Background: Bartonella henselae, Bartonella quintana, and Bartonella bacilliformis are responsible for the majority of cases of bartonellosis in humans. These species have various unique epidemiologic characteristics, clinical manifestations, and treatment approaches. The objective of this study was to summarize the evidence on the treatment for the three most common species of Bartonella in humans. Methods: We searched electronic databases through August 2011 for randomized controlled trials and observational studies designed to evaluate the efficacy and safety of the regimens used to treat diseases produced by B. henselae, B. quintana, and B. bacilliformis. Study selection and appraisal were done in duplicate. Results: We found two randomized and seven non-randomized studies at high risk of bias. For cat scratch disease, antibiotics did not significantly affect the cure rate or time to achieve cure. In chronic bacteremia, gentamicin and doxycycline significantly increased the resolution rate. The recommended treatment was not better than other regimens for infectious endocarditis and bacillary angiomatosis. Conclusions: Current clinical practice for the treatment of bartonellosis relies mostly on expert opinion and antimicrobial susceptibility data. Randomized controlled trials are needed in the field to compare different treatment options. © 2013 International Society for Infectious Diseases.