Pentoxifylline is effective in prevention of hepatorenal syndrome (HRS) in alcoholic hepatitis. The aim of this study was to assess the efficacy of pentoxifylline for prophylaxis of HRS in patients with cirrhosis and ascites. One hundred and seventy-six consecutive patients with cirrhosis and ascites were screened. Patients with creatinine clearance (Ccl) between 41 and 80 ml/min and serum creatinine of less than 1.5 mg/dl in absence of renal disease were randomized to receive either pentoxifylline (group A, 1200 mg/day) or placebo (group B) for 6 months. Patients were followed monthly for 6 months, and kidney function tests were carried out at baseline, 1, 3, and 6 months. Primary endpoint was the development of HRS within 6-month follow-up. Thirty-five patients each were randomized to group A and group B. Of the 70 patients, 61 completed follow-up (group A, n = 30 and group B, n = 31). In group A, serum creatinine remained stable at 1 month (0.94±0.2 vs. 0.90±0.4 mg/dl, P = 0.43), at 3 months (0.94±0.2 vs. 0.80±0.3 mg/dl, P = 0.10), and at 6 months (0.94±0.2 vs. 0.8±0.2 mg/dl, P = 0.06). Improvement occurred in Ccl at 1 month (61.7±16.0 vs. 82.0±30.0 ml/min, P = 0.001) and at 3 months (61.7±16.0 vs. 86.2±30.7 ml/min, P = 0.001). Serum sodium in group A showed progressive improvement at 1 month (134.7±4.4 vs. 137.0±5.3 mmol/l, P = 0.006), at 3 months (134.7±4.4 vs. 138.4±5.4 mmol/l, P = 0.004), and at 6 months (134.7±4.4 vs. 137.6±4.9 mmol/l, P = 0.009). Similarly, mean arterial pressure (MAP) also showed progressive improvement at 1 month (76.7±6.7 vs. 81.8±6.9 mmHg, P = 0.001), at 3 months (76.7±6.7 vs.82.3±6.2 mmHg, P = 0.001), and at 6 months (76.7±6.7 vs. 82.6±6.1 mmHg, P = 0.01). Tumor necrosis factor (TNF) level measured at baseline and at 3 months (14.4±4.8 vs. 13.1±3.7 pg/ml, P = 0.28) showed no difference in group A patients. In group B, serum creatinine increased gradually at 1 month (0.80±0.2 vs. 1.0±0.2 mg/dl, P = 0.007), at 3 months (0.80±0.2 vs. 1.1±0.3 mg/dl, P = 0.001), and at 6 months (0.80±0.2 vs.1.1±0.2 mg/dl, P = 0.001). Ccl at 1 month (63.1±14.5 vs. 66.8±28.2 ml/min, P = 0.37) decreased at 3 months (63.1±14.5 vs. 54.4±18.3 ml/min, P = 0.008). Serum sodium, in group B, showed progressive decline at 1 month (136.2±5.0 vs. 132.7±3.2 mmol/l, P = 0.02) and at 6 months (136.2±5.0 vs. 132.7±3.4 mmol/l, P = 0.002). MAP remained stable at 1 month (75.1±7.5 vs. 75.3.8±5.1 mmHg, P = 0.86), at 3 months (75.1±7.5 vs. 73.2±4.6 mmHg, P = 0.19), and at 6 months (75.1±7.5 vs. 74.1±6.1 mmHg, P = 0.06). TNF level also did not show any change at baseline and at 3 months (14.5±6.8 vs. 13.4±5.1 pg/ml, P = 0.31). Of the 12 patients who developed HRS, 10 patients were in group B (type 1 HRS, n = 9 and type 2 HRS, n = 1) and two patients (type-1 HRS, n = 2) were in group A (P = 0.01). Patients with HRS had higher baseline TNF-α (15.3±5.8 vs. 10.9±4.8 pg/ml, P = 0.01), lower MAP (68.0±3.8 vs. 77.8±6.5 mmHg, P = 0.01), and sodium level (131.2±3.0 vs. 135.6±4.7 mmol/l, P = 0.003) than those who did not develop HRS. Pentoxifylline is effective in preventing HRS in patients with cirrhosis and ascites at risk of HRS.