SHISA 1601 data
Description
Background Venous thromboembolism (VTE) is a potentially fatal perioperative complication. This study aimed to clarify the frequency and risk factors of intercurrent venous thromboembolism (VTE) in patients undergoing major curative gastric cancer surgery. Methods This prospective multicenter observational study included patients with gastric cancer who underwent radical gastrectomy at 5 hospitals between June 2016 and May 2018. Patients who were preoperatively administered anticoagulants were excluded. Results A total of 126 patients were eligible to participate. VTE occurred within 9 days postoperatively in 5 cases (4.0 %; 2 symptomatic and 3 asymptomatic). Postoperative day (POD) 1 plasma D-dimer and soluble fibrin (SF) levels were significantly higher in the VTE group than in the non-VTE group. Receiver operating characteristic curve analysis indicated a statistically significant ability of POD 1 D-dimer and SF levels to predict postoperative VTE development after gastrectomy; this finding was reflected by areas under the curve of 0.97 (95% confidence interval [CI], 0.92–1.0) and 0.87 (95% CI, 0.74–1.0). Cutoff values of D-dimer (24.6 µg/mL) and SF (64.1 µg/mL) were determined. Intraoperative blood transfusion (odds ratio [OR], 7.86), POD 1 D-dimer ≥24.6 µg/mL (OR, 17.35), and POD 1 SF ≥64.1 µg/mL (OR, 19.5) were independent risk factors for postoperative VTE (p < 0.05). Conclusion VTE occurred in 4.0% patients after gastric cancer surgery; however, with an early diagnosis and anticoagulant therapy, no patients experienced progression. Careful observation of patients with high risk factors for VTE, including intraoperative blood transfusion and high POD 1 D-dimer or SF levels, would contribute to the early detection of VTE.
Files
Steps to reproduce
The statistical analyses were performed by Fisher’s exact test or the Mann-Whitney U-test as appropriate. Statistical calculations were performed using Statistical Package for the Social Sciences version 22.0 for MAC (IBM Institute, Armonk, NY, USA) and GraphPad Prism version 8.0 (GraphPad Software, San Diego, CA, USA). For all analyses, differences were considered statistically significant at value of p < 0.05. Using these parameters, receiver operating characteristic (ROC) curves for D-dimer were constructed and quantified via area under the curve (AUC) analysis as well as the standard error of the mean and corresponding 95% confidence interval (CI). The Youden Index was calculated as sensitivity + specificity - 1. Logistic regression models were used for the multivariate analyses to identify risk factors associated with development of postoperative VTE. Two-sided p values < 0.05 were considered statistically significant.