Effect of valsartan on left ventricular anatomy and systolic function and aortic elasticity

Available online at www.sciencedirect.com Metabolism Clinical and Experimental 58 (2009) 682 – 688 Effect of valsartan on left ventricular anatomy and systolic function Lalita Khaodhiara, Aoife M. Brennanb, Christina Limaa, Jean L. Chanb, Christos S. Mantzorosb, Warren J. Manningc, Peter G. Daniasd,e, Aristidis Vevesa,⁎ aMicrocirculation Laboratory, Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, MA 02215, USA bDepartment of Medicine, Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, MA 02215, USA cDepartment of Radiology, Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, MA 02215, USA dCardiac MR Center and 2nd Cardiology Clinic, Hygeia Hospital, Maroussi, Athens 15123, Greece eTufts University School of Medicine, Boston MA 02111, USA Received 30 July 2008; accepted 21 January 2009 The objective of the study was to examine the effect of a 6-month daily treatment with 160 mg valsartan, an angiotensin II receptor blocker, on the left ventricular systolic function and aortic elasticity of patients with type 2 diabetes mellitus (T2DM) and healthy subjects.
This was a prospective, randomized, double-blind, placebo-controlled crossover study. Thirteen healthy control subjects and 11 patients withT2DM were enrolled in the study. Eight control subjects and 4 T2DM patients completed the study. Cardiovascular magnetic resonance wasused to evaluate the effect of valsartan on the left ventricular function and aortic elasticity. At baseline, T2DM patients had increased leftventricular mass (P = .006) when compared with the healthy controls. In the T2DM patients, treatment with valsartan, in comparison withreceiving placebo, resulted in a reduction of aortic radius (P = .026) and wall thickness (P = .032) of the ascending aorta. In the abdominalaorta, valsartan treatment, when compared with placebo treatment, reduced the arterial compliance (P = .014) in the T2DM patients. Valsartantreatment for 6 months decreased the diameter and wall thickness of the ascending aorta in patients with T2DM, but may decrease AC of theabdominal aorta.
2009 Elsevier Inc. All rights reserved.
be about 5-fold for women This increased risk implies adirect relationship between diabetes and cardiomyopathy Diabetes is associated with an increased risk for cardiovas- because it persisted despite controlling for other factors that cular morbidity and mortality that is considered to be related could be related to the development of heart failure such as age, mainly to changes in the lipid profile, accelerated athero- hypertension, coronary artery disease, lipid abnormalities, and sclerosis, endothelial dysfunction, increased platelet aggreg- obesity. The causes of diabetic cardiomyopathy are not clear, ability, raised fibrogen levels, and increased plasminogen and there is probably a constellation of pathogenic mechanisms activator inhibitor–1 activity . Diabetes is also associated with a distinct cardiomyopathy, the main clinical feature of Type 2 diabetes mellitus is associated with impaired which is congestive heart failure in the absence of coronary vascular elastic properties of the arterial tree, including the artery, hypertensive, valvular, congenital, or alcoholic heart large vessels such as the ascending and abdominal aorta disease Diabetic men participating in the Framingham Under normal circumstances, the bolus of blood that is Heart Study had more than twice the risk for development of propulsed into the arterial system during ventricular ejection congestive heart failure compared with nondiabetic partici- creates flow waves that travel distally at a velocity that is pants. The previously mentioned increased risk was shown to largely determined by the elastic properties of the arterialwall and the distal conduit resistances, which inducereflectance waves Reduced elasticity is believed to result in increased systolic pressure and ventricular mass and Corresponding author. Tel.: +1 617 632 7075; fax: +1 617 632 7090.
0026-0495/$ – see front matter 2009 Elsevier Inc. All rights reserved.
L. Khaodhiar et al. / Metabolism Clinical and Experimental 58 (2009) 682–688 Animal studies indicate that treatment with angiotensin II receptor blockers (ARB), in combination with angiotensin-converting enzyme inhibitors or alone, has a beneficial effect This was a prospective, randomized, double-blind, on cardiac remodeling and ventricular function after placebo-controlled, crossover study. All participants were myocardial infarction . These beneficial effects are evaluated during an initial screening visit and, if suitable, mediated by nitric oxide in heart failure . There is, were asked to return and be enrolled in the study. The design however, little information regarding humans. In addition, no of the study is depicted in . The randomization process information is available regarding the effects of angiotensin occurred within groups; and as a result, it was separated for II blockade in diabetic patients where ventricular dysfunc- the 2 groups. The first period lasted 6 months, the washout period lasted 2 months, and the second period lasted 6 The primary hypothesis of this trial was that indices of left months. The baseline visit included physical examination, ventricular (LV) function and aortic elasticity are impaired in blood tests, and CMR. Participants were randomized at this diabetes, even in the absence of coronary artery disease, and visit and were started on treatment with either placebo or 160 that valsartan, an ARB, may improve LV function and aortic mg valsartan daily. The second visit, at the end of the first 6- elasticity in the early stages of diabetic cardiomyopathy.
month period, included a physical examination, blood tests,and CMR. The third visit, at the end of the 2-month washoutperiod, included physical examination, blood tests, and CMR. Each participant was switched to the oppositetreatment of that of the first period. The exit visit included physical examination, blood tests, and CMR. In all visits, Twenty healthy nondiabetic subjects and 20 patients with laboratory tests were performed after an overnight fast.
T2DM, aged 21 to 80 years, were recruited. Healthy Compliance was evaluated by counting the returned tablets.
nondiabetic subjects underwent an oral glucose tolerance Plasma glucose, total serum cholesterol, low-density test to exclude unknown diabetes. Oral fasting glucose of less lipoprotein cholesterol, high-density lipoprotein cholesterol, than 100 mg/dL and a 2-hour post–oral glucose tolerance test triglycerides, liver function tests, electrolytes, blood urea plasma glucose less than 140 were required Exclusion nitrogen, and creatinine were measured using the Synchron criteria included clinical coronary artery disease; arrhythmia; CX analyzer (Beckman/Coulter, Brea, CA). Hemoglobin A1c heart failure (New York Heart Association class III and IV); (HbA1c) (reference range, 4%-6%) was determined in whole stroke or transient ischemic attack; uncontrolled hyperten- blood using ion exchange high-performance liquid chroma- sion; macroalbuminuria; severe dyslipidemia (triglycerides N600 mg/dL or cholesterol N350 mg/dL); serious chronic disease that could affect the ability of the subject to participatein the study; treatment with ARBs, glucocorticoids, anti- Cardiovascular magnetic resonance imaging was per- neoplastic agents, and bronchodilators; claustrophobia; and formed on a 1.5-T whole-body scanner (Gyroscan NT/ACS; subjects unable to have cardiovascular magnetic resonance Philips Medical Systems, Best, the Netherlands), which is (CMR) scan (eg, pacemaker, defibrillator).
equipped with the Powertrack 6000 gradient hardware (23 The protocol was approved by the institutional review mT/m, 219-μs rise time), and advanced cardiac software. A board at the Beth Israel Deaconess Medical Center. All 5-element phase array coil was used as the radiofrequency participants gave written informed consent. Participants for receiver for imaging of the heart and thoracic aorta, and the the study were recruited through local advertisement.
body coil was used for imaging of the abdominal aorta.
Fig. 1. Scheme of the clinical trial.
L. Khaodhiar et al. / Metabolism Clinical and Experimental 58 (2009) 682–688 After initial localizing “scout” images, cine LV long-axis P(d)]}, where D(s) and D(d) are the systolic and diastolic (2-chamber), 4-chamber, and contiguous short-axis images diameters of the artery and P(s) and P(d) are the systolic were obtained using a steady-state free processor breath-hold and diastolic blood pressures, respectively. Arterial cine sequence. Temporal resolution of the images was 30 to compliance is measured in square millimeters per kilo- 35 milliseconds, and the duration of the breath-hold was 10 to 12 seconds (depending on the heart rate). Oblique transverse Stiffness index (SI): defined as the natural logarithm images were obtained perpendicular to the long axis of the of the ratio of systolic to diastolic blood pressure aorta at the sinotubular junction (ascending aorta) and divided by the circumferential arterial strain (CAS), immediately proximal (cephalad) to the renal arteries which is the fractional increase in arterial diameter (abdominal aorta). Imaging was performed using a retro- during the cardiac cycle. Thus, SI is a unitless spective electrocardiographic-gated phase-encoded gradient- quantity and considered to be relatively independent echo sequence with the following parameters: field of view, of blood pressure. SI = ln[P(s)/P(d)]/CAS, where 210 × 300 mm; matrix, 96 × 128; echo time = 6.5 milliseconds; repetition time = 15 milliseconds; flip angle, Pressure-strain elastic modulus (Ep): defined as the 30°; slice thickness, 6 mm; and velocity encoding, 300 cm/s.
arterial pulse pressure divided by the CAS: Ep = [P(s) − Finally, to determine aortic wall thickness, a high-resolution P(d)]/CAS and is measured in kilopascals.
black blood image was obtained at both the ascending Young elastic modulus (YEM): defined as the ratio of thoracic and abdominal aorta using a turbo spin-echo (TSE) stress (force per unit area) to strain and measures arterial sequence with a dual inversion pulse with the following stiffness controlling for vessel wall thickness. YEM = parameters: field of view, 320 × 400 mm; matrix, 336 × 512; (R/WT){[P(s) − P(d)]/CAS}, where R is the outer echo time = 20 milliseconds; repetition time equal to the RR arterial radius and WT is the wall thickness (intima interval in milliseconds; TSE factor=12; flip angle, 90°; and plus media). The YEM is measured in kilopascals.
slice thickness, 6 mm. The latter scan was performed during breath-holding to minimize respiratory artifacts (breath-holdduration of 10-12 seconds). For the phase-encoded images, The Minitab statistical package (Minitab, State College, respiratory motion compensation was accomplished by PA) was used for the statistical analysis. The analysis for the measuring multiple signal averages (NSA = 4).
effect of valsartan treatment was performed using a parametric During the examination, blood pressure was noninva- test (ie, paired t test) for normally distributed data and a sively measured using an automated sphygmomanometer nonparametric test (ie, Wilcoxon matched-pair signed rank (Dinamap; GE Medical Systems, Madison, WI), with the test) for data that are not distributed normally to compare the cuff placed at the calf. The mean of 3 values was used for changes during the placebo and active period treatments in calculations of vascular elasticity.
each group. The t test was used to compare the baseline Image data were transferred off-line to a ViewForum characteristics between the healthy nondiabetic subjects and workstation (Philips Medical Systems) for further analysis.
the T2DM patients for normally distributed data and the Left and right ventricular endocardial and epicardial Mann-Whitney for nonparametric data. The results are contours for all short-axis end-diastolic and end-systolic presented as mean ± SD for normally distributed data and images were manually traced. Using the commercially median (25-75 percentile) for data that are not distributed available analysis package, volumetric assessment of mass, normally. Single and multiple regression analysis was also end-diastolic and end-systolic volumes, ejection fraction performed. Statistical significance was accepted at the 95% (EF), stroke volume, and cardiac output were derived for the Aortic maximal and minimal cross-sectional areas were determined from the phase-encoded flow scans using semiautomated ViewForum software. Thus, the phases 3.1. Comparisons in CMR measurements between healthy with the maximal and minimal aortic diameter/area were determined and used in the calculations for aortic elasticity.
For assessment of the aortic wall thickness, 2 measurements Thirteen healthy control subjects and 11 patients with were obtained in areas free of artifact; and the mean of the 2 T2DM were enrolled. The baseline demographics of those who completed the study in each group are shown in Vascular elasticity was described with the follow- Type 2 diabetes mellitus patients had higher body mass index (BMI) (P = .002), systolic blood pressure (P = .03), fastingblood glucose (P = .0001), and HbA1c (P = .0001). Baseline Arterial compliance (AC): defined as the absolute volume CMR data are shown in The T2DM patients had increase within an arterial segment during the cardiac higher LV mass (P = .006). When all subjects were cycle divided by the arterial pulse pressure. The AC per considered as 1 group, significant correlations were found unit length (1 mm) is AC = π[D(s)2 − D(d)2]/{4[P(s) − between LV mass and HbA1c (r = 0.58, P = .005), fasting L. Khaodhiar et al. / Metabolism Clinical and Experimental 58 (2009) 682–688 between changes observed during the valsartan and placebo treatment periods in both the healthy control subjects and the T2DM patients. In the ascending aorta measurements of T2DM patients who completed the study, treatment with valsartan, in comparison with receiving placebo, resulted in a reduction of aortic radius (P = .026) and wall thickness (P = .032). In the abdominal aorta, valsartan treatment, when compared with placebo treatment, reduced the AC (P = .014) The main finding of this study is that administration of 160 mg of valsartan, an ARB, for 6 months decreased the radius and wall thickness of the ascending aorta in T2DM patients, whereas it had no measurable effects in healthy control subjects. In the abdominal aorta, valsartan decreased the AC in the T2DM patients, but had no effect in the controls.
Previous studies have suggested that the development of cardiomyopathy starts early in the course of diabetes and that diabetic patients without any clinical findings suggestive of heart failure may have significant abnormalities of both systolic and diastolic function In the present study, we included T2DM patients with no clinical cardiovascular Mean ± SD or median (25:75 percentile). BP indicates blood pressure; LDL, low-density lipoprotein; HDL, high-density lipoprotein; BUN, blood urea nitrogen; ACE, angiotensin-converting enzyme; NS, not significant.
blood glucose (r = 0.53, P = .008), and BMI (r = 0.64, P = .001); but in multiple regression analysis, only BMI retained statistical significance. There were no differences in any of 3.2. Effects of valsartan treatment on CMR measurements in T2DM patients and the healthy control subjects Eight control subjects and 4 T2DM patients completed the study. Five subjects were lost to follow-up, 6 subjects withdrew consent because they were unable to comply with the study protocol, and 1 subject withdrew because of mild dizziness that resolved after stopping the study medication.
There were no differences between those who completed the study and those who failed to complete the study in any of the clinical characteristics that are listed in .
Treatment with valsartan did not have any effect on systolic and diastolic blood pressure, fasting blood glucose levels, and HbA1c in both groups. The comparison between difference in the CMR measurements during the valsartan and placebo period treatment is shown in There Data are presented as mean ± SD or median (25:75 percentiles). EDV indicates were no differences in all performed cardiac measurements end-diastolic volume; ESV, end-systolic volume; MR, mitral regurgitant.
L. Khaodhiar et al. / Metabolism Clinical and Experimental 58 (2009) 682–688 Table 3Changes with the valsartan treatment when compared with changes during placebo treatment period in the healthy controls and the T2DM patients Data are presented as mean and 95% confidence intervals.
disease and a relatively short duration of diabetes (mean, 5 ± sclerosis, this finding, if confirmed at a larger cohort, may 3 years); but our results are in agreement with the previous be very important because it indicates a possible beneficial studies because we found that the diabetic patients had pleiotropic effect of valsartan on atherosclerosis, poten- increased wall thickness and LV mass when compared with tially independent of reduction of the peripheral blood There are very limited data regarding the effect of ARBs An unexpected finding was the decrease of AC, which is on cardiac function in diabetes. A recent study that defined as the absolute volume increase within an arterial included diabetic patients without hypertension or heart segment during the cardiac cycle divided by the arterial disease who were treated for 6 months with candesartan, pulse pressure, in the abdominal aorta of the valsartan- another ARB, showed an improvement in diastolic treated T2DM patients. The calculation of compliance is dysfunction and attenuation of myocardial fibrosis, sug- based on the measurement of both the systolic and diastolic gesting that ARBs may regulate collagen turnover by aortic diameter and blood pressure. Although CMR can facilitating collagen degradation Previous studies in accurately measure changes in the aortic diameter during our unit have shown that valsartan treatment for 3 months, valsartan treatment, previous studies have indicated that at the same dose as in the present study, results in changes in the central blood pressure are not reflected by considerable improvement of blood flow in the skin changes in the peripheral blood pressure measurements; microcirculation of diabetic patients but had no effect in and this may be the main reason for the obtained results the skin microcirculation of healthy control subjects .
. Furthermore, all other indices of vascular elasticity Furthermore, the same studies indicated that valsartan were not affected; and there are no pathophysiologic exerts its beneficial effects by reducing the activity of poly mechanisms that would explain such deterioration in (adenosine diphosphate-ribose) polymerase, which is vascular compliance. As a result, we believe that the increased in diabetes and is associated with endothelial physiologic significance of this finding is doubtful. Never- theless, this finding deserves further investigation in a The most interesting finding of the present study was larger sample of patients and possibly during longer periods the effect of valsartan on the elasticity of the ascending aorta. In the T2DM patients, valsartan reduced the vessel In the present study, we have used CMR for evaluating radius and wall thickness of the ascending aorta. As the the effect of valsartan on LV function and aortic elasticity.
above measurements are correlates of subclinical athero- The main reason for this is that this technique is currently L. Khaodhiar et al. / Metabolism Clinical and Experimental 58 (2009) 682–688 considered as the “criterion standard” for measurement of LV [5] Kannel WB, Hjortland M, Castelli WP. Role of diabetes in mass and volumes [Because of extremely high congestive heart failure: the Framingham Study. Am J Cardiol1974;34:29-34.
accuracy and reproducibility, CMR allows for high statistical [6] Zarich SW, Nesto RW. Diabetic cardiomyopathy. Am Heart J 1989; power to detect small differences in cardiac structure and function between study groups, with even a small sample [7] Henry RM, Kostense PJ, Spijkerman AM, Dekker JM, Nijpels G, size. Accordingly, for clinical trials, CMR compares Heine RJ, et al. Hoorn Study. Arterial stiffness increases with favorably to echocardiography and is becoming the deteriorating glucose tolerance status: the Hoorn Study. Circulation2003;107:2089-95.
dominant imaging modality in cardiovascular clinical [8] Lehmann ED, Watts GF, Fatemi-Langroudi B, Gosling RG. Aortic research Cardiovascular magnetic resonance can compliance in young patients with heterozygous familial hypercho- also most accurately measure small changes in the aortic lesterolaemia. Clin Sci (Lond) 1992;83:717-21.
lumen cross-sectional area with very high temporal resolu- [9] O'Rourke MF, Staessen JA, Vlachopoulos C, Duprez D, Plante GE.
tion and can quantify blood flow through large vessels.
Clinical applications of arterial stiffness; definitions and referencevalues. Am J Hypertens 2002;15:426-44.
Therefore, CMR is uniquely suited for the evaluation of the [10] Groenink M, de Roos A, Mulder BJ, Spaan JA, van der Wall EE.
Changes in aortic distensibility and pulse wave velocity assessed with The current study has its limitations, of which the most magnetic resonance imaging following beta-blocker therapy in the prominent is the considerable number of subjects who did Marfan syndrome. Am J Cardiol 1998;82:203-8.
not complete the study (9; 37% of the participants). We [11] Westerhof N, O'Rourke MF. Haemodynamic basis for the development of left ventricular failure in systolic hypertension and for its logical believe that the main reason for this was the long duration therapy. J Hypertens 1995;13:943-52.
of the study and the crossover design. This, combined with [12] Mankad S, d'Amato TA, Reichek N, McGregor WE, Lin J, Singh D, the small number of subjects that were recruited, resulted in et al. Combined angiotensin II receptor antagonism and angiotensin- a rather small study cohort. However, the fact that valsartan converting enzyme inhibition further attenuates postinfarction left was found to improve certain indexes of ascending aortic ventricular remodeling. Circulation 2001;103:2845-50.
[13] Kim S, Yoshiyama M, Izumi Y, Kawano H, Kimoto M, Zhan Y, et al.
function in T2DM patients but not in controls indicates that, Effects of combination of ACE inhibitor and angiotensin receptor despite these limitations, the study has accomplished its blocker on cardiac remodeling, cardiac function, and survival in rat primary scope, which was to provide proof of concept and heart failure. Circulation 2001;103:148-54.
justify the conduction of additional studies in the future. As [14] Liu YH, Xu J, Yang XP, Yang F, Shesely E, Carretero OA. Effect of the conduction of such studies is cumbersome and carries a ACE inhibitors and angiotensin II type 1 receptor antagonists onendothelial NO synthase knockout mice with heart failure. Hyperten- heavy cost, we believe that the initial conduction of a small study that aims primarily to provide proof of concept is [15] American Diabetes Association. Report of the Expert Committee on fully justified and that the current study has achieved its the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care In summary, our results indicate that valsartan treatment [16] Kahn JK, Zola B, Juni JE, Vini AI. Radionuclide assessment of left ventricular diastolic filling in diabetes mellitus with and without for 6 months decreased the diameter and wall thickness of cardiac autonomic neuropathy. J Am Coll Cardiol 1986;7:1303-9.
the ascending aorta in T2DM patients but may decrease AC [17] Airaksinen KEJ, Koistinen MJ, Ikaheimo MJ, et al. Augmentation of atrial contribution to left ventricular filling in IDDM subjects asassessed by Doppler echocardiography. Diabetes Care 1989;12: [18] Danias PG, Tritos NA, Stuber M, Kissinger KV, Salton CJ, Manning WJ. Cardiac structure and function in the obese: a cardiovascular The present study was an investigator-initiated research magnetic resonance imaging study. J Cardiovasc Magn Reson 2003;5: protocol and was supported by a clinical research grant to Aristidis Veves, MD, from Novartis. This research was also [19] Friberg P, Allansdotter-Johnsson A, Ambring A, Ahl R, Arheden H, supported in part by grant RR 01032 to the Beth Israel Framme J, et al. Increased left ventricular mass in obese adolescents.
Deaconess Medical Center General Clinical Research Center [20] Kawasaki D, Kosugi K, Waki H, Yamamoto K, Tsujino T, Masuyama from the National Institutes of Health.
T. Role of activated renin-angiotensin system in myocardial fibrosisand left ventricular diastolic dysfunction in diabetic patients—reversal by chronic angiotensin II type 1A receptor blockade. Circ J 2007;71:524-9.
[1] Deliargyris EN, Nesto RW. Autonomic neuropathy and heart disease.
[21] Shrikhande G, Khaodhiar L, Scali S, Lima C, Hubbard M, Dudley K, In: Veves A, editor. The clinical management of diabetic neuropathy.
et al. Valsartan improves resting skin blood flow in type 2 diabetic Totowa (NJ): Humana Press; 1998. p. 209-26.
patients and reduces poly(adenosine diphosphate-ribose) polymerase [2] Stamler J, Vaccaro O, Neaton J, Wentworth D. Diabetes, other risk activation. J Vasc Surg 2006;43:760-70.
factors, and 12-yr cardiovascular mortality for men screened in the [22] Szabó C, Zanchi A, Komjáti K, Pacher P, Krolewski AS, Quist WC, Multiple Risk Factor Intervention Trial. Diabetes Care 1993;16:434-44.
et al. Poly(ADP-Ribose) polymerase is activated in subjects at risk of [3] Rubler S, Dlugash J, Yuceoglu YZ, Kumral T, Branwood AW, developing type 2 diabetes and is associated with impaired vascular Grishman A. New type of cardiomyopathy associated with diabetic reactivity. Circulation 2002;106:2680-6.
glomerulosclerosis. Am J Cardiol 1972;30:595-602.
[23] Garcia Soriano F, Virag L, Jagtap P, Szabo E, Mabley JG, Liaudet L, [4] Raman M, Nesto RW. Heart disease in diabetes mellitus. Endocrinol et al. Diabetic endothelial dysfunction: the role of poly (ADP-ribose) polymerase activation. Nat Med 2001;7:108-13.
L. Khaodhiar et al. / Metabolism Clinical and Experimental 58 (2009) 682–688 [24] Danias PG, Tritos NA, Stuber M, Botnar RM, Kissinger KV, Manning by the use of cardiovascular magnetic resonance. J Cardiovasc Magn WJ. Comparison of aortic elasticity determined by cardiovascular magnetic resonance imaging in obese versus lean adults. Am J Cardiol [29] Stuber M, Nagel E, Fischer SE, et al. Quantification of the local heart wall motion by magnetic resonance myocardial tagging. Comput Med [25] Williams B, Lacy PS, Thom SM, Cruickshank K, Stanton A, Collier D, et al. Differential impact of blood pressure–lowering drugs on central [30] Fischer SE, McKinnon GC, Scheidegger MB, et al. True myocardial aortic pressure and clinical outcomes: principal results of the Conduit motion tracking. Magn Reson Med 1994;31:401-13.
Artery Function Evaluation (CAFE) Study. Circulation 2006;113: [31] Stuber M, Manning WJ. CSPAMM assessment of left ventricular diastolic function. In: Manning WJ, Pennell DJ, editors. Clinical [26] Cranney GB, Lotan CS, Dean L, Baxley W, Bouchard A, Pohost GM.
cardiac magnetic resonance imaging. Churchill Livingstone; 2002.
Left ventricular volume measurement using cardiac axis nuclear magnetic resonance imaging. Validation by calibrated ventricular [32] Resnick LM, Militianu D, Cunnings AJ, Pipe JG, Evelhoch JL, Soulen angiography. Circulation 1990;82:154-63.
RL. Direct magnetic resonance determination of aortic distensibility in [27] Sakuma H, Fujita N, Foo TK, et al. Evaluation of left ventricular essential hypertension: relation to age, abdominal visceral fat, and in volume and mass with breath-hold cine MR imaging. Radiology 1993; situ intracellular free magnesium. Hypertension 1997;30:654-9.
[33] Rogers WJ, Hu YL, Coast D, Vido DA, Kramer CM, Pyeritz RE, et al.
[28] Bellenger NG, Davies LC, Francis JM, Coats AJ, Pennell DJ.
Age-associated changes in regional aortic pulse wave velocity. J Am Reduction in sample size for studies of remodeling in heart failure

Source: http://cardiacmr.hms.harvard.edu/wp-content/uploads/2011/07/Khaodhiar-2009-Effect-of-valsartan.pdf