Predicting the prognosis of operable gastric cancer patients by dynamic changes in platelets before and after surgery: a retrospective cohort study.

Abstract

Since the relationship between postoperative platelet count and prognosis is still unclear, we designed a standardized index of platelet count to predict the prognosis of gastric cancer (GC). We designed a development validation cohort for the pre/post platelet ratio. We determined the ability of PPR to predict mortality in gastric cancer patients and validated them by a separate cohort. Survival was assessed by Kaplan-Meier analysis and associations explored by multivariate and multivariate analyses. The usefulness of the prediction was estimated by measuring the time-dependent ROC. Decision-curve analysis was used to validate the net clinical benefit. The sample distribution was similar in the two cohorts, and the 1-, 3-, and 5-year OS evaluation of the postoperative/preoperative platelet ratio was the largest for AUC in the two cohorts. Meanwhile, PPR has a good predictive value and a net clinical benefit. PPR has been identified and validated to be independently concerned about OS of patients with GC and was a reliable and economic indicator to evaluate the prognosis.