NEW YORK (Reuters Health) – For patients with a hematocrit nadir of 24%-30% following percutaneous coronary intervention (PCI), the benefits of blood transfusion are unlikely to outweigh the risks, researchers say.
Antiplatelet and antithrombin treatment during and after PCI increase the risk of bleeding, red blood cell transfusion and mortality, Dr. Ron Waksman and colleagues at Washington Hospital Center in Washington, DC, remark in their paper in the American Journal of Cardiology for October 15.
However, they add, “it remains unclear whether transfusion is useful, neutral, or harmful when the hematocrit level after PCI is 24% to 30%.”
To study this issue, the research team analyzed data from a registry of catheter-based coronary procedures maintained at their institution. They identified 625 patients (mean age 71 years) with hematocrit levels in that range after PCI, including 189 who received a transfusion and 436 who did not.
Those given transfusion were more likely than the control group to present with acute MI and cardiogenic shock, and the mean hematocrit decrease following PCI and the rate of major bleeding were significantly greater in the transfused group.
At 30 days and 1 year, mortality rates were higher in the transfused group (15.3% vs 6.9% and 27.5% vs 18.5%, respectively). Rates of MI at both end-points were also significantly higher among those treated with red blood cells.
Nonetheless, after adjusting for baseline significant predictors and mean decrease in hematocrit, transfusion itself was no longer an independent predictor for death and nonfatal MI (hazard ratio at 30 days and 1 year = 1.4, p = NS).
“The potential benefit of transfusion must be balanced against the risk of infection, immune transfusion reactions, blood volume overexpansion, and afterload increases,” Dr. Waksman and his associates caution.
They conclude: “Our data do not support the routine use of transfusion in patients who present with a nadir hematocrit of 24% to 30% after PCI.”
Reference:
Am J Cardiol 2009;104:1069-1073.
