To investigate the value of drug intervention for difficult weaning from mechanical ventilation. A prospective single-blind randomized controlled trial was conducted. 120 patients with difficult weaning from mechanical ventilation encountered in Department of Critical Care Medicine of Peking University Third Hospital from January 2008 to December 2013 were included, and the patients were divided into treatment group and control group according to random number table, with 60 cases in each group. Patients received furosemide therapy in the treatment group 3 days before weaning up to 48 hours after weaning in order to control negative liquid balance. Enema was given the day before weaning to reduce abdominal pressure. On the weaning day, all of the patients received nitroglycerin and beta blocker or cedilanid to prevent or control elevation of blood pressure and heart rate in the process of weaning. All patients in treatment group received anisodamine in small dosage 2 hours before extubation. The patients in control group received conventional treatment without drug intervention. Baseline indexes of two groups were compared, including the heart rate, respiration rate (RR), mean arterial pressure (MAP), pulse blood oxygen saturation (SpO₂), blood gas, hemoglobin (HG), albumin (ALB) and creatinine (Cr). The main reasons of difficulty in weaning, sedative and analgesic drug selection, presence of abdominal discomfort before weaning, interval between sputum suction before extubation, liquid balance at the beginning of the investigation and at time of weaning, 24 hours and 48 hours after weaning, failures of spontaneous breathing test (SBT), length of mechanical ventilation,length of ICU stay, and total length of mechanical ventilation and total length of ICU stay during hospitalization. There was no statistically significant difference in the heart rate, RR, MAP, SpO₂, blood gas, HG, ALB, Cr at the beginning of the investigation between the two groups. The main reasons for difficult weaning in both groups of patients were respiratory dysfunction, cardiac insufficiency, and central nervous system dysfunction. The use of propofol combined dexmedetomidine in the treatment group was more frequent than the control group [16.7% (10/60)vs. 1.7% (1/60), χ² = 8.107, P=0.004], and there was no statistically significant difference in the use of other combinations of sedative drugs between the two groups. Abdominal discomfort before weaning was milder in treatment group as compared with control group [10.0% (6/60) vs. 25.0% (15/60), χ² = 4.675, P=0.031]. The interval between sputum suction before extubation in the treatment group was significantly longer than that of the control group [hours: 1 (1, 2) vs. 1 (1, 1), Z=-2.209, P= 0.027]. SBT failure was less frequent in treatment group compared with control group [times: 0 (0, 1) vs. 1 (1, 2), Z=-6.561, P=0.000]. Liquid balance was better in the treatment group than the control group at time of weaning, 24 hours and 48 hours after weaning [at time of weaning: -567.71(-755.95,-226.41) vs. 1 256.76 (472.48, 1 796.63), Z=-9.038, P=0.000; 24 hours after weaning: -5.03 (-530.28, 245.09) vs. 342.28 (125.36, 613.25), Z=-4.711, P=0.000; 48 hours after weaning: 115.50 (-450.26,485.00) vs. 330.00 (16.25,575.25), Z=-1.932, P=0.053]. Compared with control group, length of mechanical ventilation [days: 1.0 (1.0, 2.0) vs. 2.0 (2.0, 3.0), Z=-6.545, P=0.000], ICU stay time [days: 3.0 (3.0, 4.0) vs. 4.0 (4.0, 5.0), Z=-6.545, P=0.000], and total length of mechanical ventilation [days: 8.0 (6.0,12.0) vs. 11.0 (8.0, 15.0), Z=-4.091, P=0.000] and total length of ICU stay during hospitalization [days: 12.5 (9.2, 19.0) vs. 17.0 (12.0, 29.5), Z=-2.722, P=0.000] were all significantly shorter in the treatment group. Adjuvant drugs therapy is helpful in patients weaning from the mechanical ventilation, and can shorten length of mechanical ventilation and ICU stay time. Propofol, combined dexmedetomidine, is helpful for weaning.