RESUMEN
PURPOSE: We developed a predictive model to assess the risk of major bleeding (MB) within 6 months of primary venous thromboembolism (VTE) in cancer patients receiving anticoagulant treatment. We also sought to describe the prevalence and incidence of VTE in cancer patients, and to describe clinical characteristics at baseline and bleeding events during follow-up in patients receiving anticoagulants. METHODS: This observational, retrospective, and multicenter study used natural language processing and machine learning (ML), to analyze unstructured clinical data from electronic health records from nine Spanish hospitals between 2014 and 2018. All adult cancer patients with VTE receiving anticoagulants were included. Both clinically- and ML-driven feature selection was performed to identify MB predictors. Logistic regression (LR), decision tree (DT), and random forest (RF) algorithms were used to train predictive models, which were validated in a hold-out dataset and compared to the previously developed CAT-BLEED score. RESULTS: Of the 2,893,108 cancer patients screened, in-hospital VTE prevalence was 5.8% and the annual incidence ranged from 2.7 to 3.9%. We identified 21,227 patients with active cancer and VTE receiving anticoagulants (53.9% men, median age of 70 years). MB events after VTE diagnosis occurred in 10.9% of patients within the first six months. MB predictors included: hemoglobin, metastasis, age, platelets, leukocytes, and serum creatinine. The LR, DT, and RF models had AUC-ROC (95% confidence interval) values of 0.60 (0.55, 0.65), 0.60 (0.55, 0.65), and 0.61 (0.56, 0.66), respectively. These models outperformed the CAT-BLEED score with values of 0.53 (0.48, 0.59). CONCLUSIONS: Our study shows encouraging results in identifying anticoagulated patients with cancer-associated VTE who are at high risk of MB.
RESUMEN
BACKGROUND: The clinical characteristics of patients with factor V Leiden or prothrombin G20210A presenting with a first episode of venous thromboembolism (VTE) have not been thoroughly studied. METHODS: RIETE is an ongoing registry of consecutive patients with acute VTE. We compared the clinical characteristics of patients with factor V Leiden, prothrombin G20210A, or no thrombophilia, at presentation with a first episode of VTE. RESULTS: As of May 2009, 22428 patients had been enrolled with a first episode of VTE. Of these, 345 had factor V Leiden, 261 had prothrombin G20210A, and 2399 tested negative. Sixty-two percent of the VTE episodes in women with factor V Leiden or prothrombin G20210A (40% in men) were associated with an acquired risk factor. Among women, pregnancy or contraceptive use accounted for 63% and 67% of such risk factors. Patients with factor V Leiden presented with pulmonary embolism (PE) less likely than those with prothrombin G20210A (31% vs. 51%; p<0.001) or with negative testing (31% vs. 45%, p<0.001). In addition, PE patients with Factor V Leiden presented with hypoxaemia (Sat O(2) levels<90%) less likely than those with prothrombin G20210A (4.5% vs. 17%; p<0.001) or with no thrombophilia (4.5% vs. 20%; p<0.001). CONCLUSIONS: Most VTE episodes in women (not men) with factor V Leiden or prothrombin G20210A were associated with an acquired risk factor (mostly pregnancy or contraceptive use). Only 4.5% of patients with factor V Leiden presenting with acute PE had hypoxaemia.
Asunto(s)
Factor V/genética , Mutación Puntual , Protrombina/genética , Tromboembolia Venosa/genética , Tromboembolia Venosa/fisiopatología , Adulto , Anciano , Anciano de 80 o más Años , Femenino , Heterocigoto , Humanos , Masculino , Persona de Mediana Edad , Factores de Riesgo , Tromboembolia Venosa/epidemiología , Adulto JovenRESUMEN
A significant association between elevated white blood cell (WBC) count and mortality in patients with cancer has been reported, but the predictive value of elevated WBC on mortality in cancer patients with acute venous thromboembolism (VTE) has not been explored. RIETE is an ongoing registry of consecutive patients with acute VTE. We compared the three-month outcome of cancer patients with acute VTE according to their WBC count at baseline. As of May 2007, 3805 patients with active cancer and acute VTE had been enrolled in RIETE. Of them, 215 (5.7%) had low- (<4,000 cells/microl), 2,403 (63%) normal- (4,000-11,000 cells/microl), 1,187 (31%) elevated (>11,000 cells/microl) WBC count. During the study period 190 patients (5.0%) had recurrent VTE, 156 (4.1%) major bleeding, 889 (23%) died (399 of disseminated cancer, 113 of PE, 46 of bleeding. Patients with elevated WBC count at baseline had an increased incidence of recurrent VTE (odds ratio [OR]: 1.6; 95% confidence interval [CI]: 1.2-2.2), major bleeding (OR: 1.5; 95% CI: 1.1-2.1) or death (OR: 2.7; 95% CI: 2.3-3.2). Most of the reported causes of death were significantly more frequent in patients with elevated WBC count. Multivariate analysis confirmed that elevated WBC count was independently associated with an increased incidence of all three complications. In conclusion, cancer patients with acute VTE and elevated WBC count had an increased incidence of VTE recurrences, major bleeding or death. This worse outcome was consistent among all subgroups and persisted after multivariate adjustment.