Heart and Metabolism

Clinical relevance of left ventricular hypertrophy regression

Christian G. Brilla
Division of Cardiology, Philipps University, Marburg, Germany
Correspondence: Prof. Dr Christian G Brilla, Division of Cardiology, Philipps University of Marburg, Baldingerstr, D-35033 Marburg, Germany. Tel: +49-6421-2864980, fax: +49-6421-2868954.

Introduction
According to the Framingham Heart Study, left ventricular hypertrophy (LVH) is a primary risk factor associated with the appearance of all major cardiovascular events, including the development of heart failure.[1,2] In the presence of LVH determined by electrocardiographic criteria, the risk of developing heart failure is increased, for men and women alike, 6- to 18-fold. A prospective, longitudinal community-based cohort study, similar to that of the Framingham Heart Study, consisted of 459 subjects aged 75–85 years.[3] Electrocardiograms obtained at baseline and on an annual basis over 10 years revealed that 9.2% of subjects had LVH on electrocardiogram at baseline and a mortality rate of 11.7/100 person-years versus 4.9/100 person-years for subjects without LVH. Therefore, the question arises whether pharmacologically mediated regression of LVH would improve patients’ morbidity and mortality. Before addressing this important clinical question, the growth process leading to the development of LVH needs to be characterized.

Cell population of the myocardium: myocyte and non-myocyte cells
The myocardium is composed of different cells. Cardiac myocytes are the largest of these cells and occupy 75% of the structural space of the myocardium. These parenchymal cells, however, comprise only one-third of all cardiac cells.[4] Two-thirds of all myocardial cells are non-myocyte cells: endothelial cells, vascular smooth muscle cells, cardiac fibroblasts including pericytes, myofibroblasts and interstitial fibroblasts, which are responsible for both producing and degrading the structural components of the myocardial extracellular matrix including types-I and -III collagen, the major fibrillar collagens of the myocardium.[5,6] These collagens are involved in the interstitial and perivascular fibrosis of the myocardium[7] and the replacement scarring that follows cell death.[8] Finally, macrophages and mast cells involved in inflammatory processes during a wound healing response to any injury, i.e. hypertension, myocardial infarction, or infective heart disease, are present in the myocardium.

Myocardial growth during cardiovascular disease
Myocardial growth that leads to LVH involves all cellular compartments, i.e. myocyte and non-myocyte cells. It is the hypertrophic growth of cardiac myocytes with increased expression of contractile proteins that accounts for any increment in myocardial mass and, therefore, for the development of LVH. This is the case for all etiologies of LVH, including arterial hypertension, myocardial infarction with hypertrophy of the remote viable myocardium, valvular heart disease with chronic pressure and/or volume overload of the left ventricle and dilated cardiomyopathy. The major physiologic stimulus for myocyte hypertrophy is the hemodynamic load.[9] Why the growth of these muscle cells would prove either adaptive, as in the athlete’s heart, or pathologic, as in hypertensive heart disease, has remained an enigma. One reason could be the potential occurrence of pathologic myocyte phenotypes with, for example, decreased density of sarcoplasmic Ca2+ pumps leading to reduced pump function of the left ventricle due to impaired intracellular Ca2+ cycling.[10]
Another line of reasoning has suggested that it is not the hypertrophic growth of myocytes that is responsible for pathologic LVH.[11] Instead, pathologic LVH has been attributed to the growth and altered behavior of non-myocyte cells. Medial wall thickening of intramyocardial resistance vessels due to proliferation and/or load-dependent hypertrophy of vascular smooth muscle cells may occur and leads to a decrease in coronary reserve.[12] Indeed, in patients with hypertensive heart disease suffering from angina pectoris, normal coronary angiograms but abnormal medial wall thickening of intramyocardial resistance vessels examined by endomyocardial biopsies have been reported.[13] Non-myocyte cells also include cardiac fibroblasts, the growth and enhanced collagen synthesis and/or suppressed collagen degradation of which are responsible for the accumulation of collagen within the cardiac interstitium, where type-I collagen is the major fibrillar component of the extracellular matrix that accounts for myocardial stiffness.

Myocardial collagen matrix remodeling and left ventricular function
The structural remodeling and accumulation of fibrillar collagen that occurs in different disease states have been examined in various species[7,14] and the following identified: (1) signals mediating fibroblast and cardiac myocyte growth are largely independent of one another;[12,14] (2) fibrous tissue accumulation occurs as either a reactive or a reparative process,[15] based on whether or not there is parenchymal cell loss (i.e. myocyte necrosis); and (3) activation of the renin-angiotensin-aldosterone system (RAAS) with elevations in circulating and/or local angiotensin II and aldosterone is related to the abnormal fibrous tissue response in acquired or genetic arterial hypertension and in congestive heart failure.[12,14] The subsequent remodeling of myocardial structure with progressive reactive fibrosis alters its mechanical behavior. During early remodeling, diastolic stiffness of the left ventricle is increased[16] while systolic function (e.g. ejection fraction) is preserved. During late remodeling, in addition to diastolic dysfunction at rest with elevated left ventricular filling pressure, systolic dysfunction with chamber dilatation and reduced ejection fraction appears.[17] In patients with hypertensive heart disease due to primary hypertension, myocardial stiffness measured by the stiffness constant correlates with collagen volume fraction but not with myocyte hypertrophy of endomyocardial left ventricular biopsies. Thus, the accumulation of fibrillar collagen is a major determinant of myocardial stiffness and pump dysfunction and its progressive accumulation accounts for ventricular dysfunction that first appears during diastole and subsequently involves systole.
Collagen concentration remains normal in the hypertrophied myocardium seen in low-renin states, i.e. in arteriovenous fistula, atrial septal defect or chronic anemia, as well as with thyroxine or growth hormone administration.[18–21] This is also the case when arterial hypertension is created by infrarenal aorta banding, when renal perfusion is not impaired and therefore RAAS is not activated.[14] Therefore, mechanical factors would not appear to account for the disproportionate accumulation of collagen that occurs with LVH in some conditions but not in others, despite comparable elevations in wall stress due to ventricular pressure or volume overload. Instead, in cultured adult rat cardiac fibroblasts, the effector hormones of RAAS, angiotensin II and aldosterone, have shown to directly stimulate collagen synthesis under serum-free conditions.[22] In addition, angiotensin II inhibits the activity of matrix metalloproteinase I, which is the key enzyme for interstitial collagen degradation.[22] The net effect is excessive collagen accumulation. Furthermore, chronic administration of aldosterone in uninephrectomized animals receiving enhanced dietary sodium, is associated with increased cardiac expression of types-I and -III collagen mRNAs[23] and fibrosis in the myocardium of the right and left ventricles[14] while fibrosis has also been found in the pancreas, adrenals and other organs.[24] When subhypotensive doses of the angiotensin converting enzyme (ACE) inhibitor lisinopril or the aldosterone antagonist spironolactone were used in genetic or aldosterone/salt hypertension, respectively, myocardial fibrosis could be either reversed or prevented, while myocyte hypertrophy remained until larger, i.e. antihypertensive doses, were applied.[12,25] Thus, trophic factors which mediate myocyte and non-myocyte cell growth in the myocardium can be independent of one another (Table 1).

Table 1. Myocardial fibrosis in various experimental rat models with and without systemic hypertension and
cardiac hypertrophy.

Homogenous versus heterogenous regression of LVH
There is now broad evidence that regression of LVH can be achieved. In various meta-analyses of antihypertensive drugs it appeared that ACE inhibitors are particularly capable of reversing LVH while other antihypertensive drugs such as diuretics may work as well.[26] In prospective clinical trials, such as the Treatment of Mild Hypertension Study, diuretics were even more powerful at reversing myocyte hypertrophy that determines LVH.[27,28] Since the remodeling of the myocardium during various disease states is a complex process involving all myocardial tissue and cellular compartments, any approach to reversing LVH appears to be a complicated matter. For instance, it would not be meaningful to reverse myocyte hypertrophy while fibrotic tissue remains in the myocardium. The functional outcome would be even worse. Therefore, we would need to differentiate the effects of various antihypertensive agents on myocardial tissue because any disproportionate growth between myocyte and non-myocyte cells would set the stage for abnormal myocardial function.

Table 2. Antihypertensive drugs and myocardial
structure.

Based on the diverse effects of antihypertensive agents on the various tissue compartments in the heart, three classes of antihypertensive agents may be considered (Table 2): (1) drugs with no evidence of reversing LVH and fibrosis (direct vasodilators);[29] (2) drugs with clear effects on LVH, i.e. regression of myocyte hypertrophy (diuretics, a- and b-adrenergic receptor antagonists and verapamil);[28,30,31] and (3) agents with proven effects on regression of LVH and fibrosis (ACE inhibitors, angiotensin II type 1 receptor or aldosterone antagonists, dihydropyridine Ca2+ channel blockers and centrally acting antiadrenergic agents).[25,28,30,32,33] Removal of reactive myocardial fibrosis represents a means by which myocardial failure due to collagen accumulation would be reversible.

Reversal of LVH and patient prognosis
Several medium-sized clinical trials have been performed to answer the question whether regression of LVH is associated with improved clinical outcome. In 430 patients with primary hypertension and during an observation period of 1217 patient-years, those with an echocardiographically determined decrease in left ventricular mass during follow-up showed a significantly (P < 0.03) decreased rate of cardiovascular morbid events (1.8/100 patient-years) compared with those whose left ventricular mass increased (3.0/100 patient-years) (Figure 1).[34]

Figure 1. Cardiovascular (CV) morbid events per 100 person-years in 430 patients with primary hypertension during antihypertensive treatment;34 a significant clinical improvement, i.e. reduction in CV events, was found in all patients if left ventricular (LV) mass could be reduced compared with patients where no change occurred (yellow), which was even more impressive in patients with left ventricular mass >125 g/m2 (26% of subjects) at baseline (red).

In a 10-year follow-up study of 151 patients with uncomplicated arterial hypertension, the incidence of cardiovascular events was significantly greater (P < 0.01) in patients without an echocardiographically determined reduction in left ventricular mass (relative risk 3.5) than in patients with LVH regression (relative risk 1.4) after adjusting for traditional risk factors.[35] In the Bronx Longitudinal Aging Study,3 subjects in whom the electrocardiographic LVH pattern disappeared over time had fewer cardiovascular mortality and morbidity events than those with persistent or newly developed LVH (Figure 2).

Figure 2. Total mortality rate per 100 person-years in the Bronx Longitudinal Aging Study3 in which 459 elderly subjects (mean age 79 years) were followed over 10 years in a prospective trial. In the presence of left ventricular hypertrophy (LVH) at baseline, mortality was significantly increased compared with subjects with no electrocardiographic evidence of LVH. After 10 years, subjects without LVH showed a significantly improved prognosis compared with those in whom LVH developed.

Persistent LVH from baseline was an independent predictor of myocardial infarction, overall cardiovascular disease and total mortality. In the Framingham Heart Study, 524 subjects with electrocardiographic evidence of LVH were free of cardiovascular disease at baseline. During follow-up there were 269 new cardiovascular events. Subjects with a serial decline in voltage were at lower risk for cardiovascular disease than were those with no serial change.[36]
These findings strongly indicate that the lack of decrease in left ventricular mass following antihypertensive treatment is associated with a higher risk for cardiovascular events that is markedly improved by regression of LVH. Such improvement in patient prognosis has been achieved even with unselected antihypertensive agents. Based on the diverse myocardial tissue findings following treatment with various antihypertensive drugs we may speculate that patient prognosis with LVH can be even further improved by choosing agents which lead to a homogenous reversal of LVH associated with improved cardiac function, termed cardioreparation.[28,32] Various prospective, randomized clinical trials are underway (PRESERVE, LIFE) to prove whether ACE inhibitors or angiotensin II type 1 receptor antagonists are particularly capable of improving the prognosis of patients with LVH.

REFERENCES

1: J Cardiovasc Pharmacol 1987;10 Suppl 6:S135-40