Data on anti-platelet therapy (APT) for prevention of atherothrombotic events in thrombocytopaenic cancer patients is lacking. We aimed to identify patient and physician characteristics associated with APT management in thrombocytopaenic patients with haematological malignancy. A clinical vignette-based experiment was designed. Eleven haematologists were interviewed, identifying five variable categories. Next, 18 hypothetical vignettes were generated. Each physician received three vignettes and chose to: hold all APT; continue APT without platelet transfusion support; or continue APT with platelet transfusion support. The survey was distributed to haematologists and thrombosis specialists in three countries. Multivariate cluster robust Poisson regression models were used to calculate relative risks (RRs) of using one management option (over the other) for each variable in comparison to a reference variable. A total of 145 physicians answered 434 cases. Clinicians were more likely to hold APT in case of 20,000/µL platelets (vs. 40,000/µL; RR for continuing: 0.82 [95% confidence interval: 0.75-0.91]), recent major gastrointestinal bleeding (vs. none; RR 0.81 [0.72-0.92]) and when the physician worked at a university-affiliated community hospital (vs. non-academic community hospital; RR 0.84 [0.72-0.98]). Clinicians were more likely to continue APT in ST elevation myocardial infarction with dual APT (vs. unstable angina with single APT; RR 1.31 [1.18-1.45]) and when there were institutional protocols guiding management (vs. none; RR 1.15 [1.03-1.27]). When APT was continued, increased platelet transfusion targets were used in 34%. In summary, the decision process is complex and affected by multiple patient and physician characteristics. Platelet transfusions were frequently chosen to support APT, although no evidence supports this practice.
Georg Thieme Verlag KG Stuttgart · New York.