Evaluation of Subclinical Target Organ Damage for Risk Assessment and Treatment in the Hypertensive Patients: Left Ventricular Hypertrophy

Enrico Agabiti-Rosei, Maria Lorenza Muiesan, and Massimo Salvetti Internal Medicine, University of Brescia, Brescia, Italy

At some point in the natural history of hypertension, the compensatory increase in left ventricular mass ceases to be beneficial. Left ventricular hypertrophy (LVH) becomes a preclinical disease and an independent risk factor for congestive heart failure, ischemic heart disease, arrhythmia, sudden death, and stroke. In addition to elevated BP, several mechanisms are involved, including body size, age, gender, race, fibrogenic cytokines, and neurohumoral factors, notably angiotensin II, which favor interstitial collagen deposition and perivascular fibrosis. These tissue changes are responsible for the insidious contractile dysfunction that is associated with LVH, consequent to decreased coronary reserve and altered diastolic ventricular filling and relaxation. The cardinal investigations are echocardiography and electrocardiography. All antihypertensive drugs regress LVH, notably those that act on the renin-angiotensin-aldosterone system, which also could target the detrimental tissue changes. Regression enhances systolic midwall performance, normalizes autonomic function, and restores coronary reserve. The resulting improvement in prognosis has enshrined the detection, prevention, and reversal of LVH in the current guidelines of hypertension management. J Am Soc Nephrol 17: S104–S108, 2006. doi: 10.1681/ASN.2005121336

t some point in the natural history of hypertension, hemodynamic load, echocardiographic LVM could be assessed the compensatory increase in left ventricular mass in the individual patient as deviation from the value that is A (LVM) is not beneficial anymore. In fact, it becomes a appropriate for a given cardiac workload, corrected for gender preclinical disease and an independent risk factor for conges- and body size. LVM that overcompensates for hemodynamic tive heart failure, ischemic heart disease, arrhythmia, sudden load tends to cluster with metabolic risk factors and is associ- death, and stroke (1). ated with high cardiovascular risk (6,7). The definition and the clinical evaluation of “inappropriate” LVM require further Determinants of Hypertensive Left study. Ventricular Hypertrophy Among the main nonhemodynamic factors that may contrib- The high prevalence of LV hypertrophy (LVH) in hyperten- ute to the development of LVH, a variety of neurohumoral and sion reflects the increased afterload imposed on the left ventri- growth factors (e.g., catecholamines, angiotensin II [AngII], al- cle, although other important determinants are demographic dosterone, endothelins) are included. The effect of sympathetic characteristics (e.g., age, gender, race), neurohumoral and nervous system activity is evident in experimental models but growth factors, and underlying genetic factors. Hypertension is less clear-cut in humans: in pheochromocytoma LVH preva- the fundamental trigger to the sequence of biologic events that lence is relatively low and LVM seems to increase proportion- lead to the development of LVH. LVM is more closely related ately to BP, then in essential hypertension, LVH is associated to average 24-h BP (2–4). Volume load, inotropy, and arterial with altered autonomic activity and a blunted response to compliance also are important determinants of the develop- ␤-adrenoceptor stimulation (8–10). Experimental studies also ment and the degree of LVH. Other contributing factors include highlighted the role of the renin-angiotensin-aldosterone sys- stage of hypertensive disease, genetics, demographic factors, tem (RAAS) in mediating LVH (11,12). AngII induces hyper- comorbid diseases (e.g., diabetes, obesity, coronary artery dis- trophy and hyperplasia in myocytes and vascular smooth mus- ease), possibly mediated via cardiac load. Obesity, which com- cle cells and may regulate collagen synthesis. Excess AngII pounds hemodynamic load independent of a clear-cut increase production may regulate the expression of fibrogenic cytokine in BP, is a major determinant of LVM, because it is associated TGF-␤1 and favor perivascular and interstitial fibrosis. In ad- with increased plasma volume and cardiac output (5). It has dition, AngII may interfere with the process of collagen degra- been suggested that considering these measurable factors and dation, modulating the activity of metalloproteinases (a family of zinc-containing proteins that include stromelysins, collag- enases, and gelatinases) and their inhibitors (12). Aldosterone Address correspondence to: Prof. Enrico Agabiti-Rosei, Internal Medicine, Med- also may stimulate extracellular collagen deposition and myo- ical and Surgical Sciences, University of Brescia, c/o 2a Medicina Spedali Civili di Brescia, Piazza Spedali Civili 1, 25100 Brescia, Italy. Phone: ϩ30-396044; Fax: cardial fibrosis. ϩ30-3388147; E-mail: [email protected] The pathogenic role of the RAAS in the development of

Copyright © 2006 by the American Society of Nephrology ISSN: 1046-6673/1704-0104 J Am Soc Nephrol 17: S104–S108, 2006 Pathophysiology and Treatment of Hypertensive LVH S105 hypertensive LVH requires confirmation, although LVM is sig- more expensive or time-consuming and are still of limited nificantly increased in renovascular hypertension and primary availability. aldosteronism compared with essential hypertension (13,14). Methods have been developed to quantify tissue composi- Increased activity of the RAAS and sympathetic nervous sys- tion. Studies in animals and humans have shown that LV tem, hypervolemia, and anemia all may influence the increase acoustic properties under physiologic and pathologic condi- of LVM in patients with renal disease. tions are influenced by several tissue components (myocar- Hypertensive LVH often is associated with insulin resistance dium, contractile and elastic tissue, collagen and inelastic tis- and high insulin levels (15). The involvement of IGF- I could sue, arteries, veins, myocytes, and sarcomeres). Results with clarify the link among obesity, BP elevation, LVH, and the videodensitometry and integrated backscatter to characterize metabolic syndrome. Leptin is another possible neuroendocrine tissue in several diseases that are associated with abnormal determinant. LVH in an animal model of leptin deficiency (the myocardial tissue, including hypertensive LVH and diabetes, ob/ob mouse) reversed rapidly in response to exogenous leptin, indicate that these techniques can complement clinical evalua- therefore indicating the involvement of myocardial leptin re- tion by revealing preclinical end-organ damage (25,26). Further ceptors in cardiac remodeling (16). Other major metabolic car- reproducibility and feasibility studies are required to assess the diovascular risk factors, notably hypercholesterolemia and hy- clinical applications of these techniques. perglycemia, also may have an influence on LVM and the prevalence of LVH (17). Prognostic Significance of LVH Several studies have suggested that approximately 30% of Whether assessed by ECG or echocardiography, LVH is a LVM variance is genetically determined (18). Studies of genetic well-documented harbinger of morbidity and mortality. In sev- influence on LVM have focused mainly and are ongoing on the eral studies, the adjusted risk for cardiovascular morbidity role of candidate genes, including those that are related to the associated with baseline LVH ranges from 1.5 to 3.5 with a RAAS, ␣- and ␤-adrenoceptors, and components of the signal weighted risk ratio of 2.3 for all studies combined (27). Con- transduction mechanisms involved in cardiac hypertrophy. centric hypertrophy seems to carry the highest risk and eccen- tric hypertrophy an intermediate risk, whereas concentric re- modeling is probably associated with a smaller, albeit Methods of Assessing LVH noteworthy, risk. As recommended by the European Society of Hypertension The structural remodeling of cardiomyocytes, nonmyocytes, and European Society of Cardiology (19), LVH is diagnosed and fibroblasts that occurs in cardiac hypertrophy contributes most commonly using electrocardiography (ECG) and M-mode to perivascular fibrosis, initially around intramural coronary and two-dimensional echocardiography. New ECG criteria in arteries and thereafter in the interstitial space, leading to pro- addition to repolarization abnormalities and increased voltage gressive abnormalities of diastolic ventricular filling and relax- have been proposed, the Cornell method probably being the ation, systolic dysfunction, arrhythmias, and conduction dis- most sensitive (20). The ECG also can be used to detect patterns turbances, thereby greatly compounding the risk that is of ventricular overload (“strain”) or ischemia, which indicate associated with LVH (1). higher cardiovascular risk. ECG and echocardiographic LVH The resulting pathophysiologic and clinical changes that ac- both predict mortality independent of each other and other count for increased risk in hypertensive LVH include both cardiovascular risk factors, thus conveying, at least in part, diastolic and systolic dysfunction. LVH and failure are fre- different prognostic information (21,22). Echocardiography has quently associated with coronary artery disease, and hyperten- some advantages because it provides comprehensive wall, sion is a major risk factor for coronary atherosclerosis. In ECG chamber, and LVM measures, together with systolic and dia- LVH, use of a “definite LVH” pattern that comprises ST-seg- stolic performance indices, while remaining reasonably cheap, ment and T-wave abnormalities was strongly associated with widely available, and wholly noninvasive. The relation be- an increased incidence of acute myocardial infarction and sud- tween LVM and cardiovascular risk is continuous, although the den death (3–6), suggesting that altered repolarization reflects threshold of 125 g/m2 for men and 110 g/m2 for women reduced coronary perfusion. currently is used most widely for conservative estimates of LVH is associated with structural and functional changes in LVH, according to the recent European Society of Hypertensio- both large (28,29) and small (14,30,31) arteries. These structural n–European Society of Cardiology guidelines (19). Despite its changes are particularly evident in concentric LVH. The asso- advantages, echocardiography may entail a possible technical ciation between LVH and extracranial carotid atherosclerosis error as a result of the method itself, the quality of the exami- also might explain the increased risk for cerebrovascular nation, or observer inexperience. It has been established that a events. The vascular changes that consistently are observed in biologic significance can be attributed to changes in LVM that LVH are also responsible for the reduced coronary reserve. exceed 10 to 15% (23). Concomitant atherosclerosis in epicardial coronary vessels and Echocardiography also is useful in assessing the different structural alterations and rarefaction of small coronary vessels types of LV geometric adaptation to increased cardiac load (24), (32) limit blood supply when oxygen demand is increased. evaluating the increase in mass and/or in relative wall thick- Compensatory angiogenesis is inadequate during the develop- ness. More accurate and sophisticated techniques, such as mag- ment of adult LVH. Decreased subendocardial coronary perfu- netic resonance imaging or cine computerized tomography, are sion leads to myocyte necrosis and reparative fibrosis, favoring S106 Journal of the American Society of Nephrology J Am Soc Nephrol 17: S104–S108, 2006 the progression to heart failure. Other extravascular mecha- addition, most major intervention trials that have compared the nisms that compound the impairment of coronary reserve in- effects of single antihypertensive drugs on LVM in fact largely clude changes of oxygen demand related to wall tension, heart have been comparisons of combination therapies, because most rate, and contractility. Functional changes further weaken the patients were taking more than one drug. vasodilator response of the coronary microcirculation. In fact, There is increasing interest in the effect of antihypertensive endothelial dysfunction precedes morphologic changes in the treatment on myocardial tissue composition, particularly on vascular wall and triggers remodeling. In summary, LVH is a perivascular and interstitial fibrous tissue. Recent experimental state of potential or actual myocardial ischemia. and human evidence suggests that angiotensin-converting en- There is a predisposition to ventricular arrhythmia in hyper- zyme inhibitors and AngII antagonists are particularly effective tensive LVH, explaining the risk for sudden death. Impaired in inducing regression of myocardial fibrosis. ventricular filling, left atrial enlargement, and slowing of atrial conduction velocity all encourage atrial fibrillation, increasing Clinical and Prognostic Significance of LVH the risk for thromboembolism. In addition, in patients with Regression LVH (and abnormal LV geometry), higher urinary albumin Because LVH is such an important independent risk factor in excretion has been observed, suggesting that cardiac and glo- hypertension, there is consensus as to the desirability of its merular vascular damage may parallel, independent of the regression and prevention. Regression is associated with nu- hemodynamic load. Because hypertensive LVH is an indepen- merous benefits, such as improved systolic midwall perfor- dent risk factor for cardiovascular morbidity and mortality, the mance, normalized autonomic function, enhanced coronary re- possibility of reversal or even prevention by lowering BP and serve, and, possibly, improved diastolic filling and decreased modifying other pathogenetic factors is a major goal of antihy- ventricular arrhythmia. It remains to be assessed extensively pertensive therapy. whether LVM changes may improve parallel vascular and/or renal damage modifications. LVH Regression by Antihypertensive The improved prognosis associated with LVH regression has Treatment been demonstrated in several studies using ECG measures. The LVM can be decreased by nonpharmacologic intervention, large long-term Losartan Intervention for Endpoint (LIFE) notably weight loss, which is effective in obese hypertensive study showed that the greater regression of LVH with losartan patients independent of BP changes. As shown by the multi- was associated with fewer cardiovascular events (36). center Treatment of Mild Hypertension Study (TOMHS), life- Further observations using the more sensitive echocardio- style intervention may reduce BP significantly and decrease graphic technique have shown that patients who achieve LVH LVM substantially in 30% of patients. However, there is still no regression during follow-up are much less likely to experience hard evidence of an independent effect by dynamic exercise, morbid events as compared with those with persistence of LVH dietary sodium, or alcohol restriction. (odds ratio 0.41) (37). In the echocardiographic substudy of the Multiple studies have shown that BP reduction reverses LIFE trial that included 960 patients who were followed for Ͼ4 LVH. The main determinants are treatment duration and de- yr, the better prognosis that was associated with the significant gree of BP reduction, in particular of average 24-h BP; subse- decrease in LVM from baseline to end of study was due mainly quent evidence also has shown the importance of homogeneity, to a decrease in the incidence of stroke (38). or minimal daily fluctuation, in BP control, as expressed in the These cumulative findings highlight the prognostic value of “smoothness index” (33,34). the LVM response to treatment. BP was not significantly asso- Because BP is not the sole determinant of LVH and fibrosis, ciated with the incidence of cardiovascular events in these the differing response of LVM to various classes of antihyper- studies, although it cannot be excluded that the changes that tensive drugs can be ascribed to interference with nonhemody- were observed in the LVM index at least partially reflected BP namic factors such as the RAAS and sympathetic nervous sys- control. tem. Several meta-analyses have been conducted, including the Whereas complete regression significantly reduces cardio- main studies demonstrating reversal of echocardiographic vascular risk, an increase in echocardiographic LVM during LVH using various antihypertensive drugs, and they have antihypertensive therapy or a failure to decrease confers a shown that baseline LVM and the degree of BP reduction are worse prognosis. In addition, the response of LV geometry to the main determinants of LVH regression; in addition angio- treatment may have prognostic significance, independent of tensin-converting enzyme inhibitors, AngII receptor blockers, changes of LVM (39). and calcium channel blockers have been more effective than ␤ blockers and diuretics given the same decrease in BP (35). Future Goals Large, randomized, blinded studies that have compared two or Focuses of future interest will include the biochemistry of the more different antihypertensive drugs have provided further adaptive changes in energy metabolism and contractile pro- data, confirming, at least in large part, the results of meta- teins, notably the role of transmitters and transductional fac- analyses. tors, as well as the timing of these responses to BP changes, However, it should be kept in mind that interdrug differ- neurohumoral activation, and the development of structural ences tend to fade with time, because treatment duration is alterations in other organs. Techniques such as tissue charac- associated with progressive BP control and decrease in LVM. In terization and noninvasive quantitative analysis of coronary J Am Soc Nephrol 17: S104–S108, 2006 Pathophysiology and Treatment of Hypertensive LVH S107 flow will describe the respective contributions of perivascular velli G, Agabiti-Rosei E: Relations between cardiac and and intraventricular fibrosis and myocardial ischemia to the vascular structure in patients with primary and secondary mechanisms of LVH risk and, hopefully, indicate ways in hypertension. J Am Coll Cardiol 32: 985–992, 1998 which these advances can be translated into the individual 14. Rossi GP, Sacchetto A, Pavan E, Palatini P, Graniero GR, patient benefit. However, we already know more than enough Canali C, Pessina AC: Remodeling of the left ventricle in primary aldosteronism due to Conn’s adenoma. Circulation to realize that a major goal in the management of hypertension 195: 1471–1478, 1997 is the detection, prevention, and reversal of LVH. 15. Verdecchia P, Reboldi G, Schillaci G, Borgioni C, Ciucci A, Telera MP, Santeusanio F, Porcellati C, Brunetti P: Circu- lating insulin and insulin growth factor-1 are independent References determinants of left ventricular mass and geometry in 1. Frohlich ED: Risk mechanisms in hypertensive heart dis- essential hypertension. Circulation 100: 1802–1807, 1999 ease. Hypertension 34: 782–789, 1999 16. Barouch LA, Berkowitz DE, Harrison RW, O’Donnell CP, 2. Parati G, Pomidossi G, Albini E, Malaspina D, Mancia G: Hare JM: Disruption of lectin signaling contributes to car- Relationship of 24-hour blood pressure mean and variabil- diac hypertrophy independently of body weight in mice. ity to severity of target-organ damage in hypertension. Circulation 108: 754–759, 2003 J Hypertens 5: 93–98, 1987 17. Cuspidi C, Meani S, Fusi V, Severgnini B, Valerio C, Catini 3. Verdecchia P, Schillaci G, Guerrieri M, Gatteschi C, Ben- E, Leonetti G, Magrini F, Zanchetti A: Metabolic syndrome emio G, Boldrini F, Porcellati C: Circadian blood pressure and target organ damage in untreated essential hyperten- change and left ventricular hypertrophy in essential hyper- sives. J Hypertens 22: 1991–1998, 2004 tension. Circulation 81: 528–536, 1990 18. Post W, Larson M, Myers RH, Galderisi M, Levy D: Heri- 4. Rizzoni D, Muiesan ML, Montani G, Zulli R, Calebich S, tability of left ventricular mass. Hypertension 30: 1025–1028, Agabiti-Rosei E: Relationship between initial cardiovascu- 1997 lar structural changes and daytime and nighttime blood 19. Guidelines Committee: 2003 European Society of Hyper- pressure monitoring. Am J Hypertens 5: 180–186, 1992 tension–European Society of Cardiology guidelines for the 5. de Simone G, Devereux RB, Roman MJ, Alderman MH, management of arterial hypertension. J Hypertens 21: 1011– Laragh JH: Relation of obesity and gender to left ventric- 1053, 2003 ular hypertrophy in normotensive and hypertensive 20. Devereux RB, Roman MJ: Evaluation of cardiac and vas- adults. Hypertension 23: 600–606, 1994 cular structure and function by echocardiography and 6. de Simone G, Palmieri V, Koren M, Mensah G, Roman MJ, other non-invasive techniques. In: Hypertension: Pathophys- Devereux RB: Prognostic implications of the compensatory iology, Diagnosis and Management, New York, Raven Press, nature of left ventricular mass in arterial hypertension. 1995, pp 1969–1985 J Hypertens 19: 119–125, 2001 21. Sundstrom J, Lind L, Arnlov J, Zethelius B, Andren B, 7. Palmieri V, Wachtell K, Gerdts E, Bella JN, Papademetriou Lithell HO: Echocardiographic and electrocardiographic V, Tuxen C, Nieminen MS, Dahlof B, de Simone G, De- diagnoses of left ventricular hypertrophy predict mortality vereux RB: Left ventricular function and hemodynamic independently of each other in a population of elderly features of inappropriate left ventricular hypertrophy in men. Circulation 103: 2346–2351, 2001 patients with systemic hypertension: The LIFE study. Am 22. Okin P, Roman MJ, Lee ET, Galloway JM, Howard B, Heart J 141: 784–791, 2001 Devereux RB: Combined echocardiographic left ventricu- 8. Grassi G, Gianattasio C, Cleroux J, Cuspidi C, Sampieri L, lar hypertrophy and electrocardiographic ST depression Mancia G: Cardiopulmonary reflex before and after regres- improve prediction of mortality in American Indians. The sion of left ventricular hypertrophy in essential hyperten- Strong Heart Study. Hypertension 43: 769–774, 2004 sion. Hypertension 12: 227–237, 1988 23. de Simone G, Muiesan ML, Ganau A, Longhini C, Verdec- 9. Rizzoni D, Agabiti-Rosei E, Castellano M, Beschi M, Mon- chia P, Palmieri V, Agabiti-Rosei E, Mancia G: Reliability tani G, Pizzocolo G, Poiesi C, Rodella A, Agabiti-Rosei E: and limitations of measurement of echocardiographic mea- Antidiuretic hormone and atrial natriuretic peptide during surement of left ventricular mass for risk stratification and lower body negative or positive pressure in hypertensive follow-up in single patients: The RES trial. Working Group patients with and without left ventricular hypertrophy. on Heart and Hypertension of the Italian Society of Hy- Clin Exp Hypertens A 14: 717–732, 1992 pertension. Reliability of M-mode Echocardiographic Stud- 10. Trimarco B, De Luca N, Ricciardelli B, Rosiello G, Volpe M, ies. J Hypertens 17: 1960–1964, 1999 Condorelli G, Lembo G, Condorelli M: Cardiac function in 24. Ganau A, Devereux RB, Roman MJ, de Simone G, Picker- systemic hypertension before and after reversal of left ven- ing TG, Saba PS, Vargiu P, Simongini I, Laragh JH: Patterns tricular hypertrophy. Am J Cardiol 62: 745–750, 1988 of left ventricular hypertrophy and geometric remodeling 11. Schlaich MP, Schobel HP, Hilgers K, Schmieder RE: Impact in arterial hypertension. J Am Coll Cardiol 19: 1550–1558, of aldosterone on left ventricular structure and function in 1992 young normotensive and mildly hypertensive subjects. 25. Di Bello V, Pedrinelli R, Giorgi D, Bertini A, Talarico L, Am J Cardiol 85: 1199–1206, 2000 Caputo MT, Massimiliano B, Dell’Omo G, Paterni M, 12. Woessner JF: Matrix metalloproteinases and their inhibi- Giusti C: Ultrasonic videodensitometric analysis of two tors in connective tissue remodeling. FASEB J 5: 2145–2154, different models of left ventricular hypertrophy: Athlete’s 1991 heart and hypertension. Hypertension 29: 937–944, 1997 13. Rizzoni D, Muiesan ML, Porteri E, Salvetti M, Castellano 26. Ciulla M, Paliotti R, Hess DB, Tjahja E, Campbell SE, M, Bettoni G, Tiberio G, Giulini SM, Monteduro C, Gara- Magrini F, Weber KT: Echocardiographic patterns of myo- S108 Journal of the American Society of Nephrology J Am Soc Nephrol 17: S104–S108, 2006

cardial fibrosis in hypertensive patients: Endomyocardial of Blood Pressure and Lisinopril Evaluation. Circulation 95: biopsy versus ultrasonic tissue characterization. JAmSoc 1464–1470, 1997 Echocardiogr 10: 657–664, 1997 34. Parati G, Omboni S, Rizzoni D, Agabiti-Rosei E, Mancia G: 27. Vakili B, Okin P, Devereux RB: Prognostic implications of The smoothness index: A new reproducible and clinically left ventricular hypertrophy. Am Heart J 141: 334–341, 2001 relevant measure of the homogeneity of the blood pressure 28. Roman MJ, Pickering TG, Schwartz JE, Pini R, Devereux reduction with treatment for hypertension. J Hypertens 16: RB: Association of carotid atherosclerotic and left ventric- 1685–1693, 1998 ular hypertrophy. J Am Coll Cardiol 25: 83–90, 1995 35. Klingbeil A, Schneider M, Martus P, Messerli F, Schmieder 29. Muiesan ML, Pasini G, Salvetti M, Calebich S, Zulli R, R: A meta-analysis of the effects of treatment on left ven- Castellano M, Rizzoni D, Bettoni G, Cinelli A, Porteri E, tricular mass in essential hypertension. Am J Med 115: Corsetti V, Agabiti-Rosei E: Cardiac and vascular struc- 41–46, 2003 tural changes: Prevalence and relation to ambulatory blood 36. Okin PM, Devereux RB, Jern S, Kjeldsen SE, Julius S, pressure in a middle-aged general population in Northern Nieminen MS, Snapinn S, Harris KE, Aurup P, Edelman JM, Dahlof B; Losartan Intervention for Endpoint reduc- Italy. The Vobarno Study. Hypertension 27: 1046–1052, 1996 tion in hypertension Study Investigations: Regression of 30. Lucarini AR, Spessot M, Picano E, Marini C, Lattanzi F, electrocardiographic left ventricular hypertrophy by losar- Pedrinelli R, Salvetti A: Lack of correlation between car- tan versus atenolol. The Losartan Intervention For End- diac mass and arteriolar structural changes in mild-to- point reduction in hypertension. (LIFE) Study. Circulation moderate hypertension. J Hypertens 9: 1187–1191, 1991 108: 684–690, 2003 31. Niteberg A, Anthony I: Epicardial coronary arteries are not 37. Muiesan ML, Salvetti M, Rizzoni D, Castellano M, Donato adequately sized in hypertensive patients. J Am Coll Cardiol F, Agabiti-Rosei E: Association of change in left ventricular 27: 115–123, 1996 mass with prognosis during longterm antihypertensive 32. Rizzoni D, Palombo C, Porteri E, Muiesan ML, Kozakova treatment. J Hypertens 13: 1091–1097, 1995 M, La Canna G, Nardi M, Guelfi D, Salvetti M, Morizzo C, 38. Devereux RB, Wachtell K, Gerdts E, Boman K, Nieminen Vittone F, Rosei EA: Relationships between coronary flow MS, Papademetriou V, Rokkedal J, Harris K, Aurup P, vasodilator capacity and small artery remodelling in hy- Dahlo¨f B: Prognostic significance of left ventricular mass pertensive patients. J Hypertens 21: 615–621, 2003 change during treatment of hypertension. JAMA 292: 33. Mancia G, Zanchetti A, Agabiti-Rosei E, Benemio G, De 2350–2356, 2004 Cesaris R, Fogari R, Pessina A, Porcellati C, Rappelli A, 39. Muiesan ML, Salvetti M, Monteduro C, Bonzi B, Paini A, Salvetti A, Trimarco B: Ambulatory blood pressure is su- Viola S, Poisa P, Rizzoni D, Castellano M, Agabiti-Rosei E: perior to clinic blood pressure in predicting treatment- Left ventricular concentric geometry during treatment ad- induced regression of left ventricular hypertrophy. versely affects cardiovascular prognosis in hypertensive SAMPLE Study Group. Study on Ambulatory Monitoring patients. Hypertension 43: 1–8, 2004