Platelet count as a predictor of thrombotic events following left atrial appendage occlusion: the proof is in the platelet?

FUNDING ACKNOWLEDGEMENTS: Type of funding sources: None. BACKGROUND: Patients undergoing left atrial appendage occlusion (LAAO) often are at increased risk for bleeding or thromboembolic events, but no specific risk score for these events in patients after LAAO exists. Concurrently, biomarkers are of growing importance in risk stratification for AF-patients. Therefore, hemostatic biomarkers may have added value in predicting what patients are prone to develop thrombotic and bleeding events and aid in personalizing antithrombotic treatment strategy following LAAO. METHODS: Six implanting centers retrospectively gathered data on pre-procedural hemostasis markers (i.e. platelet count (PLC), mean platelet volume (MPV), and fibrinogen). Prespecified composite endpoints on thrombotic events (i.e. stroke, transient ischemic attack, systemic embolism and device-related thrombus) and bleeding (major, minor and intracranial hemorrhage) were collected. RESULTS: In total, 1315 patients were included (74±9 years, 64% male, 72% Watchman 2.5/FLX, 25% Amplatzer Cardiac Plug/AMULET, 4% other device). Patients were at increased risk for thrombotic and bleeding events with a CHA2DS2-VASc score of 4.3±1.5 and a HAS-BLED score of 3.3±1.1. Baseline data on PLC, MPV, and fibrinogen was available in 92%, 50%, and 21%, respectively. Over a total follow-up duration of 2682 patient years, 111 thrombotic events and 140 bleeding events occurred. Univariate Cox regression analysis showed increased thrombotic risk in patient with higher PLC (HR 1.16 [1.03-1.31] per 50*109 increment), while no statistically significant association with MPV was found (Table 1). Fibrinogen tended to be higher both in patients developing a thrombotic event (HR 1.30 [0.94-1.78] per fL increment) and bleeding event (HR 1.23 [0.98-1.54]). No association of MPV or PLC with bleeding was observed. An optimal PLC cut-off point of 200*109 was identified that was able to decently discriminate between patients developing a thrombotic event and those who did not (Figure 1). CONCLUSION: PLC may be of added value in thrombotic risk stratification following LAAO. Validation of this biomarker by prospective studies and within multivariate prediction models is warranted. [Figure: see text] [Figure: see text]