Objective To investigate the efficiency and safety of argatroban for anticoagulation therapy of critical ill patients. Methods Prospective randomized controlled study was adopted. A total of 62 patients with high risk of thrombosis who were expected to stay in ICU for more than 7 d were selected and randomly divided into the argatroban group (n=32) and enoxaparin group (n=30). Patients of the argatroban group were intravenously injected with 20 mg of argatroban twice a day for 7 d and patients of the enoxaparin group were injected with 4 000 IU (0.4 mL) of enoxaparin once a day for 7 d. Other drugs and organ support therapy of two groups were the same. Venous blood of two groups was drawn before treatment and at 24 h, 72 h, and in the morning of 7 d after treatment. Coagulation indexes and liver and kidney functions were detected. Color ultrasound examination was conducted to detect whether deep vein thrombosis (DVT) has developed. Clinical symptoms of DVT, bleeding, ICU stay time, and mortality of hospitalization time of 28 d of patients were observed. Results Prothrombin time, international normalized ratio, and activated partial thromboplastin time of two groups after treatment were significantly higher than those of before treatment. Platelet count, fibrinogen, and D-dime were at low end of normal range. The argatroban group and enoxaparin group met the goal of anticoagulation therapy 24 h and 72 h after treatment, respectively. Both groups met the goal of therapy 7 d after treatment. The differences between two groups at 24 h and 72 h were statistically significant (P<0.05), while the differences between two groups on 7 d were not statistically significant (P>0.05). For the argatroban group, 1 patient developed DVT, 1 patient developed bleeding, and no allergic reactions and liver dysfunction occurred. For the enoxaparin group, 6 patients developed DVT, 4 patients developed bleeding, 1 patient developed allergic reactions, and 1 patient developed liver damage. Compared with the enoxaparin group, the mechanical ventilation time (h) and ICU stay time (d) of the argatroban group were significantly shorter (160.50±30.36 vs 183.60±29.85, 11.41±3.51 vs 15.83±4.95, P<0.05). However, the mortality of hospitalization time of 28 d remained the same. Conclusion Argatroban can quickly meet the goal of anticoagulation therapy, satisfactorily prevent the development of DVT, and shorten the ICU stay time and mechanical ventilation time with low incidence of adverse events, but does not change the mortality of hospitalization time of 28 d.