Journal of Human Hypertension (2008) 22, 380–388 & 2008 Nature Publishing Group All rights reserved 0950-9240/08 $30.00 www.nature.com/jhh REVIEW Hypertension and cardiac arrhythmias: a review of the epidemiology, pathophysiology and clinical implications

K-H Yiu and H-F Tse Division of Cardiology, Department of Medicine, Queen Mary Hospital, The University of Hong Kong, Hong Kong, China

Hypertension is commonly associated with cardiac cular ectopy and sudden cardiac death. Recent arrhythmias in patients with and without concomitant prospective clinical trials reveal that antihyper- cardiovascular disease. Experimental and epidemiolo- tensive therapy may delay or prevent the occurrence gical studies have demonstrated potential links of cardiac arrhythmias and sudden cardiac death in between hypertension and atrial and ventricular arrhyth- patients with hypertension. Although antihypertensive mias, although the underlying pathophysiological me- agents that block the renin–angiotensin–aldosterone chanism remains unclear. Nonetheless, the importance system appear to protect against cardiac arrhythmias, of hypertension as a cause of atrial and ventricular this needs to be confirmed by current ongoing clinical arrhythmias is not well recognized. In particular, trials. the occurrence of left ventricular hypertrophy is a Journal of Human Hypertension (2008) 22, 380–388; strong predictor for the development of AF, ventri- doi:10.1038/jhh.2008.10; published online 13 March 2008

Keywords: atrial fibrillation; arrhythmia; sudden cardiac death

Introduction Epidemiology Concomitant cardiac arrhythmias are commonly Atrial fibrillation seen in patients with hypertension, although the Atrial fibrillation is the most common sustained mechanism of this association is unclear. The arrhythmia in adults and is associated with an contribution of hypertension to the development of increased risk for cardiovascular morbidity, mortality atrial and ventricular arrhythmias is unrecognized and stroke.1,2 The incidence of AF increases with age; and thus undertreated. Recent clinical trials have a prevalence of 0.1% in adults younger than 55 years demonstrated the possible preventive role of anti- increases to 9% in adults older than 80 years.3,4 In hypertensive agents, in particular those that induce the Framingham Heart Study, the lifetime risk for blockade of the renin–angiotensin–aldosterone sys- developing AF was one in four in men and women tem (RAAS). The aim of this paper is to review the aged 40 years or above.5 Among different risk epidemiology, underlying mechanisms and clinical factors, hypertension is one of the important implications of atrial and ventricular arrhythmias independent risk factors for the development of in hypertensive patients. We searched MEDLINE AF.6 In the Cardiovascular Health Study of subjects using the following terms individually and/or in older than 65 years of age, the risk for developing combination: ‘atrial fibrillation (AF)’, ‘left ventricu- new-onset AF was higher in hypertensive patients: lar hypertrophy (LVH)’, ‘hypertension’, ‘ventricular the relative risk was 1.11 after adjustment for age arrhythmias’ and ‘sudden cardiac death (SCD)’. In and other covariates.7 addition, abstracts from international cardiovascular The Framingham Study cohort8 also demonstrated meetings were studied to identify unpublished that hypertension and diabetes were the only studies. cardiovascular risk factors that were responsible for the development of new-onset AF. Hypertension was responsible for more AF in the population (14%) than any other risk factor due to its high Correspondence: Professor H-F Tse, Division of Cardiology, prevalence. After adjusting for age, hypertension Department of Medicine, Queen Mary Hospital, The University increased the risk of AF by 80% in female patients of Hong Kong, Room 1928, Block K, Hong Kong, China. E-mail: [email protected] and 70% in male patients. In addition, evidence of Received 28 November 2007; revised 19 January 2008; accepted LVH on ECG was associated with at least a threefold 26 January 2008; published online 13 March 2008 increased risk for AF. In the Progetto Ipertensione Hypertension and cardiac arrhythmias K-H Yiu and H-F Tse 381 Umbria Monitoraggio Ambulatoriale (PIUMA) Study, age and left ventricular (LV) mass as measured by echocardiogram were independent predictors for new-onset AF after a mean follow-up period of 5.3 years.9 The high prevalence of coexistent hypertension and AF is also reflected by the high proportion of hypertensive patients who require antihypertensive treatment (470%) and who are participating in major clinical trials on AF, such as the Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) Study10 and the Stroke Prevention Using an Oral Thrombin Inhibitor in Atrial Fibrillation (SPORTIF III) Study.11

Ventricular arrhythmia In patients with hypertension, the most important Figure 1 Potential mechanisms of atrial fibrillation related to determinant for the occurrence of ventricular ar- hypertension. rhythmia is the presence of LVH. Several studies have shown that hypertensive patients with LVH Renin–angiotensin–aldosterone system. It has been have an increased frequency of premature ventricu- well demonstrated that activation of the RAAS lar ectopic beats and ventricular arrhythmias.12–15 In occurs in patients with hypertension21 and is the Framingham Heart Study, echocardiographic strongly implicated in the development of AF.22 evidence of LVH was associated with ventricular Evidence is accumulating that explains the link arrhythmia independent of conventional cardio- vascular risk factors.12 In untreated hypertensive between RAAS and AF: there are two receptor subtypes that act as binding sites for angiotensin patients, non-sustained ventricular arrhythmia II, namely AT1 and AT2. Most of the known effects was observed in up to 5% of patients during 24-h of angiotensin II in adult tissue are attributable to Holter monitoring.14 the actions AT1 receptor and acts antagonistically by the action of AT2.23 In addition to its known Sudden cardiac death physiological effects (vasoconstriction, aldosterone The presence of LVH is strongly correlated with and vasopressin release, sodium and water retention SCD, although a direct causal relationship has not and sympathetic facilitation), AT1 also induces been identified.16,17 In the Framingham Heart atrial myocyte hypertrophy and fibrosis that predis- Study,17 an incremental increase in LV mass by pose an individual to development of AF.24 Upre- 50 g mÀ2 conferred a 45% higher risk of SCD in gulation of the AT1 receptor in the left atrium in hypertensive subjects aged 440 years. Increased LV patients with AF, but not the AT2 receptor, has been mass and LVH were also associated with a long-term demonstrated.25 In contrast, there is downregulation risk of sudden death. The presence of asymptomatic of AT1 and upregulation of AT2 in the right complex or frequent ventricular arrhythmia in the atrium.26 Angiotensin II is known to induce pro- absence of coronary artery disease was also asso- liferation of fibroblasts and extracellular matrix ciated with a 62% increase in mortality.18 Electro- protein accumulation via activation of the mitogen- physiological studies in hypertensive patients with activated protein kinases.27 Experimental studies LVH have nonetheless yielded conflicting show that inhibition of this angiotensin II pathway results.19,20 In symptomatic hypertensive patients, attenuates the formation of fibrosis and diminishes the presence of LVH increased susceptibility to the the incidence of AF.28 In a hypertensive rat heart development of malignant ventricular arrhythmia model, the degree of atrial fibrosis correlated with during programmed ventricular stimulation.20 the susceptibility for development of atrial tachy- cardia.29 Structural remodelling of the atria related to the activation of RAAS can thus contribute to the Pathophysiology development of AF in hypertension. In an animal model, rapid atrial pacing resulted in The development of atrial and ventricular arrhyth- atrial electrical remodelling, characterized by short- mias in hypertensive patients and patients with LVH ening of the atrial effective refractory period and appears to be complex. It is, nonetheless, clearly loss of the normal rate adaptation of refractoriness.30 evident that hypertension and LVH exert a robust Blockade of RAAS can prevent the development of proarrhythmic effect on the heart. electrical remodelling for short (3 h),31 but not long periods of time (1–5 weeks).32 It has been postulated Atrial fibrillation that an increased angiotensin II level is proarrhyth- In patients with hypertension, several mechanisms mic due to increased intracellular calcium. The may contribute to the development of AF (Figure 1). latter is due to increased intake of extracellular

Journal of Human Hypertension Hypertension and cardiac arrhythmias K-H Yiu and H-F Tse 382 calcium and increased release from the sarcoplas- mic reticulum through the activation of membrane L-type calcium channels or phosphatidylinositol– phospholipase C pathways.33,34 In addition, activa- tion of RAAS may lead to inflammation: this has also been implicated in the pathogenesis of AF.35 Experimental studies have shown that angiotensin II possesses proinflammatory properties, inducing the production of reactive oxygen species, inflammatory cytokines and adhesion molecules.36 In accordance with this observation, angiotensin II receptor block- ade has been shown to significantly reduce multiple markers of inflammation (C-reactive protein, TNF-a, IL-6 and monocyte chemotactic protein-1) in hyper- tensive patients.37

Haemodynamic changes in atria. The excessive increase in afterload imposed by hypertension will Figure 2 Potential mechanisms of ventricular arrhythmias lead to cardiac compensation with progressive related to hypertension. thickening of the LV wall and consequent LVH. Histologically, LVH is characterized by hypertrophy Electrophysiology disturbance. It has been well of existing cardiomyocytes surrounded by connec- 38,39 established that prolongation and dispersion of LV tive tissue with cardiac fibrosis. In addition to repolarization are associated with an increased risk the established relationship between LVH and 40 of ventricular tachyarrhythmias including torsade adverse cardiovascular events, previous studies de pointe and ventricular tachycardia. One of the have shown that the presence of LVH may salient features in myocardial hypertrophy is the also predispose to the development of cardiac 8,12 presence of early after-depolarization and triggered arrhythmia. Increased stiffness of the LV after activity that may increase the dispersion of repolar- LVH will inevitably lead to diastolic dysfunction ization to cause sustained arrhythmia.47 Indeed, that subsequently contributes to the occurrence of 41,42 prolongation of the QT interval, similar to a AF. Long-standing increased LA (left atrial) pharmacologically induced proarrhythmic effect, pressure due to impaired LV relaxation in the has also been observed in LVH.48 In addition, QT hypertensive heart will also eventually lead to LA 43 dispersion that represents the degree of repolariza- enlargement. In the Framingham Heart Study, both tion inhomogeneity in the heart is associated with the duration of elevated blood pressure and the level LVH 49 and hypertension.50 At a cellular level, of systolic blood pressure were associated with the structural remodelling induced by hypertension is development of LA dilatation. The consequent associated with impaired cell–cell communication distention and stretching of the atria related to at the gap junction and increased susceptibility to hypertension can alter atrial electrophysiological ventricular arrhythmia induced by hypokalaemia properties, including shortening of the effective in an animal model.51 Nonetheless, vulnerability refractory period and increased dispersion of to ventricular fibrillation is not related to abnormal refractoriness with subsequent increased vulner- 52 44,45 calcium channel handling. Although there ability to AF. remains much speculation about the underlying mechanism, it is established that electrical instabil- ity induced by hypertension and LVH is Ventricular arrhythmia an important mechanism in inducing ventricular The presence of LVH is the most important determi- arrhythmia. nant for the occurrence of ventricular arrhythmia and/or SCD in hypertensive patients. The pattern of Myocardial ischaemia. Hypertension and LVH LVH may also have important implications for the may also cause myocardial ischaemia through other prognosis and type of cardiovascular complications mechanisms. Hypertrophy of the LV will lead to a in hypertensive patients. Previous studies have mismatch between myocardial blood supply and shown that hypertensive patients with chronic oxygen consumption. In hypertensive patients with LVH may be more prone to severe ventricular or without LVH, decreased diastolic coronary blood arrhythmias, whereas concentric hypertrophy is flow may cause subendocardial ischaemia.53 Micro- more related to ischaemic events that result from vascular dysfunction with myocardial ischaemia abnormal coronary autoregulation.46 As shown in has been reported in both prehypertensive and Figure 2, several mechanisms have been proposed to hypertensive patients even in the absence of explain the relationship between the presence of LVH, 54,55 suggesting that impaired myocardial perfu- LVH and ventricular arrhythmias in hypertension. sion may develop in the early stage of hypertension.

Journal of Human Hypertension Hypertension and cardiac arrhythmias K-H Yiu and H-F Tse 383 The presence of myocardial ischaemia may trigger ACEI attenuates atrial structural remodelling with the onset of ventricular arrhythmia and SCD in fibrosis and atrial electrical remodelling, both patients with hypertension. factors that predispose to AF.68,69

Myocardial fibrosis. Experimental studies have shown that pressure overloading of the LV is Atrial fibrillation associated with increased collagen synthesis and The effect of specific class of antihypertensive degradation.56 Excessive accumulation of fibrillar agents on the prevention of AF remains unclear. 70 collagen is a characteristic feature of hypertensive Retrospective studies suggest that blockade of heart disease. The increased myocardial fibrosis RAAS with ACEI in patients with hypertension may lead to distortion of the tissue structure and reduces the relative risk of developing AF by 15% increased stiffness of the myocardium to cause LV compared with CCB (calcium channel blocker). diastolic dysfunction.57 These structural changes in Patients with a prior history of AF appear to derive the myocardium related to increased myocardium greater benefit from ACEI-based therapy in the 66,71,72 fibrosis can lead to nonhomogeneous propagation of prevention of AF. Emerging data from the electrical impulses and give rise to ventricular analysis of large prospective clinical studies also arrhythmias due to re-entry. demonstrate the beneficial effects of blockade of RAAS in preventing AF in patients with hyperten- Neuroendocrine effects. Excessive activation of the sion. In the Losartan Intervention For Endpoint sympathetic nervous system and the RAAS has been Reduction in Hypertension (LIFE) Study, ARB-based demonstrated in the pathogenesis of essential therapy (losartan) was associated with a 33% hypertension as well as in the development of reduction in the incidence of AF and with a LVH.22,58 Sympathetic activation has been shown significant reduction in the incidence of subsequent stroke compared with b-blocker-based therapy to exert a direct proarrhythmic effect that may lead 73 to ventricular arrhythmia and SCD.59,60 No direct (atenolol). Similarly, in the Valsartan Antihyper- link has been demonstrated between angiotensin II tensive Long-term Use Evaluation (VALUE) Trial, and ventricular arrhythmias. Nevertheless, angio- ARB-based therapy (valsartan) had a 16% risk tensin II can lead to LVH by increasing blood reduction in the incidence of AF compared with CCB-based therapy (amlodipine) after adjustment pressure or possibly via direct trophic actions on 71 the heart61 that subsequently increase vulnerability for confounding covariates. In patients with con- to ventricular arrhythmias and SCD. comitant hypertension and AF, ARB-based therapy (losartan) in combination with amiodarone was also more effective than CCB-based therapy (amlodipine) Clinical implication in preventing the recurrence of AF.74 Nevertheless, meta-analysis of three major clinical trials (Captopril While the pathophysiology of cardiac arrhythmias Prevention Project (CAPP), Swedish Trial in Old in hypertension is unknown, the identification of a Patients with Hypertension-2 (STOP-H-2) and LIFE) close link between blood pressure control and the including 12 114 patients with hypertension have presence of LVH with the development of cardiac failed to show a consistent reduction in AF with arrhythmias offers possibilities for therapeutic inter- ACEI and ARB.75 In addition to the blockade of vention and prevention of AF, ventricular arrhyth- RAAS with specific agents, both the blood pressure- mias and SCD in hypertensive patients by means of lowering effect and LVH regression of different blood pressure control and LVH regression. Existing antihypertensive agents make an important contri- evidence demonstrates that optimal blood pressure bution to the prevention of AF. Recent analysis of control and regression of LVH with antihypertensive data from the LIFE Study reveals that regression of therapy are associated with improved LV diastolic electrocardiographic LVH during antihypertensive function, reduced LA size and changes in plasma therapy was a major determinant for the prevention catecholamines that may contribute to the preven- of AF, independent of the blood pressure lowering tion of cardiac arrhythmias.62–64 Among different and the effect of ARB-based therapy.72 classes of antihypertensive agents, angiotensin Several prospective clinical studies (Table 1) are receptor blockers (ARBs) and angiotensin-convert- currently underway to address whether blockade of ing enzyme inhibitors (ACEIs) appear to have RAAS with ARB can prevent AF in patients with better efficacy in terms of reduction of cardiac hypertension. The GISSI-AF Trial intends to exam- arrhythmias.65 Blockade of RAAS has been proven ine the incidence of recurrent AF in a broad to regress LVH and likely explains, in part, its spectrum of cardiovascular risk patients including protective effect in arrhythmias. The regression of hypertension (n ¼ 1402) with valsartan versus pla- LVH by ARB in patients with hypertension is cebo.76 The Irbesartan for the Prevention of Atrial independent of its blood pressure-lowering effect.66 Arrhythmias and Cardiac Electrical Remodeling Furthermore, regression of hypertensive LVH is seen (I-PACE) Trial investigates the effect of irbesartan independently of lowering blood pressure.67 Experi- versus placebo in hypertensive patients with atrial mental studies have shown that blockade of ARB or tachyarrhythmias and implanted pacemakers.77 The

Journal of Human Hypertension Hypertension and cardiac arrhythmias K-H Yiu and H-F Tse 384 Table 1 Ongoing studies of AF involving blockade of renin–angiotensin–aldosterone system in hypertension patients

Name of study Study design Study population Duration AF end points

GISSI-Atrial Fibrillation A prospective, multicentre, (1) Symptomatic AF in recent 1 year Primary end points: Trial76 randomized, double-blind, 6 months or successful (1) the first recurrence of AF controlled trial of valsartan cardioversion between (2) the number of patients vs placebo 14 days and 48 h with more than one AF (2) History of heart failure episode over the whole with left ventricular ejection follow-up period fraction o40%/hypertension/ diabetes/stroke/cardiovascular disease Irbesartan for the Prevention A prospective, multicentre, (1) Permanent, atrial or dual- 6 months Primary end point: of Atrial Arrhythmias randomized, double-blind, chamber pacemaker implanted time to recurrent atrial high and Cardiac Electrical controlled trial of irbesartan 42 months before enrolment rate episodes (220 minÀ1 for Remodeling in Patients with vs placebo (2) History of prior diagnosis 42 min) Hypertension and Permanent of hypertension and/or treated Pacemakers for hypertension or two (I-PACE)77 documented BP4130/85 Atrial Fibrillation ACTIVE-I is a partial (1) Permanent AF or at least Varies, to Secondary end point: Clopidogrel Trial with factorial, double-blind, two episodes of intermittent achieve the recurrence of AF Irbesartan for prevention of controlled trial of irbesartan AF in the past 6 months target numbers Vascular Events-I vs placebo in patients (2) Systolic BP of at least of primary (ACTIVE-I) Trial78 participating in ACTIVE-A 110 mm Hg outcome events or ACTIVE-W Japanese Rhythm A randomized prospective (1) History of paroxysmal 1 year Primary end point: Management Trial II for multicentre controlled AF within preceding 6 the difference in the Atrial Fibrillation trial of candesartan vs months; and number of days with (J-RHYTHM II study)79 amlodipine (2) Hypertension, defined symptomatic and as a systolic asymptomatic AFs BPX140 mm Hg and/or a recorded on diastolic BPX90 mm Hg transtelephonic monitoring ONgoing Telmisartan Alone A randomized prospective Age455 years and with 3.5–5.5 years Secondary end point: and in Combination with multicentre trial to evaluate history of coronary artery newly diagnosed AF Ramipril Global Endpoint (1) telmisartan+ramipril vs disease/stroke/diabetes/ Trial (ONTARGET)80 ramipril peripheral artery disease (2) telmisartan vs ramipril

Abbreviations: AF, atrial fibrillation; BP, blood pressure. Atrial fibrillation Clopidogrel Trial with Irbesartan and rhythm-control strategy, in terms of quality of for prevention of Vascular Events-I (ACTIVE-I) Trial life, stroke and mortality.81–84 These studies high- assesses the effect of irbesartan versus placebo on lighted the low clinical efficacy of current antiar- the prevention of AF recurrence in patients with rhythmic agents to maintain sinus rhythm. This multiple cardiovascular risk factors and paroxysmal probably contributes to the stroke risk associated AF.78 The Japanese Rhythm Management Trial II for with rhythm control, especially in a high-risk Atrial Fibrillation (J-RHYTHM II Study) compares hypertensive population. Therefore, in patients with the effect of candesartan versus amlodipine for hypertension and AF, the use of oral anticoagulant treatment of paroxysmal AF in patients with with warfarin should be considered for stroke hypertension.79 The Ongoing Telmisartan Alone prevention, irrespective of whether rate or rhythm- and in Combination with Ramipril Global Endpoint control strategy is used.85 However, one of the major Trial (ONTARGET) Study investigates the efficacy of drawbacks of warfarin is the potential of increasing telmisartan, ramipril and combination therapy with risk of haemorrhagic events. Therefore, the risk and telmisartan plus ramipril for reducing cardiovascu- benefit of warfarin therapy must be assessed lar events, including AF in high-risk patients with individually for every patient. Compared with controlled blood pressure. This study will provide warfarin, aspirin is less effective in preventing important data on the potential incremental benefit stroke, and can only be recommended for patients of ACEI, ARB or both in preventing AF, independent who refuse or cannot safely take warfarin. of blood pressure lowering.80 Finally, data concern- In those patients who need ventricular rate ing the potential effect of blockade of RAAS with control, either b-blocker and/or a rate-limiting CCBs aldosterone antagonists, such as spironolactone and (diltiazem or verapmail) is a suitable initial choice eplerenone, for AF prevention in patients with of therapy in patients with hypertension. In patients hypertension are lacking. Further studies are needed with impaired LV function or LVH, the use of Class I to discern their association. and III antiarrhythmic agents is associated with an In regard to the treatment of AF in patients with increase risk of proarrhythmia. Therefore, only hypertension, recent randomized trials have demon- amiodarone is recommended for rhythm control in strated similar clinical outcomes with rate-control patients with hypertension and LVH.85

Journal of Human Hypertension Hypertension and cardiac arrhythmias K-H Yiu and H-F Tse 385 Ventricular arrhythmia different agents to lower blood pressure and, more Data on the prevention of ventricular arrhythmias importantly, to regress LVH can prevent AF and and SCD in patients with hypertension are limited. SCD. Whether the use of ACEI or ARB to block the Previous studies suggested that regression of LVH RAAS is more effective in preventing cardiac with captopril reduced the number of ventricular arrhythmias in patients with hypertension remains ectopy.86 Until recently, the only data available on to be determined by current ongoing clinical trials. the impact of LVH regression and SCD came from LIFE.87 In this study, the presence of regression of electrocardiographic LVH during antihypertensive References therapy was associated with a 30% lower risk of SCD independent of blood pressure lowering and 1 Benjamin EJ, Wolf PA, D’Agostino RB, Silbershatz H, other known predictors for SCD. Furthermore, ARB- Kannel WB, Levy D. Impact of atrial fibrillation on the based therapy seemed to provide similar cardiopro- risk of death: the Framingham Heart Study. Circulation tection against SCD compared with b-blocker-based 1998; 98: 946–952. therapy. Although b-blocker has been shown to 2 Stewart S, Hart CL, Hole DJ, McMurray JJ. A popula- prevent SCD in patients with heart failure and tion-based study of the long-term risks associated with 88,89 atrial fibrillation: 20-year follow-up of the Renfrew/ myocardial infarction, case–control studies in Paisley study. Am J Med 2002; 113: 359–364. patients with hypertension suggested that b-blocker 3 Go AS, Hylek EM, Phillips KA, Chang Y, Henault LE, 90 was associated with an increased risk of SCD. Selby JV et al. Prevalence of diagnosed atrial fibrilla- Similarly, the use of thiazide diuretic has been tion in adults: national implications for rhythm linked to an increased risk of cardiac arrhythmia. management and stroke prevention: the AnTicoagula- Siscovick et al.91 had shown a dose-dependent tion and Risk Factors In Atrial Fibrillation (ATRIA) increase in SCD in patients treated with thiazide Study. JAMA 2001; 285: 2370–2375. diuretic. Although the exact mechanism remains 4 Heeringa J, van der Kuip DAM, Hofman A, Kors JA, unknown, it is possible that undetected hypokalae- van Herpen G, Stricker BHC et al. Prevalence, mia related to the use of high-dose diuretic can lead incidence and lifetime risk of atrial fibrillation: the Rotterdam study. Eur Heart J 2006; 27: 949–953. to ventricular arrhythmia including torsade de 5 Lloyd-Jones DM, Wang TJ, Leip EP, Larson MG, Levy D, pointes. Indeed, the use of non-potassium-sparing Vasan RS et al. Lifetime risk for development of atrial diuretic was associated with a twofold increased fibrillation: the Framingham Heart Study. Circulation risk of SCD compared with the use of potassium- 2004; 110: 1042–1046. sparing diuretic.90 In contrast, recent studies 6 Benjamin EJ, Levy D, Vaziri SM, D’Agostino RB, demonstrated that candesartan-based and amlodi- Belanger AJ, Wolf PA. Independent risk factors for pine-based regimens produced no statistical atrial fibrillation in a population-based cohort. The differences in the risk of SCD in high-risk hyperten- Framingham Heart Study. JAMA 1994; 271: 840–844. sive patients.92 Nevertheless, future studies are 7 Psaty BM, Manolio TA, Kuller LH, Kronmal RA, needed to confirm whether alleviation of LV Cushman M, Fried LP et al. Incidence of and risk factors for atrial fibrillation in older adults. Circulation structural remodelling related to hypertension by 1997; 96: 2455–2461. LVH regression with different classes of antihyper- 8 Kannel WB, Wolf PA, Benjamin EJ, Levy D. Prevalence, tensive agents can prevent ventricular arrhythmias incidence, prognosis, and predisposing conditions for and SCD. atrial fibrillation: population-based estimates. Am In hypertensive patients with asymptomatic and J Cardiol 1998; 82(Suppl 1): 2N–9N. non-sustained ventricular arrhythmias, there is no 9 Verdecchia P, Reboldi G, Gattobigio R, Bentivoglio M, role for prophylactic use of antiarrhythmic agents. Borgioni C, Angeli F et al. Atrial fibrillation in In these hypertensive patients, especially those with hypertension: predictors and outcome. Hypertension LVH or LV dysfunction and those who suffered from 2003; 41: 218–223. sustained and symptomatic ventricular tachycardia 10 Wyse DG, Waldo AL, DiMarco JP, Domanski MJ, Rosenberg Y, Schron EB et al. A comparison of rate not related to a reversible causes, implantable control and rhythm control in patients with atrial cardioverter defibrillator should be considered for fibrillation. N Engl J Med 2002; 347: 1825–1833. prevention of SCD. 11 Olsson SB, Executive Steering Committee of the SPORTIF III Investigators. Stroke prevention with the oral direct thrombin inhibitor ximelagatran compared Conclusions with warfarin in patients with non-valvular atrial fibrillation (SPORTIF III): randomised controlled trial. Experimental and epidemiological studies have Lancet 2003; 362: 1691–1698. established a close link between hypertension, 12 Messerli FH, Ventura HO, Elizardi DJ, Dunn FG, LVH and cardiac arrhythmias. In patients with Frohlich ED. Hypertension and sudden death. Increased ventricular ectopic activity in left ventricu- hypertension, alteration of cardiac haemodynamic, lar hypertrophy. Am J Med 1984; 77: 18–22. structural and electrophysiological properties im- 13 Levy D, Anderson KM, Savage DD, Balkus SA, Kannel posed by hypertension and LVH can contribute to WB, Castelli WP. Risk of ventricular arrhythmias in left the development of AF, ventricular arrhythmias ventricular hypertrophy: the Framingham Heart Study. and SCD. Effective antihypertensive therapy with Am J Cardiol 1987; 60: 560–565.

Journal of Human Hypertension Hypertension and cardiac arrhythmias K-H Yiu and H-F Tse 386 14 McLenachan JM, Henderson E, Morris KI, Dargie HJ. 29 Choisy SCM, Arberry LA, Hancox JC, James AF. Ventricular arrhythmias in patients with hypertensive Increased susceptibility to atrial tachyarrhythmia in left ventricular hypertrophy. N Engl J Med 1987; 317: spontaneously hypertensive rat hearts. Hypertension 787–792. 2007; 49: 498–505. 15 Ferrara N, Furgi G, Longobardi G, Nicolino A, 30 Wijffels MC, Kirchhof CJ, Dorland R, Allessie MA. Acanfora D, Leosco D et al. Relation between age, left Atrial fibrillation begets atrial fibrillation: a study in ventricular mass and ventricular arrhythmias in awake chronically instrumented goats. Circulation patients with hypertension. J Hum Hypertens 1995; 1995; 92: 1954–1968. 9: 581–587. 31 Nakashima H, Kumagai K, Urata H, Gondo N, Ideishi 16 Cupples LA, Gagnon DR, Kannel WB. Long- and short- M, Arakawa K. Angiotensin II antagonist prevents term risk of sudden coronary death. Circulation 1992; electrical remodeling in atrial fibrillation. Circulation 85(1 Suppl): I11–I18. 2000; 101: 2612–2617. 17 Haider AW, Larson MG, Benjamin EJ, Levy D. 32 Kumagai K, Nakashima H, Urata H, Gondo N, Arakawa Increased left ventricular mass and hypertrophy are K, Saku K. Effects of angiotensin II type 1 receptor associated with increased risk for sudden death. JAm antagonist on electrical and structural remodeling Coll Cardiol 1998; 32: 1454–1459. in atrial fibrillation. J Am Coll Cardiol 2003; 41: 18 Bikkina M, Larson MG, Levy D. Asymptomatic 2197–2204. ventricular arrhythmias and mortality risk in subjects 33 Marrero MB, Paxton WG, Duff JL, Berk BC, Bernstein with left ventricular hypertrophy. J Am Coll Cardiol KE. Angiotensin II stimulates tyrosine phosphoryla- 1993; 22: 1111–1116. tion of phospholipase C-gamma 1 in vascular smooth 19 Kowey PR, Friechling TD, Sewter J, Wu Y, Sokil A, muscle cells. J Biol Chem 1994; 269: 10935–10939. Paul J et al. Electrophysiological effects of left 34 Griendling KK, Rittenhouse SE, Brock TA, Ekstein LS, ventricular hypertrophy. Effect of calcium and Gimbrone MA et al. Sustained diacylglycerol forma- potassium channel blockade. Circulation 1991; 83: tion from inositol phospholipids in angiotensin II- 2067–2075. stimulated vascular smooth muscle cells. J Biol Chem 20 Vester EG, Kuhls S, Ochiulet-Vester J, Vogt M, Strauer 1986; 261: 5901–5906. BE. Electrophysiological and therapeutic implications 35 Engelmann MD, Svendsen JH. Inflammation in the of cardiac arrhythmias in hypertension. Eur Heart genesis and perpetuation of atrial fibrillation. J 1992; 13(Suppl D): 70–81. Eur Heart J 2005; 26: 2083–2092. 21 Wright J, Mizutani S, Harding JW. Pathways 36 Brasier AR, Recinos III A, Eledrisi MS. Vascular involved in the transition from hypertension to inflammation and the renin–angiotensin system. hypertrophy to heart failure. Treatment strategies. Arterioscler Thromb Vasc Biol 2002; 22: 1257–1266. Heart Fail Rev 2007; e-pub ahead of print 7 November 37 Lopez B, Querejeta R, Varo N, Gonzalez A, Larman M, 2007. Martinez Ubago JL et al. Usefulness of serum 22 Goette A, Staack T, Rocken C, Arndt M, Geller JC, Huth carboxy-terminal propeptide of procollagen type I in C et al. Increased expression of extracellular signal- assessment of the cardioreparative ability of antihy- regulated kinase and angiotensin-converting enzyme pertensive treatment in hypertensive patients. in human atria during atrial fibrillation. J Am Coll Circulation 2001; 104: 286–291. Cardiol 2000; 35: 1669–1677. 38 Lorell BH, Carabello BA. Left ventricular hypertrophy: 23 Kaschina E, Unger T. Angiotensin AT1/AT2 receptors: pathogenesis, detection, and prognosis. Circulation regulation, signalling and function. Blood Press 2003; 2000; 102: 470–479. 12: 70–88. 39 Weber KT. Targeting pathological remodeling: 24 Rogg H, de Gasparo M, Graedel E, Stulz P, Burkart F, concepts of cardioprotection and reparation. Circulation Eberhard M et al. Angiotensin II-receptor subtypes 2000; 102: 1342–1345. in human atria and evidence for alterations in 40 Vakili BA, Okin PM, Devereux RB. Prognostic implica- patients with cardiac dysfunction. Eur Heart J 1996; tions of left ventricular hypertrophy. Am Heart J 2001; 17: 1112–1120. 141: 334–341. 25 Boldt A, Wetzel U, Weigl J, Garbade J, Lauschke J, 41 Tsang TS, Barnes ME, Gersh BJ, Bailey KR, Seward JB. Hindricks G et al. Expression of angiotensin II Risks for atrial fibrillation and congestive heart failure receptors in human left and right atrial tissue in atrial in patients X65 years of age with abnormal left fibrillation with and without underlying mitral valve ventricular diastolic relaxation. Am J Cardiol 2004; disease. J Am Coll Cardiol 2003; 42: 1785–1792. 93: 54–58. 26 Goette A, Arndt M, Rocken C, Spiess A, Staack T, 42 Tsang TS, Gersh BJ, Appleton CP, Tajik AJ, Barnes ME, Geller JC et al. Regulation of angiotensin II receptor Bailey KR et al. Left ventricular diastolic dysfunction subtypes during atrial fibrillation in humans. Circula- as a predictor of the first diagnosed nonvalvular atrial tion 2000; 101: 2678–2681. fibrillation in 840 elderly men and women. J Am Coll 27 Yano M, Kim S, Izumi Y, Yamanaka S, Iwao H. Cardiol 2002; 40: 1636–1644. Differential activation of cardiac c-Jun amino-terminal 43 Vaziri SM, Larson MG, Lauer MS, Benjamin EJ, Levy D. kinase and extracellular signal-regulated kinase in Influence of blood pressure on left atrial size. The angiotensin II mediated hypertension. Circ Res 1998; Framingham Heart Study. Hypertension 1995; 25: 83: 752–760. 1155–1160. 28 Li D, Shinagawa K, Pang L, Leung TK, Cardin S, Wang 44 Ravelli F, Allessie M. Effects of atrial dilatation on Z et al. Effects of angiotensin-converting enzyme refractory period and vulnerability to atrial fibrillation inhibition on the development of the atrial fibrillation in the isolated Langendorff-perfused rabbit heart. substrate in dogs with ventricular tachypacing- Circulation 1997; 90: 1686–1695. induced congestive heart failure. Circulation 2001; 45 Satoh T, Zipes DP. Unequal atrial stretch in dogs 104: 2608–2614. increases dispersion of refractoriness conducive to

Journal of Human Hypertension Hypertension and cardiac arrhythmias K-H Yiu and H-F Tse 387 developing atrial fibrillation. J Cardiovasc Electrophysiol 61 Harrap SB, Mitchell GA, Casley DJ, Mirakian C, 1996; 7: 833–842. Doyle AE. Angiotensin II, sodium, and cardiovascular 46 Levy D, Anderson KM, Savage DD, Kannel WB, hypertrophy in spontaneously hypertensive rats. Christiansen JC, Castelli WP. Echocardiographically Hypertension 1993; 21: 50–55. detected left ventricular hypertrophy: prevalence and 62 Klingbeil AU, Schneider M, Martus P, Messerli FH, risk factors. The Framingham Heart Study. Ann Intern Schmieder RE. A meta-analysis of the effects of Med 1988; 108: 7–13. treatment on left ventricular mass in essential hyper- 47 Hart G. Cellular electrophysiology in cardiac hyper- tension. Am J Cardiol 2003; 115: 41–46. trophy and failure. Cardiovasc Res 1994; 28: 933–946. 63 Devereux RB, Palmieri V, Liu JE, Wachtell K, Bella JN, 48 Savelieva I, Yap YG, Yi G, Guo XH, Hnatkova K, Camm Boman K et al. Progressive hypertrophy regression AJ et al. Relation of ventricular repolarization to with sustained pressure reduction in hypertension: the cardiac cycle length in normal subjects, hypertrophic Losartan Intervention For Endpoint Reduction study. cardiomyopathy, and patients with myocardial infarction. J Hypertens 2002; 20: 1445–1450. Clin Cardiol 1999; 22: 649–654. 64 Manolis AJ, Beldekos D, Handanis S, Haralabidis G, 49 Davey PP, Bateman J, Mulligan IP, Forfar C, Barlow C, Hatzissavas J, Foussas S et al. Comparison of spirapril, Hart G. QT interval dispersion in chronic heart failure isradipine, or combination in hypertensive patients and left ventricular hypertrophy: relation to autonomic with left ventricular hypertrophy: effects on LVH nervous system and Holter tape abnormalities. Heart regression and arrhythmogenic propensity. Am J 1994; 71: 268–273. Hypertens 1998; 11: 640–648. 50 Matsumura K, Takata Y, Ansai T, Awano S, Takehara T, 65 Gottdiener JS, Reda DJ, Massie BM, Materson BJ, Fujii K et al. Association of QT interval with blood Williams DW, Anderson RJ. Effect of single-drug pressure in 80-year-old subjects. Hypertens Res 2004; therapy on reduction of left ventricular mass in mild 27: 387–391. to moderate hypertension: comparison of six antihy- 51 Tribulova N, Okruhlicova L, Novakova S, Pancza D, pertensive agents. Circulation 1997; 95: 2007–2014. Bernatova I, Pechanova O et al. Hypertension-related 66 Weber KT, Brilla CG, Janicki JS. Myocardial fibrosis: intermyocyte junction remodelling is associated functional significance and regulatory factors. Cardiovasc with a higher incidence of low-K(+)-induced lethal Res 1993; 27: 341–348. arrhythmias in isolated rat heart. Exp Physiol 2002; 87: 67 Wachtell K, Bella JN, Rokkedal J, Palmieri V, 195–205. Papademetriou V, Dahlof B et al. Change in diastolic 52 Zaugg CE, Wu ST, Lee RJ, Wikman-Coffelt J, Parmley left ventricular filling after one year of antihyperten- WW. Intracellular Ca2+ handling and vulnerability to sive treatment: the Losartan Intervention For Endpoint ventricular fibrillation in spontaneously hypertensive Reduction in Hypertension (LIFE) Study. Circulation rats. Hypertension 1997; 30: 461–467. 2002; 105: 1071–1076. 53 Ohtsuka S, Kakihana M, Watanabe H, Enomoto T, 68 Shi Y, Li D, Tardif JC, Nattel S. Enalapril effects on Ajisaka R, Sugishita Y. Alterations in left ventricular atrial remodeling and atrial fibrillation in experimental wall stress and coronary circulation in patients with congestive heart failure. Cardiovasc Res 2002; 54: isolated systolic hypertension. J Hypertens 1996; 14: 456–461. 1349–1355. 69 Choudhury A, Varughese GI, Lip GY. Targeting the 54 Burke AP, Farb A, Liang YH, Smialek J, Virmani R. renin–angiotensin–aldosterone-system in atrial fibril- Effect of hypertension and cardiac hypertrophy on lation: a shift from electrical to structural therapy? coronary artery morphology in sudden cardiac death. Expert Opin Pharmacother 2005; 6: 2193–2207. Circulation 1996; 94: 3138–3145. 70 L’Allier PL, Ducharme A, Keller PF, Yu H, Guertin MC, 55 Erdogan D, Yildirim I, Ciftci O, Ozer I, Caliskan M, Tardif JC. Angiotensin-converting enzyme inhibition Gullu H et al. Effects of normal blood pressure, in hypertensive patients is associated with a reduction prehypertension, and hypertension on coronary micro- in the occurrence of atrial fibrillation. J Am Coll vascular function. Circulation 2007; 115: 593–599. Cardiol 2004; 44: 159–164. 56 Bishop JE, Lindahl G. Regulation of cardiovascular 71 Schmieder R, Kjeldsen SE, Julius S, McInnes GT, collagen synthesis by mechanical load. Cardiovasc Res Zanchetti A, Hua T et al. Reduced incidence of new 1999; 42: 27–44. onset atrial fibrillation with angiotensin II receptor 57 Mu¨ ller-Brunotte R, Kahan T, Lo´pez B, Edner M, blockade: the VALUE trial. J Hypertens 2006; 24: S3. Gonza´lez A, Dı´ez J et al. Myocardial fibrosis and 72 Okin PM, Wachtell K, Devereux RB, Harris KE, Jern S, diastolic dysfunction in patients with hypertension: Kjeldsen SE et al. Regression of electrocardiographic results from the Swedish Irbesartan Left Ventricular left ventricular hypertrophy and decreased incidence Hypertrophy Investigation versus Atenolol (SILVHIA). of new-onset atrial fibrillation in patients with hyper- J Hypertens 2007; 25: 1958–1966. tension. JAMA 2006; 296: 1242–1248. 58 Mancia G, Chalmers G, Ferrari A, Julius S, Saruta T, 73 Wachtell K, Lehto M, Gerdts E, Olsen MH, Hornestam Weber M (eds). Etiology and pathophysiology of B, Dahlof B et al. Angiotensin II receptor blockade hypertension: neural factors. Manual of Hypertension reduces new-onset atrial fibrillation and subsequent 2002. stroke compared to atenolol: the Losartan Intervention 59 Schwartz PJ, La Rovere MT, Vanoli E. Autonomic For End Point Reduction in Hypertension (LIFE) study. nervous system and sudden cardiac death. J Am Coll Cardiol 2005; 45: 712–719. Experimental basis and clinical observations for post- 74 Fogari R, Mugellini A, Destro M, Corradi L, Zoppi A, myocardial infarction risk stratification. Circulation Fogari E et al. Losartan and prevention of atrial 1992; 85: 177–197. fibrillation recurrence in hypertensive patients. 60 Barron HV, Lesh MD. Autonomic nervous system and J Cardiovasc Pharmacol 2006; 47: 46–50. sudden cardiac death. J Am Coll Cardiol 1996; 27: 75 Healey JS, Baranchuk A, Crystal E, Morillo CA, 1053–1060. Garfinkle M, Yusuf S et al. Prevention of atrial

Journal of Human Hypertension Hypertension and cardiac arrhythmias K-H Yiu and H-F Tse 388 fibrillation with angiotensin-converting enzyme 83 Carlsson J, Miketic S, Windeler J, Cuneo A, Haun S, inhibitors and angiotensin receptor blockers: a meta- Micus S et al. Randomized trial of rate-control versus analysis. J Am Coll Cardiol 2005; 45: 1832–1839. rhythm-control in persistent atrial fibrillation: the 76 Disertori M, Latini R, Maggioni AP, Delise P, Di Strategies of Treatment of Atrial Fibrillation (STAF) Pasquale G, Franzosi MG et al. Rationale and design study. J Am Coll Cardiol 2003; 41: 1690–1696. of the GISSI-Atrial Fibrillation Trial: a randomized, 84 Hohnloser SH, Kuck KH, Lilienthal J. Rhythm or rate prospective, multicentre study on the use of valsartan, control in atrial fibrillation—Pharmacological Inter- an angiotensin II AT1-receptor blocker, in the preven- vention in Atrial Fibrillation (PIAF): a randomised tion of atrial fibrillation recurrence. J Cardiovasc Med trial. Lancet 2000; 356: 1789–1794. (Hagerstown) 2006; 7: 29–38. 85 Lip GY, Tse HF. Management of atrial fibrillation. 77 Irbesartan for the prevention of atrial arrhythmias and Lancet 2007; 370: 604–618. cardiac electrical remodeling in patients with hyper- 86 Yurenev AP, Dyakonova HG, Novikov ID, Vitols A, tension and permanent pacemakers. NCT002255667; Pahl L, Haynemann G et al. Management of essential 002252006: http://www.clinicaltrials.gov/. hypertension in patients with different degrees of 78 The ASC. Rationale and design of ACTIVE: the left ventricular hypertrophy. Multicenter trial. Am J atrial fibrillation clopidogrel trial with irbesartan for Hypertens 1992; 5: 182S–189S. prevention of vascular events. Am Heart J 2006; 151: 87 Wachtell K, Okin PM, Olsen MH, Dahlof B, 1187–1193. Devereux RB, Ibsen H et al. Regression of electro- 79 Yamashita T, Ogawa S, Aizawa Y, Atarashi H, Inoue H, cardiographic left ventricular hypertrophy during Ohe T et al. Randomized study of angiotensin II antihypertensive therapy and reduction in sudden type 1 receptor blocker vs dihydropiridine calcium cardiac death: the LIFE Study. Circulation 2007; 116: antagonist for the treatment of paroxysmal atrial 700–705. fibrillation in patients with hypertension. Circ J 2006; 88 Hjalmarson A. Effects of beta blockade on sudden 70: 1318–1321. cardiac death during acute myocardial infarction 80 Teo K, Yusuf S, Sleight P, Anderson C, Mookadam F, and the postinfarction period. Am J Cardiol 1997; 80: Ramos B et al. Rationale, design, and baseline 35J–39J. characteristics of 2 large, simple, randomized trials 89 Egan BM, Basile J, Chilton RJ, Cohen JD. Cardioprotec- evaluating telmisartan, ramipril, and their combina- tion: the role of beta-blocker therapy. J Clin Hypertens tion in high-risk patients: the Ongoing Telmisartan (Greenwich) 2005; 7: 409–416. Alone and in Combination with Ramipril Global 90 Hoes AW, Grobbee DE, Lubsen J. Sudden cardiac death Endpoint Trial/Telmisartan Randomized Assessment in patients with hypertension. An association Study in ACE Intolerant Subjects with Cardiovascular with diuretics and beta-blockers? Drug Saf 1997; 16: Disease (ONTARGET/TRANSCEND) trials. Am Heart 233–241. J 2004; 148: 52–61. 91 Siscovick DS, Raghunathan TE, Psaty BM, Koepsell 81 Wyse DG, Waldo AL, DiMarco JP, Domanski MJ, TD, Wicklund KG, Lin X et al. Diuretic therapy for Rosenberg Y, Schron EB et al. A comparison of rate hypertension and the risk of primary cardiac arrest. control and rhythm control in patients with atrial N Engl J Med 1994; 330: 1852–1857. fibrillation. N Engl J Med 2002; 347: 1825–1833. 92 Ogihara T, Nakao K, Fukui T, Fukiyama K, Ueshima K, 82 Van Gelder IC, Hagens VE, Bosker HA, Kingma JH, Oba K et al. Effects of candesartan compared Kamp O, Kingma T et al. A comparison of rate control with amlodipine in hypertensive patients with high and rhythm control in patients with recurrent cardiovascular risks. Candesartan Antihypertensive persistent atrial fibrillation. N Engl J Med 2002; 347: Survival Evaluation in Japan Trial. Hypertension 1834–1840. 2008; 51: 393–398.

Journal of Human Hypertension