Large volume paracentesis (4.8 to 15.5 liters) was performed in 42 patients with post-hepatitic cirrhosis and massive ascites, not only to derive parameters capable of predicting the development of severe clinical hypotension after large volume paracentesis, but also to determine the optimal time to introduce preventive volume expanders. Systemic hemodynamics were sequentially measured for 72 hours in thirty-two patients. Severe clinical hypotension occurred in 13 (31.0%) patients 4-62 hours from the start of paracentesis. Univariate analysis, with the Mantel-Cox test used to compare Kaplan-Meier curves, and the subsequent multivariate analysis by stepwise Cox regression procedure were utilized to identify two variables, withdrawn ascitic fluid greater than 7.5 liters (p = 0.0121) and the absence of peripheral edema (p = 0.0148), reaching statistical significance to predict the occurrence of severe clinical hypotension. Compared to the baseline value, the cardiac output of patients not developing severe clinical hypotension increased (6.26 +/- 0.66 vs. 6.65 +/- 0.69 liter/min, p < 0.01) one hour from the start of paracentesis and right atrial pressure decreased (11.2 +/- 2.4 vs. 8.7 +/- 2.3 mmHg, p < 0.05). The cardiac output returned to the baseline value at the 9th hour. Based on the results presented herein, we can conclude that severe clinical hypotension occurs in a high percentage of patients with post-hepatitic cirrhosis and massive ascites within 72 hours from the start of large volume paracentesis. At potential risk of this occurring are those patients without peripheral edema and withdrawn ascitic fluid greater than 7.5 liters. Volume expanders should be introduced before 4th hour from the start of large volume paracentesis.