Objective To evaluate the application of bilateral thoracic paravertebral block (bTPVB) combined with general anesthesia in thoracoscopic‑laparoscopic esophagectomy (TLE) and its effects on early postoperative recovery. Methods Seventy patients with esophageal cancer scheduled for TLE were divided into two groups according to the random number table method (n=35): a general anesthesia group (group G) and a bTPVB combined with general anesthesia group (group B). Five patients were excluded due to intraoperative blood transfusion, traditional open surgery or other reasons. Finally, 32 patients were included into group G and 33 patients in group B, respectively. Standardized general anesthesia was performed in both groups. Paravertebral blocks at the right T7‒T8 level and the left T8‒T9 level were performed 15 min before anesthesia induction in group B. Then, perioperative hemodynamics, the length of anesthesia and surgery, the use of anesthetics and vasoactive agents during operation in both groups were recorded. The use of sufentanil in the post‑anesthetic care unit (PACU) and flurbiprofen axetil in the thoracic intensive care unit (ICU) for rescue analgesia were recorded. The eye‑opening time, extubation time, sedation‑agitation scale scores of PACU admission and discharge, the length of PACU stay, postoperative delirium, and pulmonary function 48 h after surgery were recorded. The Visual Analogue Scale (VAS) scores at resting and upon coughing and the cumulative doses of sufentanil were assessed 0, 4, 8, 12, 24, 36 h and 48 h after surgery. Results There was no statistical difference in hemodynamic parameters between the two groups at each time point (P>0.05). Compared with group G, the intraoperative doses of sufentanil, the length of PACU stay, the rate of intraoperative nitroglycerin use, and the rate of sufentanil use for rescue analgesia in PACU and the rate of flurbiprofen use for rescue analgesia in thoracic ICU significantly reduced in group B (P<0.05). Compared with pre‑operative levels, the forced vital capacity (FVC) and the forced expiratory volume in first second significantly decreased in both groups 48 h after surgery (P<0.05), where a more remarkable decrease was found in group G (P<0.05). Group G presented marked higher VAS scores 0, 4, 8 h and 12 h after surgery at resting and upon coughing, and 24 h after surgery upon coughing, compared with group B (P<0.05). The cumulative doses of sufentanil in group B were obviously lower than those in group G 4, 8, 12, 24, 36 h and 48 h after surgery (P<0.05). Conclusions The use of bTPVB combined with general anesthesia in TLE can effectively decrease the doses of analgesics during the perioperative period, alleviate early postoperative pain, and improve postoperative pulmonary function, so as to promote early postoperative recovery.