Background Association of baseline hemoglobin levels with long-term adverse events after percutaneous coronary interventions has not been yet thoroughly defined. We aimed to assess the clinical impact of baseline hemoglobin on long-term ischemic and bleeding risk after percutaneous coronary intervention. Methods and Results Using the pooled individual patient-level data from the 3 percutaneous coronary intervention studies, we categorized 19 288 patients into 4 groups: high-normal hemoglobin (≥14.0 g/dL; n=7555), low-normal hemoglobin (13.0-13.9 g/dL in men and 12.0-13.9 g/dL in women; n=5303), mild anemia (11.0-12.9 g/dL in men and 11.0-11.9 g/dL in women; n=4117), and moderate/severe anemia (<11.0 g/dL; n=2313). Median follow-up duration was 3 years. Low-normal hemoglobin, mild anemia, and moderate/severe anemia correlated with significant excess risk relative to high-normal hemoglobin for GUSTO (Global Utilization of Streptokinase and Tissue Plasminogen Activator for Occluded Arteries Trial) moderate/severe bleeding, with adjusted hazard ratios of 1.22 (95% CI, 1.04-1.44), 1.73 (95% CI, 1.47-2.04), and 2.31 (95% CI, 1.92-2.78), respectively. Moderate/severe anemia also correlated with significant excess risk relative to high-normal hemoglobin for the ischemic composite end point of myocardial infarction/ischemic stroke (adjusted hazard ratio, 1.33; 95% CI, 1.11-1.60), whereas low-normal hemoglobin and mild anemia did not. However, the excess risk of low-normal hemoglobin, mild anemia, and moderate/severe anemia relative to high-normal hemoglobin remained significant for ischemic stroke and for mortality. Conclusions Decreasing baseline hemoglobin correlated with incrementally higher long-term risk for major bleeding, ischemic stroke, and mortality after percutaneous coronary intervention. Even within normal range, lower baseline hemoglobin level correlated with higher ischemic and bleeding risk.
Keywords: anemia; bleeding; hemoglobin; ischemia; percutaneous coronary interventions.