BACKGROUND: Declining platelet counts may reveal platelet activation and aggregation in a postoperative prothrombotic state. Therefore, we hypothesized that nadir platelet counts after on-pump coronary artery bypass grafting (CABG) surgery are associated with stroke. METHODS: We evaluated 6130 adult CABG surgery patients. Postoperative platelet counts were evaluated as continuous and categorical (mild versus moderate to severe) predictors of stroke. Extended Cox proportional hazard regression analysis with a time-varying covariate for daily minimum postoperative platelet count assessed the association of day-to-day variations in postoperative platelet count with time to stroke. Competing risks proportional hazard regression models examined associations between day-to-day variations in postoperative platelet counts with timing of stroke (early: 0-1 days; delayed: ≥2 days). RESULTS: Median (interquartile range) postoperative nadir platelet counts were 123.0 (98.0-155.0) × 10/L. The incidences of postoperative stroke were 1.09%, 1.50%, and 3.02% for platelet counts >150 × 10/L, 100 to 150 × 10/L, and <100 × 10/L, respectively. The risk for stroke increased by 12% on a given postoperative day for every 30 × 10/L decrease in platelet counts (adjusted hazard ratio [HR], 1.12; 95% confidence interval [CI], 1.01-1.24; P= .0255). On a given day, patients with moderate to severe thrombocytopenia were almost twice as likely to develop stroke (adjusted HR, 1.89; 95% CI, 1.13-3.16; P= .0155) as patients with nadir platelet counts >150 × 10/L. Importantly, such thrombocytopenia, defined as a time-varying covariate, was significantly associated with delayed (≥2 days after surgery; adjusted HR, 2.83; 95% CI, 1.48-5.41; P= .0017) but not early postoperative stroke. CONCLUSIONS: Our findings suggest an independent association between moderate to severe postoperative thrombocytopenia and postoperative stroke, and timing of stroke after CABG surgery.