CHICAGO – A high-normal hematocrit was associated with an increased risk of new-onset heart failure in a Framingham Heart Study analysis.
"To our knowledge, this is the only study to show such a relationship in men and women in middle age. ... Our results should prompt consideration of a cautious and measured approach to the aggressive treatment of low hematocrit in a variety of disease states," Dr. Erin E. Coglianese said at the annual scientific sessions of the American Heart Association.
The mechanism by which a hematocrit (HCT) within normal range is linked to heart failure is unclear. However, animal studies suggest one possibility – that a high-normal HCT could impair vasodilation as a result of scavenging of nitric oxide by hemoglobin, according to Dr. Coglianese of Massachusetts General Hospital, Boston.
To explore the relationship between HCT and the risk of heart failure, she and her coinvestigators turned to the Framingham Heart Study. They documented a strong, graded relationship between HCT level and the risk of developing heart failure in 3,523 Framingham participants aged 50-65 who were free of a history of heart failure at baseline and were followed prospectively for up to 20 years.
Indeed, individuals with a high-normal baseline HCT had almost double the risk of new-onset heart failure during follow-up, compared with those with a low HCT, even after adjustment for conventional risk factors for heart failure.
A low HCT was defined as 39% to less than 44% in men and 36% to less than 40% in women. Men with an HCT of 44% to less than 46% and women with a level of 40% to less than 42% were deemed as having a low-normal level. A normal HCT was defined as 46% to less than 50% in men and 42% to less than 46% in women. And a high-normal HCT was one greater than 50% in men or 46% in women.
When these definitions were used, the incidence of new-onset heart failure was 25/10,000 person-years in individuals with a low HCT level, 31/10,000 with a low-normal HCT, 38/10,000 with a normal HCT, and 48/10,000 in Framingham participants with high-normal HCT.
In a multivariate logistic regression analysis, the risk of new-onset heart failure, compared with the risk in those with a low HCT, was 27% greater in those with a low-normal HCT, 47% greater in those with a normal HCT, and 78% greater in those with a high-normal level. The analysis was adjusted for age, sex, total cholesterol, hypertension, body mass index, left ventricular hypertrophy, pack-years of smoking, and physical activity.
The big limitation of this study is that the original Framingham cohort, included in this analysis, looks in some ways quite different from contemporary patient populations. Specifically, roughly half of the men in the original cohort were smokers, Dr. Coglianese noted.
In contrast to these new findings regarding HCT and risk of new-onset heart failure, numerous studies have shown that in patients who already have heart failure, a low HCT is associated with an increased risk of heart failure hospitalization as well as all-cause mortality. It remains unclear, however, whether this increased risk of poor outcomes is due to pathophysiologic changes induced by low HCT, or if a low HCT is merely a marker of greater disease severity, she said.
Dr. Coglianese said he had no relevant financial disclosures.