NEW YORK (Reuters Health) – A lower than normal hematocrit level before percutaneous coronary intervention (PCI) and a drop in hematocrit after PCI are “strongly associated” with long-term mortality, the results of a retrospective study indicate.

“This analysis emphasizes the close relationship between these two factors and 1-year adverse outcome,” the study team notes in the December issue of the American Heart Journal.

The presence of anemia before PCI or the development of bleeding or anemia after PCI has been shown to increase mortality and morbidity, they explain. And while the prognostic value of these two established risk factors has been reported separately, the relationship between the two has been unclear.

To investigate, Dr. Gabriel Maluenda and colleagues from Washington Hospital Center, Washington, DC, analyzed 6,025 consecutive patients (65% male, mean age 65 years) who underwent PCI at their institution from 2003 to 2007.

They grouped the study subjects by hematocrit at baseline and the magnitude of hematocrit decline. Overall, the mean baseline hematocrit was 39.7% and the mean hematocrit drop was 3.5%.

According to the investigators, the rate of patients meeting the 1-year primary end point (a composite of death and MI) “increased continuously every time hematocrit at baseline decreased and/or hematocrit dropped after PCI.”

In a multivariable Cox regression model, both hematocrit at baseline and extent of hematocrit decline after PCI, examined as continuous variables, were significant predictors of the primary end point at 1 year.

The frequency of reaching the composite primary end point was inversely correlated with hematocrit levels at baseline and after PCI, the study team notes.

“For example, the higher the hematocrit at baseline, the less likely the occurrence of the composite primary end point at 1-year follow-up (hazard ratio, 0.92 for 1% unit increase, p < 0.001); the higher the magnitude of hematocrit drop, the higher the incidence of the composite primary end point (HR 1.11 for 1% unit decrease, p < 0.001)."

“Importantly,” according to Dr. Maluenda and colleagues, transfusion “did not seem to increase the risk of late death or MI once preprocedural anemia and magnitude of hematocrit drop were accounted for.”

The researchers suggest that a risk model be developed based on “the robust relationship between both baseline hematocrit and the hematocrit drop and…long-term mortality.”

Reference:
Am Heart J 2009;158:1024-1030.