Hematocrit is the most powerful determinant of blood viscosity and is determined by plasma volume and red cell mass. Blood viscosity is under homeostatic control via regulation of hematocrit. The regulatory pathway is as follows: increased blood viscosity increases total peripheral resistance (that is, systemic vascular resistance) and left ventricular end diastolic volume, which is sensed by stretch receptors. Their signal upregulates cardiomyocyte production of the cardiac natriuretic peptides, atrial and brain natriuretic peptide, and soluble erythropoietin receptor.
Cardiac natriuretic peptides cause vasodilation and diuresis, which decreases plasma volume, and soluble erythropoietin receptor intercepts erythropoietin before it acts in the bone marrow, decreasing red cell mass. The result is decreased total peripheral resistance and reduced blood pressure. A manifestation of this homeostatic pathway is anemia of chronic disease, in which hematocrit is decreased to maintain normal blood viscosity.
Increased blood viscosity is directly linked to primary (essential) hypertension. Multiple causes of hyperviscosity have been identified: increased erythrocyte aggregation in hypercholesterolemia, increased plasma immunoglobulins in multiple myeloma, and increased red cell mass in polycythemia vera. None of these are obvious factors in hypertension. Instead, the hyperviscosity seen in hypertension is caused by decreased plasma volume combined with normal red cell mass, resulting in increased hematocrit.1 From an insightful study led by Stevo Julius, a widely recognized leader in hypertension research from Archives of Internal Medicine, 1971:
Patients with borderline hypertension have decreased plasma volume. This is a reproducible finding. Decreased plasma volume is associated with increases of the hematocrit reading. …the decreased plasma volume was correlated with increases of the blood pressure and of the total peripheral resistance. Since the decrease of plasma volume clearly relates indices of “severity” of borderline hypertension (blood pressure and resistance), measurement of plasma volume may be important for understanding the natural history of the disease and for the assessment of an individual’s risk.2
Decreased plasma volume and normal red cell mass are not restricted to borderline hypertension, but present throughout the entire spectrum of disease severity.3-5 Early in the course of hypertension, the effect of increased total peripheral resistance is partially mitigated by the cushioning effect of aortic elasticity. However, with progressive fatigue and fracture of the elastic elements in the aorta, aortic stiffness worsens blood pressure and end organ damage.
Hemodynamic Homeostasis in Hypertension
This theory implies that homeostasis of blood viscosity requires coordinated changes of plasma volume and red cell mass, and that primary hypertension is the result of loss of homeostasis of red cell mass, not plasma volume. Plasma volume is appropriately decreased in response to increased left ventricular end diastolic volume, while red cell mass is unchanged. The result is increased hematocrit, increased blood viscosity, higher total peripheral resistance, and hypertension. Even small increases of hematocrit are rheologically significant because linear increases of hematocrit over this range cause exponential increases in blood viscosity.6
So hypertension occurs despite adequate control of cardiac natriuretic peptide activity. Supporting this is the finding that treatment of hypertension reduces cardiac natriuretic peptide levels.7 As suggested in a recent review, this decrease could be due to “a better hemodynamic profile characterized by a decrease in vascular resistances leading to decreased left ventricular afterload and circumferential wall stress.”8
Increased blood viscosity reduces erythropoietin production.9,10 The most elegant demonstration of this was reported in 1969 by a team of hematologists at the University of Illinois College of Medicine. These investigators produced hyperviscosity in mice by transfusing them with both human and mice erythrocytes and then placing them in hypoxic conditions. Hyperviscosity decreased erythropoietin production despite hypoxia. In a second experiment, mice made hyperviscous by dehydration were incubated in hypoxic conditions, and again, erythropoietin production was decreased. Rehydration restored the ability to produce erythropoietin following hypoxia.11
As the authors noted, “increased viscosity, without compensatory cardiovascular response, would result in a decreased delivery of oxygen to tissue. If this premise is valid, polycythemia states should be self-perpetuating.”11 In other words, if there was no homeostatic control of blood viscosity, tissue hypoxia would increase erythropoietin production, which would increase red cell mass and blood viscosity, which, because viscosity is inversely proportional to flow, would worsen hypoxia and lead to further increases in erythropoietin production and viscosity in a vicious cycle. Homeostatic control of blood viscosity prevents uncontrolled upregulation of erythropoietin expression. Unremitting positive feedback is incompatible with life.
Erythropoeisis and Essential Hypertension
I submit that defective homeostasis of viscosity and uncontrolled erythropoietin expression are major drivers in the pathophysiology of primary hypertension. A defect in a silencer gene is a possibility. However, defects may involve other steps in the regulatory cascade, including paracrine signaling, which normally function to increase the number of cardiomyocytes expressing soluble erythropoietin receptor, or the primary structure, expression, or post-translational modification of soluble erythropoietin receptor.
Genome-wide association studies have found 35 different single nucleotide polymorphisms (SNPs) associated with hypertension and 36 associated with systolic blood pressure.12 Multiple genetic defects must be considered because hypertension has little effect on reproductive potential, which suggests to me a complex regulatory pathway.
This theory of the etiology of primary hypertension has therapeutic implications. Therapeutic phlebotomy should be first line therapy for hypertension, at least until a pharmacologic analog of soluble erythropoietin receptor can be developed. In a study of therapeutic phlebotomy in metabolic syndrome, removing between 250cc and 500cc of blood (comparable to the volume of a single-unit whole blood donation) from subjects on two occasions, baseline and four weeks later, decreased systolic blood pressure by 18.3 mmHg, as compared to only 0.2 mmHg in controls.13 This could be sufficient therapy for Stage 1 hypertension, before irreversible loss of aortic elasticity.
This hemodynamic paradigm can help to explain why the most commonly prescribed antihypertensive medicine in the world, the diuretic hydrochlorothiazide, is poor therapy for primary hypertension. Hydrochlorothiazide decreases systolic pressure by only 6.5 mm Hg and diastolic pressure by only 4.5 mm Hg14 because plasma volume is already contracted in primary hypertension. The hemodynamic paradigm also relegates dietary sodium restriction to its proper, marginal role in management of hypertension. Reduction of dietary salt by 4.4 g/day decreased systolic blood pressure by 4.18 mm Hg and diastolic blood pressure by 2.06 mm Hg. In hypertensive patients, systolic blood pressure dropped 5.39 mm Hg, and diastolic blood pressure dropped 2.82 mm Hg.15
Some of the observations which form the basis of my theory of the pathogenesis of hypertension are decades old, namely, the decrease in plasma volume with normal red cell mass and inverse relationship of blood viscosity to erythropoietin production, which date from the late 1960's. Understanding the full significance of these observations required recognition of the role of blood viscosity in hypertension, which dates back to the late 1970's and 80's work of Chien and Letcher.16 Of course, the more recent discovery of cardiac natriuretic factors, soluble erythropoietin receptor, and recognition of the widepspread role of cardiac mechanotransduction in cardiovascular pathophysiology were also necesary to formulate this theory.
In conclusion, I submit that primary hypertension is a disorder of viscosity homeostasis, in a similar fashion as diabetes mellitus type I is a disorder of glucose homeostasis and thrombosis is a disorder of coagulation homeostasis. The treatment of choice is therapeutic phlebotomy.
9. Increased plasma viscosity as a reason for inappropriate erythropoietin formation. Singh A, Eckardt KU, Zimmermann A, Götz KH, Hamann M, Ratcliffe PJ, Kurtz A, Reinhart WH. J Clin Invest. 1993 Jan;91(1):251-6.
12. A Catalog of Published Genome-Wide Association Studies. Hindorff LA, MacArthur J (European Bioinformatics Institute), Morales J (European Bioinformatics Institute), Junkins HA, Hall PN, Klemm AK, and Manolio TA. Available at: www.genome.gov/gwastudies. Accessed 4/12/14.
13. Effects of phlebotomy-induced reduction of body iron stores on metabolic syndrome: Results from a randomized clinical trial. Houschyar KS, Lüdtke R, Dobos GJ, Kalus U, Broecker-Preuss M, Rampp T, et al. BMC Medicine 2012; 10:54.
14. Antihypertensive efficacy of hydrochlorothiazie as evaluated by ambulatory blood pressure monitoring. A meta-analysis of randomized trials. Messerli FH, Makani H, Benjo A, Romero J, Alviar C, Bangalore S. J Am Coll Card 2011; 57(5): 590-600.
16. Direct relationship between blood pressure and blood viscosity in normal and hypertensive subjects. Role of fibrinogen and concentration. Letcher RL, Chien S, Pickering TG, Sealey JE, Laragh JH. Am J Med 1981; 70(6): 1195-1202.
LAST UPDATED: 2014-12-30