Objective: To evaluate the safety and efficiency of patient-controlled intravenous analgesia (PCIA) using hydromorphone supplement with dexmedetomidine on patients undergoing transcatheter arterial chemoembolization. Methods: One hundred and eighty patients, age ranged from 40 to 65 years, body mass index from 18 to 25 kg/m(2,) ASA physical status Ⅱ-Ⅲ, who were scheduled for transcatheter arterial chemoembolization (TACE) under monitor anesthesia care (MAC) were randomly divided into 3 groups: hydromorphone group (H group), hydromorphone supplement with dexmedetomidine 1 μg/kg group (D1 group), hydromorphone supplement with dexmedetomidine 2 μg/kg group (D2 group), 60 patients in every group. All the groups of patients received PCIA pump, in the H group, the PCIA reagent was composed of 120 μg/kg hydromorphone and 5 mg tropisetron in 100 ml of normal saline. In comparison, PCIA regiment was composed of 120 μg/kg hydromorphone, 1 μg/kg dexmedetomidine and 5 mg tropisetron in 100 ml of normal saline in the D1 group, while 120 μg/kg hydromorphone, 2 μg/kg dexmedetomidine and 5 mg tropisetron in 100 ml of normal saline in the D2 group. The visual analogue scale (VAS) score, the observer's assessment of alertness/sedation scale (OAA/S) score, patients' satisfaction index, consumption of hydromorphone, the additional dose of morphine, the effective pressing times of PCIA and adverse reactions were recorded in detail at 0, 0.5, 1, 4, 12 and 24 hours after the patients underwent TACE. Results: The total consumptions of hydromorphone were (4.3±0.1), (4.1±0.1), and (3.8±0.1) mg in group H, D1, and D2, respectively, and the effective pressing times were 13±3, 6±2 and 2±1, the additional doses of morphine were (30±5), (15±3), and (3±1) mg, and adverse reaction rates were 45.0%, 28.3%, and 10.0%, respectively. The manifestations mentioned above in D2 group were significantly lower than those in group H and group D1 (P<0.05). Immediately and 5 min after embolization, at the end of surgery and 0.5, 1, 4, 12 and 24 h after surgery, the VAS scores in the D2 group were 1.9±0.2, 2.1±0.3, 1.8±0.4, 1.8±0.3, 1.7±0.3, 1.6±0.3, 1.3±0.2, 1.3±0.3, respectively, lower than those in group H and group D1 (P<0.05); The satisfaction index in D2 group at these times were 8.7±1.1, 8.9±0.8, 9.2±0.9, 9.0±0.7, 9.1±0.8, 9.0±0.6, 9.1±0.7, 9.2±0.9, respectively, higher than those in group H and group D1 (P<0.05). No breath depression happened in these three groups. Conclusion: The formula of hydromorphone combined with dexmedetomidine to patients undergoing TACE is greatly safe and efficient, with advantages in alleviating pain, reducing hydromorphone consumption and the incidence of adverse reaction of hydromorphone, and without breath depression.