Objective To investigate the effects of four different anesthesia methods on the perioperative immune function and recovery quality of patients undergoing laparoscopic radical resection of colorectal cancer. Methods A total of 120 patients who were scheduled for laparoscopic radical resection for colorectal cancer were enrolled. According to the random number table method, they were divided into four groups (n=30): a total intravenous anesthesia group (group N), a transversus abdominis plane block (TAPB) combined with general anesthesia group (group T), an epidural block (EB) combined with general anesthesia group (group S) and a TAPB combined with dexmedetomidine (Dex)‑assisted general anesthesia group (group TD). Group N was induced and maintained by total intravenous anesthesia. In group T, ultrasound‑guided TAPB was performed before induction of general anesthesia. In group S, EB was performed in intervertebral space between L2‑L3. In group TD, ultrasound‑guided bilateral TAPB was performed before induction of general anesthesia, with intravenous infusion of Dex at 0.3 μg·kg−1·h−1 during anesthesia maintenance until 30 min before operation. Then, their surgical and anesthesia times and intraoperative blood loss as well as the intraoperative dosages of propofol, remifentanil and vasoactive agents were recorded. The concentrations of T lymphocyte subsets (CD3+, CD4+ and CD8+) and natural killer (NK) cells were measured 1 day before surgery (T0), 1 day (T1), 3 days (T2) and 5 days (T3) after operation to calculate CD4+/CD8+. The Ramsay sedation scores at extubation, the highest scores of Visual Analogue Scale (VAS) pain, and the incidences of cough, chills, agitation, nausea and vomiting were recorded. Results Compared with those at T0, the concentrations of NK cells, CD3+, and CD4+, and CD4+/CD8+ in the four groups significantly decreased at T1 (P<0.05); the concentrations of NK cells, CD3+ and CD4+ in the four groups and CD4+/CD8+ in groups N and T significantly decreased at T2 (P<0.05). Compared with group T, the concentrations of NK cells, CD3+, CD4+ and CD4+/CD8+ in group N significantly decreased at T1 and T2 (P<0.05), while groups S and TD presented significant increases in the concentrations of NK cells, CD3+, CD4+ and CD4+/CD8+ (P<0.05). Compared with group N, the concentrations of NK cells, CD3+, CD4+ and CD4+/CD8+ in groups S and TD significantly increased at T1 and T2 (P<0.05). Compared with groups N and T, the dosages of propofol and remifentanil significantly reduced in groups S and TD (P<0.05). The usage of vasoactive agents in group S was more frequent than that in groups N, T and TD (P<0.05). Compared with group N, the highest scores of VAS pain in groups S, T and TD significantly decreased (P<0.05), the incidences of nausea and vomiting in groups S and TD decreased (P<0.05), the incidences of cough, chills, and agitation in group TD reduced (P<0.05) and the Ramsay sedation scores at extubation increased (P<0.05). Compared with group T, the incidences of nausea and vomiting in groups S and TD decreased (P<0.05), the incidence of cough, chills and agitation in group TD reduced, and the Ramsay sedation scores at extubation significantly increased (P<0.05). Compared with group S, the incidence of cough, chills and agitation in group TD reduced (P<0.05), but the Ramsay sedation scores at extubation increased (P<0.05). There was no significant difference in other indicators (P>0.05). Conclusions Compared with total intravenous anesthesia group, TAPB combined with general anesthesia group and EB combined with general anesthesia group, TAPB combined with Dex assisted general anesthesia group can mantain hemodynamic stability and effectively improve perioperative immune function and quality of emergence from general anesthesia in patients undergoing laparoscopic resection of colorectal cancer.