Diuretic treatment in cirrhotic patients with ascites increases ascitic fluid concentration of total protein and complement components, and opsonic activity. These changes are not observed in patients treated with paracentesis. Based on these data it has been suggested that therapeutic paracentesis may be associated with an increased risk of spontaneous bacterial peritonitis (SBP) development. To assess this possibility, 80 cirrhotic patients with tense ascites were randomly allocated in two therapeutic groups: group 1 (40 patients) was treated with total paracentesis associated with plasma volume expansion and group 2 was treated with diuretics. After mobilization of ascites, patients from both groups received diuretics to avoid reaccumulation of ascites; cases that developed tense ascites during follow-up (mean follow-up period, 60 +/- 6 and 55 +/- 4 weeks, respectively) were treated according to initial randomization. Patients from both groups had similar results regarding baseline clinical and standard laboratory data, ascitic fluid concentration of total protein, complement components, and opsonic activity. Sixteen patients (7 from group 1 and 9 from group 2) developed SBP during the study period. The 4-week and 1-year probability of SBP occurrence were 2.5% and 18.6%, respectively, in group 1 patients, and 11.9% and 24%, respectively, in group 2 patients. Therefore, our study indicates that therapeutic paracentesis does not increase the early- and long-term risk of SBP development in cirrhotic patients with tense ascites.