Platelets and Hypertension: the Pressure Is on the Clot

COMMENTARY Platelets and hypertension: the pressure is on the clot

AD Blann and GYH Lip Haemostasis, Thrombosis and Vascular Biology Unit, University Department of Medicine, City Hospital, Birmingham B18 7QH, UK

Keywords: platelets; thrombosis; hypertension; calcium antagonists

An important presumption regarding the pro- inducing hyperfibrinogenaemia (and therefore gression of cardiovascular pathology is that there are hyperviscosity), and exacerbate platelet activity (as discreet and easily identifiable stages through which measured by increased beta-thromboglobulin, sol- the disease passes. The first is clearly good health, uble P-selectin, and aggregability to ADP, collagen which is followed by the presence of one or more etc assessed in vitro), and that this relationship of the risk factors, and then ultimately symptomatic extends and also deteriorates with time.2,7,8 atherosclerosis. The latter stages are accompanied Hypertension differs from the other risk factors for by physiological changes to entire organ systems, atherosclerosis as there is no clearly identifiable such as critical organ ischaemia and left ventricular agent (such as the cigarette, hyperglycaemia or the hypertrophy. There are also notable changes in cer- high fat diet) which can be implicated in its relation- tain aspects of thrombosis and haemostasis in frank ship with any of Virchow’s triad. If a patient with atherosclerosis, and examples of this are increased hypertension also has hyperviscosity, it is not so levels of fibrinogen and coagulation factor VII.1 easy to justify a cause and effect relationship in It has also been recognised that the risk factors (in either direction. Nevertheless, patients with polycy- the absence of symptomatic disease) are also thaemia, in whom rheological abnormalities are accompanied by changes in the constituents of the common, frequently develop hypertension.9 blood, and increased fibrinogen in smoking and altered Increases in haematocrit and haemoglobin are also platelet function in diabetes are examples of this.2,3 It found in association with hypertension in patients is further presumed that these generally undesirable with renal artery stenosis and ‘stress-induced’ poly- conditions, unless reversed, predispose to or contrib- cythaemia (Gaisbok’s syndrome). Furthermore ute to the development of symptomatic disease. population studies have found a weak association Virchow’s original triad4 recognises three broadly between indices of rheology, such as haematocrit or independent factors in thrombogenesis, a viewpoint haemoglobin concentration, and blood pressure (BP) developed and extended by Duguid,5 Ross6 and across the whole range of pressures.10–12 The abnor- others towards the pathogenesis of atherosclerosis, malities of rheology and coagulation in hyperten- since both thrombogenesis and atherogenesis are sion are therefore suggestive of a prothrombotic state intimately related. These are changes in blood flow in this condition, and may be a reason why, despite (today interpreted as rheology and viscosity), the arterial tree being exposed to high pressures, the changes in the composition of the blood itself (such complications of hypertension (heart attacks, as platelet activity and an increased propensity to strokes) are thrombotic rather than haemorrhagic.13 coagulation), and changes in (and blood interactions In addition, increased levels of von Willebrand with) the blood vessel wall (which we now recog- factor in hypertension14,15 may well be cause and nise as generally undesirable changes to the effect, but in which direction? Is it that the hydro- endothelium). However, the relationship between dynamic stress damages the endothelium, causing these three aspects of the pathology of cardiovascu- the release of von Willebrand factor, or is it that an lar disease, and their likely development over time, undefined agent which first damages the endo- has still to be clarified. For example, it can with rela- thelium, thus releasing von Willebrand factor and tive simplicity be reasoned that the relationship which ultimately leads to hypertension? Hyperten- between smoking and myocardial infarction is due sion may also have a role in the third of Virchow’s to the tendency of the former to increase endothelial triad as enhanced platelet activation (which cell damage, promote hypercoagulability by responds to therapy with ACE inhibitors) has been reported.16–20 Whether or not these two aspects of Correspondence: Dr AD Blann drug therapy are linked remain to be proven, but Received and accepted 20 June 1997 may be clinically irrelevant. One can easily con- Platelets and hypertension AD Blann and GYH Lip 764 struct a paradigm for the influence of these drugs on adhesiveness in diabetes mellitus. JClinPath1967; 20: the megakaryocyte or on circulating platelets, so that 845–847. the apparent benefit of these agents in mortality out- 4 Virchow R. Gesammelte adhandlungen zur wissensch- come studies may be due equally to reduction in aftlichen. Krankfurt, Medinger Sohn & Co, 1856, 219– thrombosis as to a reduction in the consequences of 732. 5 Duguid JB. Pathogenesis of atherosclerosis. Lancet the relief of stress to the cardiovascular system. 21 1949; ii: 925–927. The paper by Smith and colleagues in the current 6 Ross R. The pathogenesis of atherosclerosis: a perspec- issue of the Journal of Human Hypertension extends tive for the 1990s. Nature 1993; 362: 801–809. this work in that the calcium channel antagonist isra- 7 Blann AD, McCollum CN. Adverse influence of ciga- dipine, but not the beta-blocker atenolol, also reduces rette smoking on the endothelium. Thromb Haemostas an index of platelet activity (beta-thromboglobulin) 1993; 70: 707–711. alongside BP. While only small numbers of patients 8 Mustard JF, Murphy EA. Effect of smoking on blood were studied, the observations are in keeping with a coagulation and platelet survival in man. Br Med J report by Pahor et al22 that use of calcium antagonists 1963; i: 846–849. in elderly patients increased the risk of bleeding. This 9 Lowe GDO. Blood rheology in arterial disease. Clin Sci 1986; 71: 137–146. may be supportive of the argument that hypertension 13 10 Bulpitt CJ, Hodes C, Everitt MG. The relationship confers a prothrombotic state, and that treating the between blood pressure and biochemical risk factors in a 23 high BP and vasodilatation by using calcium antag- general population. BrJPrevSocMed1976; 30: 158–162. onists reduces the prothrombotic state and perhaps 11 Sive PH et al. Distribution and multiple regression results in a pro-haemorrhagic state. However, recent analysis of blood pressure in 10000 Israeli men. Am J large trials involving calcium antagonists, such as the Epidemiol 1971; 93: 317–327. SYST-EUR study, have not demonstrated an excess 12 Stamler J et al. Multivariate analysis of the relation- of mortality, cancer or bleeding24 as previously ship of six variables to blood pressure. J Chron Dis reported in pharmacovigilance studies25. 1975; 28: 499–525. Whilst previous studies with isradipine have dem- 13 Lip GYH, Beevers DG. Abnormalities of rheology and coagulation in hypertension. J Hum Hypertens 1994; 8: onstrated a reduction in platelet activation, Smith et 21 693–702. al postulate that these effects on platelet were actu- 14 Blann AD, Naqvi T, Waite M, McCollum CN. Von Wil- ally minor and indirect, and was probably of doubtful lebrand factor and endothelial damage in essential clinical significance. If these changes merely reflected hypertension. J Hum Hypertens 1993; 7: 107–111. that effects of lowered BP on platelet function, per- 15 Kario K et al. Factor VII hyperactivity and endothelial haps the hypothesis should be extended to include cell damage are found in elderly hypertensives only the endothelium in the pathophysiological dis- when concomitant with microalbuminuria. Arterioscl cussion; for example, if uncontrolled high BP is Thromb Vasc Biol 1996; 16: 455–461. prothrombotic (with abnormalities of rheology and 16 Mehta J, Mehta P. Platelet function in hypertension coagulation) and damage to the endothelium occurs, and effect of therapy. Am J Cardiol 1981; 47: 331–334. this increases the tendency to thrombosis (Virchow’s 17 Someya N et al. Suppressive effect of captopril on plat- triad). However, successful lowering and good control elet aggregation in essential hypertension. J Cardiov- 9,26 asc Pharmacol 1984; 6: 840–843. of BP results in less endothelial damage; hence, we 18 James IM et al. Treatment of hypertension with capto- would predict, less platelet activation and thrombog- pril: preservation of regional blood flow and reduced enesis. The precise identification of the components platelet aggregation. J Hum Hypertens 1988; 2: 21–25. involved in the pathophysiology of hypertension and 19 Islim IF, Beevers DG, Bareford D. The effect of antihy- thrombogenesis still require further studies. pertensive drugs on in vivo platelet activity in essen- Whatever the reason, it is therefore difficult to tial hypertension. J Hypertens 1992; 10: 379–383. construct a cause and effect mechanism between 20 Lip GYH et al. Soluble adhesion molecule P-selectin platelet hyperactivity and hypertension, in either and endothelial dysfunction in essential hypertension: direction. Whether or not this has any clinical implications for atherogenesis? A preliminary report. consequence or remains merely on interesting side J Hypertens 1995; 13: 1674–1678. effect cannot yet be answered. We also need further 21 Smith A et al. Pro-haemorrhagic effects of calcium antagonists: a comparison of isradipine and atenalol details on the prognostic value of measuring these on ex-vivo platelet function in hypertensive subjects. indices, and whether demonstration of abnormal J Hum Hypertens 1997; 11: 783–788. platelet activation identifies a high risk group for 22 Pahor M et al. Risk of gastrointestinal haemorrhage cardiovascular events and stroke. More data on large with calcium antagonists in hypertensive persons over populations of hypertensives and the value of treat- 67 years old. Lancet 1996; 347: 1061–1065. ment strategies in reducing the prothrombotic state, 23 Gill JS, Zezulka AV, Horrocks PM. Rupture of a cer- as well as the prognostic implications, require ebral aneurysm associated with nifedipine treatment. close scrutiny. Postgrad Med J 1986; 62: 1029–1030. 24 Staessen JA et al for the Syst-Eur Trial Investigators. Morbidity and mortality in the placebo-controlled Eur- References opean Trial on Isolated Systolic Hypertension in the Elderly. Lancet (in press). 1 Van de Loo J. Circulating factors of the haemostatic 25 Beevers DG, Sleight P. Short acting dihydropyridine systems as indicators of increased or reduced coronary (vasodilating) calcium channel blockers for hyperten- risk. Br J Haem 1995; 91: 777–782. sion: is there a risk? Br Med J 1996; 312: 1143–1145. 2 Meade TW, Imeson J, Stirling Y. Effects of changes in 26 Blann AD, Waite MA. Von Willebrand factor and sol- smoking and other characteristics on clotting and the risk uble E-selectin in hypertension: influence of treatment of ischaemic heart disease. Lancet 1987; ii: 986–988. and value in predicting the progression of atheroscler- 3 Shaw S, Pegrum G, Wolff S, Ashton WL. Platelet osis. Coronary Artery Dis 1996; 7: 143–147.