Amphotericin B Treatment for Indian Visceral Leishmaniasis ...

MAJOR ARTICLE

Amphotericin B Treatment for Indian Visceral Leishmaniasis: Response to 15 Daily versus Alternate-Day Infusions Shyam Sundar,1 J. Chakravarty,1 V. K. Rai,1 N. Agrawal,2 S. P. Singh,1 V. Chauhan,1 and Henry W. Murray3 1

Kala-Azar Medical Research Center, Department of Medicine, Banaras Hindu University, Institute of Medical Sciences, Varanasi, and 2MLN Medical College, Allahabad, India; and 3Department of Medicine, Weill Medical College of Cornell University, New York

Amphotericin B deoxycholate has been used in India for treatment of visceral leishmaniasis (kala-azar) for 11 decade. Its rediscovery as effective treatment for Leishmania donovani infection was spawned by the development of large-scale resistance to conventional pentavalent antimony therapy in Bihar State [1]. (Approximately 90% of India’s—and ∼45% of the world’s—annual new cases of visceral leishmaniasis occur in this region [2].) In Indian kala-azar, 15 infusions of 1 mg/kg of am-

Received 2 February 2007; accepted 7 May 2007; electronically published 23 July 2007. Reprints or correspondence: Dr. Shyam Sundar at 6, SK Gupta Nagar, Lanka, Varanasi-221 005, India ([email protected]). Clinical Infectious Diseases 2007; 45:556–61 2007 by the Infectious Diseases Society of America. All rights reserved. 1058-4838/2007/4505-0005$15.00 DOI: 10.1086/520665

556 • CID 2007:45 (1 September) • Sundar et al.

photericin B (total dose, 15 mg/kg) are usually given on alternate days because of concerns about tolerability [3]; we and others have also successfully administered a dose of 0.75 mg/kg in this 30-day regimen (total dose, 11.25 mg/kg) (S.S., unpublished data) [4, 5]. Other investigators recommend once-daily infusions of 1 mg/ kg for 20 days (total dose, 20 mg/kg) [6]. In either instance, efficacy is uniformly high, with long-term cure rates of ⭓96% [3]. Nevertheless, the prolonged treatment duration (20–30 days) and dose-related adverse reactions remain recognized drawbacks of amphotericin B treatment. To address the unsettled therapeutic questions of the administration schedule (alternate-day vs. daily administration) and dose (1 vs. 0.75 mg/kg) and to determine whether the duration of amphotericin B treatment in Bihar can be abbreviated (to 15 days), we performed a randomized, controlled trial that examined these 3 issues.

Downloaded from http://cid.oxfordjournals.org/ by guest on November 14, 2016

Background. For patients with Indian visceral leishmaniasis, amphotericin B deoxycholate is usually given as 15 alternate-day infusions of 1 mg/kg over 30 days (total dose, 15 mg/kg); daily treatment with 1 mg/kg for 20 days (total dose, 20 mg/kg) is also used. This study was done to address the unsettled therapeutic questions of administration schedule (alternate-day vs. daily administration) and dose (1 vs. 0.75 mg/kg) and to determine whether the duration of amphotericin B treatment in Bihar, India, can be shortened to 15 days. Methods. To compare alternate-day versus daily administration and 1-mg/kg versus 0.75-mg/kg doses and to determine whether the duration of treatment could be abbreviated, Indian subjects randomly received 15 infusions of 1 mg/kg (group A; 245 patients) or 0.75 mg/kg (group B; 244 patients) on alternate days or 1 mg/kg (group C; 500 patients) or 0.75 mg/kg (group D; 496 patients) daily. Noninferiority testing compared 6-month cure rates using a 5% margin. Results. Overall, 1439 of the 1485 subjects completed treatment and responded. Treatment interruptions (nephrotoxicity) but not infusion-associated reactions or study removals were more common with daily administration. Final cure rates at 6 months were similar: group A, 234 patients (96%; 95% confidence interval [CI], 92%–98%); group B, 225 patients (92%; 95% CI, 88%–95%); group C, 483 patients (97%; 95% CI, 95%–98%); and group D, 476 patients (96%; 95% CI, 94%–97%; P 1 .05). Conclusions. Provided that the serum creatinine level is repeated once, daily treatment with amphotericin B, 0.75 mg/kg for 15 days (total dose, 11.25 mg/kg), is efficient and effective for visceral leishmaniasis in India. Trial registration. ClinicalTrials.gov identifier: NCT00310505.

PATIENTS AND METHODS

Figure 1. Trial profile. qd, Daily; qod, every other day. *Two subjects randomized to group C or D for every 1 subject randomized to group A or B.

Amphotericin B in Indian Kala-Azar • CID 2007:45 (1 September) • 557


Eligibility and entry and exclusion criteria. This open-label study was performed from July 2003 through June 2006 at the Kala-Azar Medical Research Center in Muzaffarpur, Bihar, and was approved by the center’s ethics committee. Patients aged 2–65 years were eligible if they had symptoms and signs of kalaazar (i.e., fever, weight loss, and splenomegaly) and if the presence of parasites was demonstrated by microscopic examination of a splenic aspirate smear [2, 7]. Pregnant or breast-feeding women, individuals who were seropositive for HIV, and individuals with a serious concurrent infection (e.g., tuberculosis or bacterial pneumonia) were excluded from the study. Exclusion criteria also included the following findings: a leukocyte count of !1000 cells/mm3, a hemoglobin concentration of !3.5 g/dL, or a platelet count of !40,000 platelets/mm3; hepatic transaminase or total bilirubin levels of 13 times the normal level; serum creatinine level, 12.0 mg/dL; and prothrombin time, 15 s greater than the control level. Number of patients. For the purposes of this study, we considered 15 alternate-day infusions of 1 or 0.75 mg/kg (groups A and B) as conventional therapy [2, 3, 5, 8]. To compare responses to these same doses given once daily (groups C and D), we used a 1:2 ratio for random assignment to treatment arms and aimed to enroll 250 subjects each in groups A and B and 500 subjects each in groups C and D (see Statistical analysis below). A total of 1578 eligible patients were screened; 9 elected not to participate, 84 were excluded from the study on the basis of the aforementioned criteria, and 1485 were enrolled and randomized (figure 1).

Trial procedures and treatments. Subjects underwent baseline testing (determination of standard biochemistry and hematologic profiles, urinalysis, chest film examination, electrocardiography, anti-HIV antibody testing by ELISA, and malaria smear examination) [7] and provided written informed consent (from a parent or guardian for minors). An independent statistician prepared sealed randomization envelopes using a computer-based random number generator. Subjects were assigned to receive 15 intravenous infusions of amphotericin B (Fungizone; Sarabhai Chemicals; stored at 2C–8C) given over 4 h, as follows: (A) on alternate days at a dose of 1 mg/kg (group A) or 0.75 mg/kg (group B) or (B) once daily at a dose of 1 mg/kg (group C) or 0.75 mg/kg (group D). Treatment commenced within 3 days of diagnosis of kala-azar by examination of a splenic aspirate specimen, and all patients received an initial 5-mg test dose on treatment day 1. An oral antipyretic (paracetamol, 500 mg) and an antihistamine (chlorpheniramine, 50 mg) were given 1 h before each infusion. Patients were kept in our inpatient unit throughout the treatment period and were examined daily. Complete blood cell counts and serum creatinine levels were determined on days 8 and 16 and on day 31 (groups A and B only) or when warranted clinically; limited resources precluded additional laboratory testing. Treatment was interrupted if the creatinine level doubled from the baseline level and exceeded 2.0 mg/dL or for any value ⭓2.5 mg/dL, and treatment was not restarted until the level decreased to 1.4 mg/dL (i.e., the upper limit of the normal range). Treatment was discontinued and subjects were removed from the study if any adverse event with a Common Toxicity

Table 1. Baseline and posttreatment clinical and laboratory data. Group, day Group A (n p 245) Characteristic

Day 0

Group B (n p 244)

Day 31

Day 0

24 1

Age, years

Group C (n p 500)

Day 31

Day 0

20 1

Group D (n p 496)

Day 16

Day 0

20 1

Day 16

16 1

Percentage of male subjects

63

…

56

…

63

…

60

…

No. (%) of recipients of prior therapy

28 (11)

…

30 (12)

…

46 (9)

…

59 (12)

…

Duration of illness, days

53 6

…

57 3

…

54 2

…

53 3

…

Splenic aspirate score

2.0 0.1

0

1.9 0.6

0

1.9 0.1

0

1.9 0.1

0

Weight, kg

35 1

36 1

31 1

31 1

27 1

27 1

32 1

32 1

Spleen size, cm

4.8 0.2

1.1 0.1

a

5.5 0.2

1.4 0.1

a

4.9 0.1

2.4 0.1

a

5.4 0.2

2.7 0.1

a

8.6 0.1

a

7.6 0.1

8.4 0.1

a

8.0 0.1

a

7.9 0.1

8.1 0.1

a

a

a

a

a

Hemoglobin concentration, g/dL WBC level, ⫻ 10 cells/mm

7.5 0.1

a

3.3 0.1

6.7 0.2

Platelet count, ⫻ 103 platelets/mm3

106 4

231 8

Creatinine level, mg/dL

0.8 .02

0.9 .03

3

3

a a

a

3.5 0.1

7.3 02

116 4

244 8

0.7 .01

0.9 .03

a a

7.8 0.1

a

3.9 0.1

6.2 0.1

117 3

197 4

0.7 .01

1.1 .02

a a

a

3.5 0.1

5.7 0.1

118 3

190 4

0.7 .01

1.1 .02

a a

a

P ! .05, compared with the day 0 value.

Criteria grade ⭓3 occurred. An additional splenic aspirate specimen was obtained for apparent cure evaluation 1 day after treatment ended. Parasite density score for pre- and posttreatment aspirate specimens was graded microscopically in a blinded fashion using a conventional logarithmic scale of 0 (indicating no parasites per 1000 oil immersion fields) to +6 (indicating 1100 amastigotes per oil immersion field) [7]. Designation of apparent cure on day 16 (groups C and D) or day 31 (groups A and B) required clinical improvement, reduction in spleen size, and a splenic aspirate score of 0 (i.e., apparent parasitologic cure) [7, 8]. Definitive cure, assessed after 6 months, required the person to be healthy and to have no signs or symptoms of relapse [7, 8]. All subjects were given 400 rupees (US$9) to offset travels costs for the 6-month visit. Patients who did not return on time were contacted in person. Thirteen of the 1439 subjects who completed treatment were lost to follow-up. Statistical analysis. Data are expressed as mean SEM. Analysis of variance and the multiple range test (Student-Newman-Keuls) were used to detect differences among clinical and laboratory results for the 4 patient groups, except for differences in sex, prior treatment, and response rates, for which the Normal test–Z was used. The paired-samples t test was used to compare pre- and posttreatment values in each group. The Fleiss Quadratic method for proportions was used to compute 95% CIs for the individual proportions of persons who responded to each treatment regimen. A P value of !.05 was considered to be statistically significant. The proportion of patients with a final cure determined the end point of this trial, which was designed to assess noninferiority of the experimental regimens. With a 1-sided a of 0.05, a power of 90%, and a margin of noninferiority of 5%, samples sizes were determined 558 • CID 2007:45 (1 September) • Sundar et al.

assuming cure rates of 91%–95% for 15 daily infusions of 0.75 or 1 mg/kg of amphotericin B and of 95%–99% for 15 alternateday infusions at the same doses and a ratio of 2:1 for random assignment to daily and alternate-day infusions. The numbers of patients required were 472 for the daily treatment arm and 236 for the alternate-day treatment arm. RESULTS Patient characteristics and initial responses. Clinical features and laboratory results at study entry (day 0) were similar in the 4 patient groups (table 1), except for lower age in group D and lower body weight and higher WBC count in group C (P ! .05). One hundred sixty-three subjects had received prior treatment elsewhere (sodium antimony gluconate, 136 patients) or in our unit (amphotericin B, 21 patients; miltefosine, 5 patients; or sitamaquine, 1 patient). Clinically severe kala-azar (spleen size, 18 cm; and hemoglobin concentration, !7 g/dL [7]), was present in 9%–13% of patients in each group. Of 1485 enrolled subjects, 1439 completed the assigned treatment regimen and were evaluable for initial responses (figure 1 and table 2). Those who did not complete the treatment regimen included 14 patients who died, 13 who were removed from the study because of toxicity (table 3), and 19 who defaulted and left the unit. Each of these 46 subjects was considered to have experienced treatment failure in the intention-totreat analysis. At the evaluation 1 day after treatment had ended (day 31 for groups A and B and day 16 for groups C and D), all 1439 patients who completed therapy had parasite-free splenic aspirate smears and fulfilled the criteria for apparent cure (table 2). Compared with baseline (day 0) data, posttreatment eval-


NOTE. Data are mean SEM , unless otherwise indicated. Group A received 1 mg/kg on alternate days for 30 days, group B received 0.75 mg/kg on alternate days for 30 days, group C received 1 mg/kg daily for 15 days, and group D received 0.75 mg/kg daily for 15 days.

Table 2. Responses to amphotericin B treatment among patients with Indian kala-azar. Treatment group Characteristic

Group A

Group B

Group C

Group D

Patients enrolled

245

244

500

496

Completed treatment

237

229

491

482

237

229

491

482

Relapse

2

2

2

2

Lost to follow-up

1

2

6

4

a

Apparent cure

1). All others were asymptomatic and appeared to be healthy at month 6 and were judged to have had a definitive cure response. Thus, in the intention-to-treat analysis, overall cure rates, which were not significantly different (P 1 .05), were as follows: group A, 96%; group B, 92%; group C, 97%; and group D, 96%. The 8 subjects who experienced relapse were re-treated with 15 alternate-day infusions at the same dose that they had been originally given; each responded and was considered to be cured 6 months later.

b

Definitive cure at 6 months No. (%) of patients 95% CI, %

234 (96)

225 (92)

483 (97)

476 (96)

92–98

88–95

95–98

94–97

NOTE. Data are no. of patients, unless otherwise indicated. Group A received 1 mg/kg on alternate days for 30 days, group B received 0.75 mg/kg on alternate days for 30 days, group C received 1 mg/kg daily for 15 days, and group D received 0.75 mg/kg daily for 15 days. a

uation of all 4 groups demonstrated significant decreases in spleen size and increases in hemoglobin concentration, WBC count, and platelet count (table 1). Adverse reactions. Despite the administration of pretherapy medication, in most patients, the initial infusions in each treatment group provoked higher fever and/or rigors—reactions typically associated with amphotericin B. In 1 subject, fever and rigors were severe enough to terminate treatment; other reactions that led to removal from the study are reported in table 3. Anticipated tolerance to infusions eventually developed in a similar fashion in all 4 groups: 24%–34% reacted to dose 10, and 7%–10% reacted to dose 15. The number of episodes of rigors and fever was lower in subjects who received the 0.75-mg/kg dose and in those who received daily treatment (table 3). The data in table 1 (overall group responses) show statistically significant but clinically modest increases in creatinine and hemoglobin levels at the end of treatment. In individual subjects, however, testing during or at the end of treatment demonstrated nephrotoxicity (table 3) and decreases in the hemoglobin concentration of ⭓0.5 g/dL, compared with the baseline concentration, in 12%–15% of patients in groups A and B and 23%–27% of those in groups C and D. These anticipated adverse effects of amphotericin B [3, 6–8] and treatment interruptions (nearly all of which were associated with increased creatinine levels) were significantly more common among patients who received daily treatment (P ! .05 ). In subjects for whom the drug was withheld, therapy was restarted after 4–8 days and completed without incident. Outcome and re-treatment. During the posttreatment period, 13 subjects were lost to follow-up, and symptomatic, parasitologically confirmed relapses occurred in 8 patients (figure

This study was performed to address 3 practical questions about the use of amphotericin B in Indian kala-azar. (1) Is alternateday administration better tolerated? (2) Can the infusion dose be reduced from 1 to 0.75 mg/kg without compromising efficacy? (3) Can the duration of therapy be safely shortened, in this case, to 15 days? For virtually all clinical indications, including Indian kalaazar [2], treatment with amphotericin B is well recognized as being arduous. Indeed, because of concerns about tolerability, we have, in treating 15000 patients in Bihar over the past decade, adhered to our untested bias that alternate-day therapy is sufficiently better tolerated to justify a lengthy 30-day treatment period [5, 7, 8]. The results presented here partially support this bias, because treatment interruptions, which are largely associated with nephrotoxicity, were more common among subjects receiving daily therapy. At the same time, however, in these same patients, infusion-associated reactions were less frequent and less lengthy (data not shown); more importantly, of 981 patients who survived and did not default in the study, 973 (99%) completed the regimen of daily treatment. Although increased creatinine and blood urea nitrogen levels, electrolyte imbalances, and anemia can certainly complicate clinical management, these adverse reactions to amphotericin B are nearly always reversible once treatment ends. Early studies of Indian kala-azar raised the possibility of using low-dose amphotericin B (0.5 mg/kg) given as 14 alternate-day infusions over 28 days [9]. However, subsequent trials, which primarily tested once-daily administration for 20 days, indicated otherwise, concluding that optimal long-term cure rates (e.g., 99%–100%) required dosing at 1 mg/kg [6, 10]. We also used the 1-mg/kg dose, but in a 15-infusion alternate-day regimen, and we achieved similar cure rates of 96–97% in recent trials in Bihar [7, 8]. However, an earlier study [4] and our own uncontrolled experience with 1500 patients (authors’ unpublished data) [5] have also suggested that a dose of 0.75 mg/ kg is as active as a dose of 1 mg/kg in the 15-infusion, alternateday regimen. Because the usefulness of a dose of !1 mg/kg had not been formally settled, the comparative data presented here are thus of additional clinical interest, because they demonstrate Amphotericin B in Indian Kala-Azar • CID 2007:45 (1 September) • 559


Apparent cure evaluation 1 day after the end of treatment, which was day 31 for groups A and B and day 16 for groups C and D. b Patients who died or were removed from study, defaulted (i.e., left against advice), experienced relapse, or were lost to follow-up at month 6 were considered to have experienced treatment failure.

DISCUSSION

Table 3. Adverse reactions during treatment. Treatment group Characteristic

Group A

No. of patients treated Infusion-associated episodes

245 6.8 0.2

Rigors

7.7 0.2

Fever Removal from study All patients

Group B 244

Group C

Group D

500

496

5.6 0.2

5.9 0.1a

4.9 0.1a

6.3 0.2

a

5.8 0.1a

6.6 0.1

3 (1)

2 (1)

4 (1)

4 (1)

Vomiting and/or diarrhea Hepatotoxicity

1 1

1 …

3 1

2 …

Infusion reaction Nephrotoxicity

1 …

… …

… …

… 1

Severe thrombocytopenia Recurrent hypothermia

… …

1 …

… …

… 1

b

Treatment interruptions

a

8 (3)

12 (5)

29 (6)

38 (8)

Elevated creatinine level

8

11

29

37

NOTE. Data are no. (%) of patients or mean no. of episodes SEM . Group A received 1 mg/ kg on alternate days for 30 days, group B received 0.75 mg/kg on alternate days for 30 days, group C received 1 mg/kg daily for 15 days, and group D received 0.75 mg/kg daily for 15 days. a

P ! .05 for group A versus group C and for group B vs. group D. Treatment interruptions were made because (A) the patient had a 12-fold increase in the serum creatinine level (to 12.0 mg/dL), compared with the baseline value, or had any creatinine level ⭓2.5 mg/dL; (B) the patient had hepatitis (1 patient in group B); or (C) the patient continued to experience vomiting (1 patient in group D). b

similar efficacy for 0.75 and 1 mg/kg in both of the treatment schedules tested. In kala-azar, amphotericin B therapy is arduous, not only because of infusion-associated reactions and other toxicities, but also because of the lengthy duration of treatment (20–30 days). Although the use of lipid formulations of amphotericin B remedied these drawbacks [2, 3, 7], high cost has rendered these effective agents largely irrelevant in the Indian subcontinent. Thus, in the present trial, we also sought to determine whether the 15-infusion regimen could be successfully abbreviated by providing amphotericin B once daily. Results for nearly 1000 patients in groups C and D clearly indicate the high-level efficacy of the 15-day regimen. This effect of daily treatment, allowing for a shorter duration of hospitalization, is particularly relevant in an impoverished region where hospital beds are at a premium and long inpatient stays equate importantly with lost wages and appreciable family life disruption [11, 12]. It is also possible that an even shorter daily regimen might be effective. However, it is worth pointing out that it is not known what role the duration of therapy (compared with the daily dose or total dose administered) plays in amphotericin B’s clear-cut efficacy in Indian kala-azar. In contrast, it is known that meaningful cost savings result from reducing the duration of treatment and using lower drug doses, even in developing countries such as India, where medical charges are low [7, 12]. Although the costs of the drug in 560 • CID 2007:45 (1 September) • Sundar et al.

India for a 25-kg patient treated with 15 infusions of 1 or 0.75 mg/kg of amphotericin B are only ∼US$50 and ∼ $40, respectively [3], this expense alone is often an obstacle to treatment. The overall economic impact of kala-azar and the hardship it poses for individuals and families in regions of India and neighboring countries where it is endemic should not be underestimated. These patients are among the poorest, and they often sell possessions or take on loans to purchase the drug and to pay for medical care [11–13]. On the basis of prior calculations for a 25-kg patient in Bihar [7], the total per-patient cost (i.e., the costs of the drug plus hospitalization) for the regimen used in group D in this study (i.e., 15 once-daily infusions of 0.75 mg/kg) would be ∼US$220, compared with ∼US$400 if the drug is given on alternate days (group B). It should also be made clear that approvals of miltefosine (in 2002) and paromomycin (in 2006) in India have provided important, new alternative treatments for kala-azar. As the first effective oral agent [3, 8], miltefosine, used in a 28-day course, has made self-administered outpatient therapy a reality. Miltefosine’s “public” price is projected to be ∼US$60 per adult treatment course for drugs supplied to the World Health Organization, the Indian government, and nongovernmental organizations in India. Paromomycin, a rediscovered aminoglycoside, is given once daily for 21 days by intramuscular injection [3] and thus may also prove suitable in the future for outpatient use. Although the drug is not yet available for distribution in


a

All

Acknowledgments Financial support. Sitaram Memorial Trust and AB Foundation. Potential conflicts of interest. All authors: no conflicts.

References 1. Sundar S, More DK, Singh MK, et al. Failure of pentavalent antimony in visceral leishmaniasis in India: report from the center of the Indian epidemic. Clin Infect Dis 2000; 31:1104–6.

2. Murray HW, Berman JD, Davies CR, Saravia NG. Advances in leishmaniasis. Lancet 2005; 366:1561–77. 3. Olliaro PI, Guerin PJ, Gerstl S, Haaskjold AA, Rottingen J-A, Sundar S. Treatment options for visceral leishmaniasis: a systematic review of clinical studies done in India, 1980–2004. Lancet Infect Dis 2005; 5: 763–74. 4. Giri OP, Singh AN. Experience with amphotericin B in sodium stibogluconate-unresponsive cases of visceral leishmaniasis in north Bihar. J Assoc Physicians India 1994; 42:690–1. 5. Sundar S, Rai M. Advances in the treatment of leishmaniasis. Curr Opin Infect Dis 2002; 15:593–8. 6. Thakur CP, Singh RK, Hassan SM, Kumar R, Narain S, Kumar A. Amphotericin B deoxycholate treatment of visceral leishmaniasis with newer modes of administration and precautions: a study of 938 cases. Trans R Soc Trop Med Hyg 1999; 93:319–23. 7. Sundar S, Mehta H, Suresh AV, Singh SP, Rai M, Murray HW. Amphotericin B treatment for Indian visceral leishmaniasis: conventional versus lipid formulations. Clin Infect Dis 2004; 38:377–83. 8. Sundar S, Jha TK, Thakur CP, et al. Oral miltefosine for Indian visceral leishmaniasis. N Engl J Med 2002; 347:1739–46. 9. Mishra M, Biswas UK, Jha AM, Khan AB. Amphotericin B versus sodium stibogluconate in first-line treatment of Indian kala-azar. Lancet 1994; 344:1599–600. 10, Thakur CP, Ahmed S. Observations on amphotericin B treatment of kala-azar given in a rural set up in Bihar, India. Indian J Med Res 2001; 113:14–8. 11. Thakur CP. Socio-economics of visceral leishmaniasis in Bihar (India). Trans Roy Soc Trop Med Hyg 2000; 94:156–7. 12. Meheus F, Boelaert M, Baltussen R, Sundar S. Costs of patient management of visceral leishmaniasis in Muzaffarpur, Bihar, India. Trop Med Int Health 2006; 11:1715–24. 13. Alvar J, Yactayo S, Bern C. Leishmaniasis and poverty. Trends Parasitol 2006; 22:552–7. 14. Sundar S, Jha TK, Thakur CP, Sinha PK, Bhattacharya SK. Injectable paromomycin for visceral leishmaniasis in India. N Engl J Med 2007; 356:2571–81.

Amphotericin B in Indian Kala-Azar • CID 2007:45 (1 September) • 561


India, clear-cut efficacy, apparently low adverse reaction rates, and a drug cost projected to be US$5–US$510 per treatment course [3, 14] indicate the promise of paromomycin. Nevertheless, until that promise is realized in the field, amphotericin B remains the first-line parenteral therapy in Bihar. The results we present here identify the usefulness of a new regimen (once-daily infusions of 0.75 mg/kg for 15 days) that should help to diminish, at least in part, several drawbacks of treatment with amphotericin B. This regimen incorporates high-level efficacy in adults and children, reasonable tolerability, shorter duration, and reduced cost. However, and as with any amphotericin B deoxycholate treatment protocol, this 15-day regimen should be accompanied by redetermining serum creatinine (and, preferably, serum potassium) levels at least once midway through the therapy regimen. In some treatment units in India, including primary community health centers, such monitoring is simply not possible; thus, giving 15 alternateday infusions of 1 mg/kg would still appear to be prudent in these settings.