Objective To observe the effect of erector spinae plane block (ESPB) and paravertebral nerve block (PVB) on postoperative pain in patients of esophageal cancer undergoing endoscopically assisted esophagectomy. Methods One hundred and thirty‑eight patients undergoing elective endoscopic assisted esophagectomy were divided into 3 groups according to the random number table method (n=46): erector spinae plane block group (group E), paravertebral nerve block group (group P) and control group (group C). Patients in group P received PVB at the T5‑T6 level before anesthesia induction. Patients in group E received ESPB at the T5 level respectively. Patients in group C only received general anesthesia. Consumption of propofol and remifentanil during operation, Visual Analogue Scale (VAS) score of resting and coughing at 4 hours (T1), 1 day (D1), 2 days (D2), 3 days (D3) after the operation, the time of the first self‑control analgesia and the first‑time ambulation, the frequency of self‑control analgesia and times of pentazocine given were assessed. Incidence of adverse reactions at 48 h after the operation was recorded too. Chronic post‑surgical pain (CPSP) was evaluated by telephone follow‑up 3 months after surgery. Results Propofol and remifentanil consumption of group P and group E were less than the consumption in group C (P<0.05), The dosage of remifentanil in group P was less than the dosage in group E (P<0.05). At T1 and D1, the resting VAS score of group C was higher than that of group E and P (P<0.05), and the cough VAS score was higher than that of group E and P (P<0.05); at D1, the resting VAS score of group P was lower than the score in group E (P<0.05), and the cough VAS score of the group P was lower than that of the group E at T1 and D1 (P<0.05). First time and total times of PCA of group E and group P were later or less than that of group C, usage of pentazocine was also reduced (P<0.05). The first ambulation in group C was longer than the ambulation of group E and group P, and group E was longer than that of group P (P<0.05). The incidence of chronic postoperative pain in group C was higher than the incidence in group P (P<0.05). There was no significant difference in other indicators among the groups (P>0.05). Conclusions ESPB and PVB can relieve postoperative pain in patients undergoing endoscopically assisted esophagectomy, PVB has a better analgesic effect than ESPB, and PVB may reduce the incidence of CPSP at 3 months after surgery.