Higher blood hematocrit predicts hypertension in men

Sae Young Jae, Sudhir Kurl, Jari A. Laukkanen, Kevin S. Heffernan, Jina Choo, Yoon Ho Choi, Jeong Bae Park

Research output: Contribution to journalArticlepeer-review

36 Scopus citations


Objective: Little is known on the relationship between blood hematocrit level as an estimate of whole blood viscosity and the risk of hypertension. We investigated the relationship of higher hematocrit level with incident hypertension in men during a mean follow-up of 5 years. Methods: Participants were 3620 men (mean age 48 ± 6 years) who participated in two health examinations during 1998-2009. All participants were free of hypertension and within the normal range of hematocrit (39-50%) at baseline. The incidence of hypertension was defined as blood pressure values exceeding 140/90 mmHg of SBP and/or DBP at time of follow-up examination. Results: During 17 856 person-years of follow-up, 351 new cases of hypertension developed. The relative risk (RR) of incident hypertension in men with hematocrit in the highest tertile vs. men with hematocrit in the lowest tertile was 1.45 (95% confidence interval, CI 1.11-1.90) after adjustment for confounding factors. Every 1% increment in hematocrit was associated with 7% (RR 1.07, 95% CI 1.02-1.13) higher incidence of hypertension after adjusting for risk factors. Men whose hematocrit increased over time had a 25% increased risk of developing hypertension (RR 1.25, 95% CI 1.01-1.57) compared with men with decreased hematocrit after adjustment for risk factors. Conclusion: These results demonstrate that higher hematocrit levels, even within the normal range, were associated with the incidence of hypertension independent of other risk factors in men.

Original languageEnglish
Pages (from-to)245-250
Number of pages6
JournalJournal of Hypertension
Issue number2
StatePublished - Feb 2014


  • epidemiology
  • hematocrit
  • hypertension


Dive into the research topics of 'Higher blood hematocrit predicts hypertension in men'. Together they form a unique fingerprint.

Cite this