RESUMO
Objective: The pattern of digestive tract reconstruction in radical gastrectomy for gastric cancer is still inconclusive. This study aims to compare mid-term and long-term quality of life after radical gastrectomy for distal gastric cancer between Billroth-I (B-I) and Billroth-II (B-II) reconstruction. Methods: A retrospective cohort study was conducted.Clinicopathological and follow-up data of 859 gastric cancer patients were colected cellected from the surgical case registry database of Gastrointestinal Surgery Center of Sichuan University West China Hospital, who underwent radical distal gastric cancer resection between January 2016 and December 2020. Inclusion criteria: (1) gastric cancer confirmed by preoperative gastroscopy and biopsy; (2) elective radical distal major gastrectomy performed according to the Japanese Society for Gastric Cancer treatment guidelines for gastric cancer; (3) TNM staging referenced to the American Cancer Society 8th edition criteria and exclusion of patients with stage IV by postoperative pathology; (4) combined organ resection only involving the gallbladder or appendix; (5) gastrointestinal tract reconstruction modality of B-I or B-II; (6) complete clinicopathological data; (7) survivor during the last follow-up period from December 15, 2021 to January 15, 2022. Exclusion criteria: (1) poor compliance to follow-up; (2) incomplete information on questionnaire evaluation; (3) survivors with tumors; (4) concurrent malignancies in other systems; (5) concurrent psychiatric and neurological disorders that seriously affected the objectivity of the questionnaire or interfered with patient's cognition. Telephone follow-up was conducted by a single investigator from December 2021 to January 2022, and the standardized questionnaire EORTC QLQ-C30 scale (symptom domains, functional domains and general health status) and EORTC QLQ-STO22 scale (5 symptoms of dysphagia, pain, reflux, restricted eating, anxiety; 4 single items of dry mouth, taste, body image, hair loss) were applied to evaluate postoperative quality of life. In 859 patients, 271 were females and 588 were males; the median age was 57.0 (49.5, 66.0) years. The included cases were divided into the postoperative follow-up first year group (202 cases), the second year group (236 cases), the third year group (148 cases), the fourth year group (129 cases) and the fifth year group (144 cases) according to the number of years of postoperative follow-up. Each group was then divided into B-I reconstruction group and B-II reconstruction group according to procedure of digestive tract reconstruction. Except for T-stage in the fourth year group, and age, tumor T-stage and tumor TNM-stage in the fifth year group, whose differences were statistically significant between the B-I and B-II reconstruction groups (all P<0.05), the differences between the B-I and B-II reconstruction groups in terms of demographics, body mass index (BMI), tumor TNM-stage and tumor pathological grading in postoperative follow-up each year group were not statistically significant (all P>0.05), suggesting that the baseline information between B-I reconstruction group and the B-II reconstruction group in postoperative each year group was comparable. Evaluation indicators of quality of life (EORTC QLQ-C30 and EORTC QLQ-STO22 scales) and nutrition-related laboratory tests (serum hemoglobin, albumin, total protein, triglycerides) between the B-I reconstruction group and B-II reconstruction group in each year group were compared. Non-normally distributed continuous variables were presented as median (Q(1),Q(3)), and compared by using the Wilcoxon rank sum test (paired=False). The χ(2) test or Fisher's exact test was used for comparison of categorical variables between groups. Results: There were no statistically significant differences in all indexes EORTC QLQ-30 scale between the B-I reconstruction group and the B-II reconstruction group among all postoperative follow-up year groups (all P>0.05). The EORTC QLQ-STO22 scale showed that significant differences in pain and eating scores between the B-I reconstruction group and the B-II reconstruction group were found in the second year group, and significant differences in eating, body and hair loss scores between the B-I reconstruction group and the B-II reconstruction group were found in the third year group (all P<0.05), while no significant differences of other item scores between the B-I reconstruction group and the B-II reconstruction group were found in postoperative follow-up of all year groups (P>0.05). Triglyceride level was higher in the B-II reconstruction group than that in the B-I reconstruction group (W=2 060.5, P=0.038), and the proportion of patients with hyperlipidemia (triglycerides >1.85 mmol/L) was also higher in the B-II reconstruction group (19/168, 11.3%) than that in the B-I reconstruction group (0/34) (χ(2)=0.047, P=0.030) in the first year group with significant difference. Albumin level was lower in the B-II reconstruction group than that in the B-I reconstruction group (W=482.5, P=0.036), and the proportion of patients with hypoproteinemia (albumin <40 g/L) was also higher in the B-II reconstruction group (19/125, 15.2%) than that in the B-I reconstruction group (0/19) in the fifth year group, but the difference was not statistically significant (χ(2)=0.341, P=0.164). Other nutrition-related clinical laboratory tests were not statistically different between the B-I reconstruction and the B-II reconstruction in each year group (all P>0.05). Conclusions: The effects of both B-I and B-II reconstruction methods on postoperative mid-term and long-term quality of life are comparable. The choice of reconstruction method after radical resection of distal gastric cancer can be based on a combination of patients' condition, sugenos' eoperience and operational convenience.