Table 3. Risk of bias assessment.

Reference	Random sequence generation	Allocation concealment	Blinding (Objective outcomes)	Blinding (Subjective outcomes)	Incomplete outcome data and withdrawals (Objective outcomes)	Incomplete outcome data and withdrawals (Subjective outcomes)	Free of selective reporting?	Other sources of bias and commentaries
Chrusciak-Talhari 2011 [73]	Low	Unclear	Low “Standardized digital photograph”	Unclear (for patient reported adverse events)	Low	NA	Low: retrospective registration: NCT00600548; Authors presented results on all outcome measures that were pre specified as relevant	Low: funding: public (FINEP and CNPq./Brazil)
Lopez 2012 [71]	Low: generated list in blocks of eight using EpiInfo	Low: “Only the clinical coordinator of the study had access to the list and was in charge of allocating treatments”	Unclear: no description of the measurement process	Low: the assessment of side effects was done according to the Common Terminology Criteria for Adverse Events	Low	Low	Low: retrospective registration: NCT00471705; authors presented results on all outcome measures that were pre specified as relevant	Low: funding, Public. Social Protection Ministry of the Republic of Colombia
Lopez-Jaramillo 2010 [81]	Low: randomization list using a computer program.	Low: randomization process was blinded and centralized. The assigned code was reported to the monitoring nurse who had no contact with the participants.	Unclear: no further description of the measurement process nor blinding methods	Unclear: no further description of the measurement process nor blinding methods	Unclear: 35/178 lost to follow up. Difference according to groups.	Unclear: 35/178 lost to follow up. Difference according to groups.	Low: authors presented results on all outcome measures that were pre specified as relevant	Low: funding: Public. Institute for Science and Technology “COLCIENCIAS
Machado 2010 [74]	Low: randomization list using a computer program.	Unclear	Low: “The area involved was calculated as the product of the two measurements. A standardized digital photograph was also taken from each patient’s lesions at the same time points”	Unclear	Low	Low	Low: Retrospective registration: NCT00600548; Authors presented results on all outcome measures that were pre specified as relevant	Low: Funding: Mix Brazilian National Research Council (CNPq). Miltefosine was supplied by Aeterna Zentaris GmbH.
Miranda-Verastegui 2009 [76]	Low: randomization list using a computer program.	Low: “Numbers and corresponding treatment packages were prepared so that both subjects and study investigators were blind to treatment allocation throughout the study”.	Low: both creams were identical in appearance. Standardized photograph of each lesion	Low: both creams were identical in Standardized photograph of each lesion	Low	Low	Low: prospectively registered: NCT00257530; authors presented results on all outcome measures that were pre specified as relevant	Low: Funding: mix. Drugs for Neglected Diseases Initiative (not-for-profit organization). Pharmaceuticals Inc. provided the randomized allocation of imiquimod and placebo creams at no cost
Neves 2011 [69]	Low: list of random distribution generated by a biostatistician	Unclear: not reported	Unclear	Unclear	Unclear: 75.7% of the patients randomized to Amphotericin B refused to continue in the study when they learned that they would need to come to the hospital for at least 20 days	Unclear: 75.7% of the patients randomized to Amphotericin B refused to continue in the study when they learned that they would need to come to the hospital for at least 20 days. 8.1% and 5.4% were loss to follow up in in the meglumine and the pentamidine groups respectively	Unclear: some pre specified outcomes were not considered (i.e. relapse); - The trials was not not registered.	Low: fundings: Financiadora de Estudos e Projetos (Research and Projects Financing) of the Ministry of Science and Technology - FINEP.
Newlove 2011 [72]	Low: randomization table	Unclear: sealed envelopes (opaque or numerated not reported)	Low: placebo was identical in form, color, and number to treatment	Low: placebo was identical in form, color, and number to treatment	Low: no loss to follow up	Low: no loss to follow up	Unclear: prospectively registered: NCT00469495; some pre specified outcomes were not considered (i.e. relapse);	Low: funding: public. NIH/FIC and NIH/NIAID
Rubiano 2012 [70]	Low: computerized balanced block randomization scheme	Low: coordinating center via phone call	Low: to eliminate ascertainment bias, treatment outcome was determined by a masked evaluator using standardized photographs of lesions.	Low: adverse events were identified by study personnel using a structured questionnaire and classified according to Common Terminology Criteria for Adverse Events	Low	Low	Low: prospective registration NCT00487253; authors presented results on all outcome measures that were pre specified as relevant	Low: funding: mix Departamento Administrativo de Ciencia, Tecnologia e Innovacion (COLCIENCIAS); National Institute of Allergy and Infectious Diseases International Collaborations in Infectious Disease Research Program; Fogarty Global Infectious Diseases Research Training Program
Soto 2008 [77]	Unclear: not reported	Unclear: not reported	Unclear	Unclear	Low	Low	Unclear: prospective registration: NCT00233545; not all pre specified outcomes were reported.	Low: funding: NGO AB Foundation
Velez 2010 [75]	Low: list generated randomly in blocks of eight (EpiInfo)	Low: only the study coordinator had access to the list	Unclear	Unclear: adverse events were evaluated according to standard criteria	Low: 14.7% were lost in the miltefosine group while 12.6% were lost in the meglumine antimoniate group. Reasons were explained.	Low	Low: not registered; authors presented results on all outcome measures that were pre specified as relevant	Low: funding: public Ministerio de la Protección Social de la República de Colombia