Low molecular weight heparins are as effective as unfractionated heparin in deep-vein thrombosis (DVT) prophylaxis for major surgery. However, there is no evidence nor consensus for prophylaxis in medical patients. We compared the efficacy and safety of nadroparin calcium (nadroparin) with placebo in medical patients at high risk of DVT. A total of 223 patients mechanically ventilated for acute, decompensated chronic obstructive pulmonary disease, were randomized to treatment with subcutaneous nadroparin adjusted for body weight (0.4 ml, i.e., 3,800 AXa IU, or 0.6 ml, i.e., 5,700 AXa IU) or placebo. The average duration of treatment was 11 d. The incidence of DVT in patients receiving nadroparin was significantly lower than that in patients receiving placebo (15.5 versus 28.2%; p = 0.045). Although the incidence of adverse events was high in both groups, there were no significant differences between nadroparin and placebo for total adverse events (46.3 versus 39.8%; p = 0.33), serious adverse events (25.0 versus 19.5%; p = 0.32), or those resulting in early permanent discontinuation of treatment (12.0 versus 8.8%; p = 0.44). The most common adverse event was hemorrhage. There was the same number of deaths in both treatment groups. Subcutaneous nadroparin resulted in 45% decrease in incidence of DVT compared with placebo.