In the mid-1980s it was discovered that serotonin (5-hydroxytryptamine; 5-HT) was at least partially responsible for producing chemotherapy-induced nausea and vomiting. It was therefore realised that serotonin receptor blockade with serotonin 5-HT3 receptor antagonists could inhibit chemotherapy-induced nausea and vomiting. 5-HT3 antagonists have different chemical structures and receptor binding affinity. Granisetron, dolasetron and its major metabolite are pure 5-HT3 antagonists, while ondansetron and tropisetron are weak antagonists at the 5-HT4 receptor. Ondansetron has also been demonstrated to bind at other serotonin receptors and to the opioid mu receptor. The half-lives of granisetron, tropisetron and the active metabolite of dolasetron are 2 to 3 times longer than that of ondansetron. These observations initially suggested that more frequent ondansetron administration would be required; however, it has now been shown that receptor blockade does not correlate with elimination half-life and all 5-HT3 antagonists can be effectively administered once daily. Clinical trials have been conducted that directly compare the 5-HT3 antagonists. To compare these studies, it is necessary to assess trial design, including known risk factors for the development of chemotherapy-induced nausea and vomiting, and response criteria. Stratification for risk factors, use of strict efficacy criteria and randomisation to a blinded trial using an appropriate comparative regimen are essential for a well designed antiemetic trial. Comparative clinical trials using various doses, routes and regimens of administration have been conducted with 5-HT3 antagonists. Despite some trial design shortcomings, most of the studies show equal efficacy between the agents, especially in moderately emetogenic chemotherapy and mild, infrequently occurring adverse effects. The addition of steroids also appears to improve outcome. However, since many doses and regimens of ondansetron were used, further study is needed to determine the optimal regimen. The efficacy of 5-HT3 antagonists in controlling delayed nausea and vomiting from chemotherapy is less well studied. Further, there is no good scientific rationale for the use of 5-HT3 antagonists in controlling delayed nausea and vomiting since serotonin has not been shown to be released during the delayed phase. In fact, most studies show no benefit or modest benefit of 5-HT3 antagonists over placebo. Because the 5-HT3 antagonists perform similarly in the clinical setting, pharmacological differences do not seem to translate into therapeutic differences. There is also no appreciable difference in the incidence or severity of adverse effects among the 5-HT3 antagonists. Determination of clinical use may then be driven by cost.