We conducted a two-part meta-analysis to assess the effectiveness of fluoroquinolones for preventing bacterial infections in granulocytopenic patients who are receiving chemotherapy for malignancies. Overall, 19 randomized studies met selection criteria and were included in this meta-analysis of 2,112 patients. Thirteen studies that compared the fluoroquinolones alone with control regimens (co-trimoxazole, oral nonabsorbable antibiotics, or placebo) and six studies that compared the fluoroquinolones plus prophylaxis for bacteremia due to gram-positive bacteria with control regimens (fluoroquinolones or oral nonabsorbable antibiotics) were included in the two meta-analyses. The results of the first meta-analysis indicate that fluoroquinolones alone are effective in preventing gram-negative bacteremia (overall odds ratio [OR], 0.09; 95% confidence interval [CI], 0.05-0.16; P < .001), but not gram-positive bacteremia (OR, 1.05; 95% CI, 0.76-1.45; P = .7), fever-related morbidity (OR, 0.76; 95% CI, 0.56-1.04; P = .09), and infection-related mortality (OR, 0.79; 95% CI, 0.47-1.34; P = .4). The results of the second meta-analysis indicate that a combination of fluoroquinolones plus prophylaxis for gram-positive bacteremia (penicillin, vancomycin, or macrolides) significantly reduces the occurrence of gram-positive bacteremia (OR, 0.46; CI, 0.33-0.63; P < .001) without affecting the incidence of fever-related morbidity (OR, 0.83; 95% CI, 0.62-1.13; P = .2) and infection-related mortality (OR, 0.74; 95% CI, 0.40-1.38; P = .3)