Objective: To compare the tolerability of intravenous azithromycin, clarithromycin, erythromycin and placebo. Design: A double-blind, double-dummy, placebo-controlled, four-way crossover study was conducted in 12 healthy male volunteers. The participants were randomised to receive I-hour intravenous infusions of azithromycin 500mg (2 mg/ml) once daily, erythromycin 500mg (1 mg/ml) three times daily, clarithromycin 500mg (2 mg/ml) twice daily, or placebo (normal saline, 500ml) three times daily, with each regimen administered for 3 days. There was a minimum 4-week washout period before participants switched to an alternative regimen, in random sequence, until all four regimens had been completed. Participants were monitored for infusion site reactions and gastrointestinal (GI) adverse events. Results: Clarithromycin caused clinically significant infusion site pain in 92% of the 12 participants evaluated and was exclusively associated with phlebitis and inflammation. Areas under score-time curves (AUSs) rating the intensity of inflammation and pain were significantly higher for clarithromycin compared with azithromycin (p < 0.0005). Erythromycin infusion caused clinically significant abdominal pain or nausea in 25% of participants. The AUSs for GI tolerability were significantly different for erythromycin compared with azithromycin (p < 0.001). Discontinuation rates due to infusion site reactions were 0% for azithromycin, 50% for clarithromycin, and 8% each for erythromycin and placebo. Treatment with erythromycin was interrupted or discontinued as a result of abdominal pain in 17% of patients and as a result of nausea in 8% of patients. Conclusions: Intravenous azithromycin had better infusion site tolerability than clarithromycin and better GI tolerability than erythromycin. The superior tolerability of azithromycin may avoid the discontinuation of intravenous antimicrobial therapy in seriously ill patients and assist in reducing the duration of hospitalisation and the cost of patient management.