Objective　To compare the effects of combined paravertebral nerve block and general anesthesia on hemodynamics and recovery period in thoracic surgery.　Methods　Eighty patients undergoing elective thoracic surgery with American Society of Anesthesiologists(ASA) Ⅱ or Ⅲ were randomly divided into 2 groups with 40 cases in each group. Paravertebral nerve block combined general anesthesia group (TPVB group) and traditional general anesthesia control group (C group). Two groups of patients were induced with the same procedure. Sevoflurane inhalation, propofol, and remifentanil were administered intravenously to maintain anesthesia. Hemodynamic parameters before surgery (T0), after induction (T1), 3 min after the operation (T2), after the end of surgery (T3) and the number of interventions requiring vasoactive drugs and dosage of analgesic drugs were measured. The time of anesthesia, operation time, and recovery time, extubation time and pain score were recorded. The recovery of consciousness was assessed by using Observer′s Assessment of Alertness/Sedation Scale (OAA/S) score.　Results　Systolic blood pressure (SBP), diastolic blood pressure (DBP) and heart rate were significantly lower in TPVB group than in group C at the T2 timepoint. SBP and DBP were significantly higher, and heart rate was significantly lower in TPVB group than in group C at the T3 timepoint (P<0.05). The number of 4 vasoactive drugs and remifentanil dosage were significantly less in TPVB group than in group C (P<0.05). The time of blinking, extubating time and pain score of TPVB group were significantly better than those of group C (P<0.05). Postoperative opening the eyes by prompting time, extubating time, and pains cores were significantly better in TPVB group than in group C (P<0.05). The OAA/S scores before and after extubating were significantly higher in TPVB group than in group C (P<0.05).　Conclusions　Combination of paravertebral nerve block with general anesthesia leads to more stable hemodynamics, earlier time for extubation, and less postoperative pain scores in patients undergoing thoracic surgery.