Well-designed retrospective study versus small-sample prospective study in research based on laparoscopic and open radical distal gastrectomy for advanced gastric cancer.

Abstract

Well-designed retrospective studies (RSs) and small-sample prospective studies (PSs) evaluating the efficacy of interventions have received much attention. This study was designed to evaluate the differences between well-designed RSs and small-sample randomized controlled trials based on the efficacy of laparoscopic distal gastrectomy (LDG) and open distal gastrectomy (ODG) for advanced gastric cancer (GC). The clinicopathological data of 1360 patients with GC who underwent DG were analysed. After propensity score matching (1:1), 380 cases (ODG = 190, LDG = 190) were finally selected in a RS. Meanwhile, data from 120 patients (ODG = 60, LDG = 60) who enrolled in a PS were analysed. In the PS, the LDG group had less intraoperative blood loss, shorter time to first flatus, and shorter time to fluid diet than the ODG group. In the RS, the LDG group had less intraoperative blood loss, and a shorter postoperative hospital stay than the ODG group. In the PS, the 3-year overall survival (OS) rate was 83.3% in the LDG group and 83.2% in the ODG group (p = 0.877). In the RS, the 3-year OS rate was 68.7% in the LDG group and 66.6% in the ODG group (p = 0.752). No significant interactions were observed between the two groups and any of the variables examined, either in the PS or RS. The recurrence patterns were similar in the two groups. Furthermore, Cox regression analysis showed that surgical method (LDG/ODG) was not a prognostic factor affecting OS or DFS, either prospectively or retrospectively. The oncologic efficacy of laparoscopic and open distal gastrectomy for advanced GC is comparable. Well-designed RSs can be similar to small sample of PSs in assessing long-term oncologic outcomes of surgical interventions, but the short-term outcomes obtained should be treated with caution.