Low On-Treatment Platelet Reactivity Predicts Long-Term Risk of Bleeding After Elective PCI.

BACKGROUND: Bleeding after percutaneous coronary interventions (PCI) is an important complication with impact on prognosis. AIM: To evaluate the predictive value of enhanced platelet responsiveness to dual antiplatelet therapy with aspirin and clopidogrel, for bleeding, after elective PCI. METHODS AND RESULTS: We performed multiple electrode aggregometry (MAE) platelet functional tests induced by arachidonic acid (ASPI) and adenosine-diphosphate (ADP) before PCI, and 24 hours after PCI, in 481 elective PCI patients who were followed-up for an average of 15.34 ± 7.19 months. Primary end point was the occurrence of any bleeding, while ischemic major adverse cardiovascular event (MACE) was a secondary endpoint. The incidence of total, BARC ≤ 2, and BARC ≥ 3 bleeding, according to BARC classification, was 19, 18, and 1%, respectively. Groups with any, and BARC ≤ 2 bleeding, had a lower average value of MAE ADP test after 24 hours, compared to the group without bleeding: 45.30 ± 18.63 U versus 50.99 ± 19.01 U; P = 0.005; and 45.75 ± 18.96 U versus 50.99 ± 18.99 U; P = 0.01; respectively. Female gender (HR 2.11; CI 1.37-3.25; P = 0.001), previous myocardial infarction (HR 0.56; CI 0.37-0.85; P = 0.006), lower body mass (HR 0.78; CI 0.62-0.98; P = 0.03), and MAE ADP test after 24 hours (HR 0.75; CI 0.61-0.93; P = 0.009) were the independent predictors for any bleeding by Cox univariate analysis. After adjustment, MAE ADP test after 24 hours, was the only independent predictor for any (HR 0.7; CI 0.56-0.87; P = 0.002), and BARC ≤ 2 (HR 0.71; CI 0.56-0.89; P = 0.003) bleeding, by Cox multivariate analysis. CONCLUSION: MAE ADP test before and after PCI, was associated with any, and BARC ≤ 2 bleeding after elective PCI.

Developing a machine learning model for bleeding prediction in patients with cancer-associated thrombosis receiving anticoagulation therapy.

BACKGROUND: Only 1 conventional score is available for assessing bleeding risk in patients with cancer-associated thrombosis (CAT): the CAT-BLEED score. OBJECTIVES: Our aim was to develop a machine learning-based risk assessment model for predicting bleeding in CAT and to evaluate its predictive performance in comparison to that of the CAT-BLEED score. METHODS: We collected 488 attributes (clinical data, biochemistry, and International Classification of Diseases, 10th Revision, diagnosis) in 1080 unique patients with CAT. We compared CAT-BLEED score, Ridge and Lasso logistic regression, random forest, and Extreme Gradient Boosting (XGBoost) algorithms for predicting major bleeding or clinically relevant nonmajor bleeding occurring 1 to 90 days, 1 to 365 days, and 90 to 455 days after venous thromboembolism (VTE). RESULTS: The predictive performances of Lasso logistic regression, random forest, and XGBoost were higher than that of the CAT-BLEED score in the prediction of bleeding occurring 1 to 90 days and 1 to 365 days after VTE. For predicting major bleeding or clinically relevant nonmajor bleeding 1 to 90 days after VTE, the CAT-BLEED score achieved a mean area under the receiver operating characteristic curve (AUROC) of 0.48 ± 0.13, while Lasso logistic regression and XGBoost both achieved AUROCs of 0.64 ± 0.12. For predicting bleeding 1 to 365 days after VTE, the CAT-BLEED score achieved a mean AUROC of 0.47 ± 0.08, while Lasso logistic regression and XGBoost achieved AUROCs of 0.64 ± 0.08 and 0.59 ± 0.08, respectively. CONCLUSION: This is the first machine learning-based risk model for bleeding prediction in patients with CAT receiving anticoagulation therapy. Its predictive performance was higher than that of the conventional CAT-BLEED score. With further development, this novel algorithm might enable clinicians to perform personalized anticoagulation strategies with improved clinical outcomes.