amastigotes within monocyte-derived macrophages in vitro. Am J Trop Med Hyg 1981; 30:31821. 15. Sacks DL, Barral A, Neva F. Thermosensitivity patterns of Old vs. New World cutaneous strains of Leishmania growing within mouse peritoneal macrophages in vitro. Am J Trop Med Hyg 1983; 32:3004. 16. Neva FA, Petersen EA, Corsey R, Bogaert H, Martinez D. Observations on local heat treatment for cutaneous leishmaniasis. Am J Trop Med Hyg 1984; 33:8004. 17. Junaid AJN. Treatment of cutaneous leishmaniasis with infrared heat. Int J Dermatol 1986; 25:4702. 18. Sharquie KE, al-Hamamy H, el-Yassin D. Treatment of cutaneous leishmaniasis by direct current electrotherapy: the Baghdadin device. J Dermatol 1998; 25:2347. 19. Aram H, Leibovici V. Ultrasound-induced hyperthermia in the treatment of cutaneous leishmaniasis. Cutis 1987; 40:3503. 20. Rodriguez ME, Inguanzo P, Ramos A, Perez J. Treatment of cutaneous leishmaniasis with CO2 laser radiation. Rev Cubana Med Trop 1990; 42:197202. 21. Babajev KB, Babajev OG, Korepanov VI. Treatment of cutaneous leishmaniasis using a carbon dioxide laser. Bull World Health Organ 1991; 69:1036.

Acknowledgments
We are grateful to HealthNet International (HNI) Khair Khana Clinic staff for logistical support in the trial. We thank Harry Noyes, for the reference strains used in the PCR assays, and Anthony Bryceson, for making valuable comments on the manuscript. Financial support. The HNI Malaria and Leishmaniasis Control Program is funded by the European Union. This study was supported by the United Nation Childrens Fund / United Nation Development Program / World Bank / World Health Organization Special Program for Research and Training in Tropical Diseases, the Afghan Research Evaluation Unit, and the Leveen Family Fund. We are grateful to Thermosurgery Technologies for lending us the instruments for the study.

1154 CID 2005:40 (15 April) Reithinger et al.

22. Meawad OB. Selective heat therapy in cutaneous leishmaniasis: a preliminary experience using the 585 nm pulsed dye laser. J Europ Acad Dermatol Venereol 1997; 8:2414. 23. Asilian A, Sharif A, Faghihi G, Enshaeieh Sh, Shariati F, Siadat AH. Evaluation of CO laser efcacy in the treatment of cutaneous leishmaniasis. Int J Dermatol 2004; 43:7368. 24. Navin TR, Arana BA, Arana FE, de Merida AM, Castillo LA, Pozuelos JL. A placebo controlled clinical trial of meglumine antimoniate (Glucantime ) vs. localized controlled heat in the treatment of cutaneous leishmaniasis in Guatemala. Am J Trop Med Hyg 1990; 42:4350. 25. Velasco-Castrejon O, Walton BC, Rivas-Sanchez B, et al. Treatment of cutaneous leishmaniasis with localized eld radio frequency in Tabasco, Mexico. Am J Trop Med Hyg 1997; 57:30912. 26. Levine N. Cutaneous leishmaniasis treated with controlled localized heating. Arch Dermatol 1992; 128:75961.

27. Tallab TM, Bahamdam KA, Mirdad S, et al. Cutaneous leishmaniasis: schedules for intralesional treatment with sodium stibogluconate. Int J Dermatol 1996; 35:5947. 28. Alkhawajah AM, Larbi E, al-Gindan Y, Abahussein A, Jain S. Treatment of cutaneous leishmaniasis with antimony: intramuscular versus intralesional administration. Ann Trop Med Parasitol 1997; 91:899905. 29. Harms G, Chehade AK, Douba M, et al. A randomized trial comparing a pentavalent antimonial drug and recombinant interferon-gamma in the local treatment of cutaneous leishmaniasis. Trans R Soc Trop Med Hyg 1991; 85:2146. 30. Arana BA, Mendoza CE, Rizzo NR, Kroeger A. Randomized, controlled, double-blind trial of topical treatment of cutaneous leishmaniasis with paromomycin plus methylbenzethonium chloride ointment in Guatemala. Am J Trop Med Hyg 2001; 65:46670.

Thermotherapy for Cutaneous Leishmaniasis CID 2005:40 (15 April) 1155