Whether extended lymphadenectomy for right colon cancer leads to increased perioperative complications or improves survival is still controversial. This trial aimed to compare the efficacy and safety of complete mesocolic excision (CME) versus D2 dissection in laparoscopic right hemicolectomy for patients with right colon cancer. This article reports the early safety results from the trial. This randomised, controlled, phase 3, superiority, trial was done at 17 hospitals in nine provinces of China. Eligible patients were aged 18-75 years with histologically confirmed primary adenocarcinoma located between the caecum and the right third of the transverse colon, without evidence of distant metastases. Central randomisation was done by means of the Clinical Information Management-Central Randomisation System via block randomisation (block size of four). Patients were randomly assigned (1:1) to CME or D2 dissection during laparoscopic right colectomy. Central lymph nodes were dissected in the CME but not in the D2 procedure. Neither investigators nor patients were masked to their group assignment but the quality control committee were masked to group assignment. The primary endpoint was 3-year disease-free survival, but the data for this endpoint are not yet mature; thus, only the secondary outcomes-intraoperative surgical complications and postoperative complications within 30 days of surgery, graded according to the Clavien-Dindo classification, mortality (death from any cause within 30 days of surgery), and central lymph node metastasis rate in the CME group only-are reported in this Article. This early analysis of safety was preplanned. The outcomes were analysed according to a modified intention-to-treat principle (excluding patients who no longer met inclusion criteria after surgery or who did not have surgery). This study is registered with ClinicalTrials.gov, NCT02619942. Study recruitment is complete, and follow-up is ongoing. Between Jan 11, 2016, and Dec 26, 2019, 1072 patients were enrolled and randomly assigned. After exclusion of 77 patients, 995 patients were included in the modified intention-to-treat population (495 in the CME group and 500 in the D2 dissection group). The postoperative surgical complication rate was 20% (97 of 495 patients) in the CME group versus 22% (109 of 500 patients) in the D2 group (difference, -2·2% [95% CI -7·2 to 2·8]; p=0·39); the frequency of Clavien-Dindo grade I-II complications were similar between groups (91 [18%] vs 92 [18%], difference, -0·0% [95% CI -4·8 to 4·8]; p=1·0) but Clavien-Dindo grade III-IV complications were significantly less frequent in the CME group than in the D2 group (six [1%] vs 17 [3%], -2·2% [-4·1 to -0·3]; p=0·022); no deaths occurred in either group. Of the intraoperative complications, vascular injury was significantly more common in the CME group than in the D2 group (15 [3%] vs six [1%], difference, 1·8 [95% CI 0·04 to 3·6]; p=0·045). Metastases in the central lymph nodes were detected in 13 (3%) of 394 patients who underwent central lymph node biopsy in the CME group; no patient had isolated metastases to central lymph nodes. Although the CME procedure might increase the risk of intraoperative vascular injury, it generally seems to be safe and feasible for experienced surgeons. The Capital Characteristic Clinical Project of Beijing and the Chinese Academy of Medical Sciences. Copyright © 2021 Elsevier Ltd. All rights reserved.