Background: Peritoneal dissemination is the most frequent mode of recurrence in patients with gastric cancer. We tried to identify factors that predict peritoneal recurrence with high sensitivity.
Study design: Clinical and pathologic data from 587 consecutive patients with gastric cancer were reviewed retrospectively. The stepwise Cox proportional hazards regression model was used to assess the prognostic significance of the magnitude of serosal changes. Multiple stepwise logistic regression analysis was used to determine factors associated with peritoneal recurrence in 375 patients who underwent curative resection.
Results: The 5-year survival rate of patients with S2 disease (greatest dimension of macroscopic serosal changes >/= 2.5 cm) was 18%, which was worse than S0 (no serosal changes) and S1 disease (macroscopic serosal changes < 2.5 cm)(p < 0.001). Patients with S0 tumors who underwent curative resection had the best 5-year survival rate. Multivariate analyses indicated that the magnitude of serosal changes was an independent prognostic factor for survival both overall and after curative resection. Logistic regression analysis showed that peritoneal recurrence was more than four times as likely with S2 than with S0 or S1 tumors. The sensitivity for predicting peritoneal recurrence was 79%; the sensitivity of cytologic examination was 38%.
Conclusions: Magnitude of serosal changes is easy to measure intraoperatively and predicts peritoneal recurrence of gastric cancer with greater sensitivity than conventional peritoneal lavage cytology.