Spontaneous bacterial peritonitis (SBP) is a severe infectious complication in cirrhotic patients, and initial antibiotic therapy must be empirical. An initial study published in 1985 found that cefotaxime administered at a dose of 2 g every 4 h was more effective and safer than the combination of tobramycin-ampicillin. Since then, cefotaxime has been considered the agent of choice in the empiric therapy of SBP. Subsequent publications showed that a dosage of 2 g every 6 h was also adequate in this infection. More recent studies have demonstrated that the high efficacy of cefotaxime in SBP can be maintained by using lower dosages than those used in initial investigations. In one of these studies, a dose of 2 g every 8 h for 5 days was found to be as effective as the same dose for 10 days. Finally, a prospective, randomized multicenter trial aimed at comparing the efficacy of two different dosages of cefotaxime, 2 g every 6 h versus 2 g every 12 in a large series of cirrhotic patients with SBP, showed that both dosages resulted in similar rates of resolution of infection and survival. Despite the reasonably adequate rate of infection resolution in SBP patients, the in-hospital mortality rate remains high as a result of complications such as renal failure. Further studies should therefore be addressed to reducing the incidence of these complications and thus improving survival.