Effects of an aging vascular model on healthy and diseased hearts

David Jegger, Rafaela F. da Silva, Isabelle Lartaud, Virginie Gaillard, Xavier Jeanrenaud, Mohammad Nasratullah, Ludwig K. von Segesser, Jeffrey Atkinson, Patrick Segers, Hendrik Tevaearai, Nikolaos Stergiopulos


The vitamin D3 and nicotine (VDN) model is a model of isolated systolic hypertension (ISH) due to arterial calcification raising arterial stiffness and vascular impedance similar to an aged and stiffened arterial tree. We therefore analyzed the impact of this aging model on normal and diseased hearts with myocardial infarction (MI). Wistar rats were treated with VDN (n = 9), subjected to MI by coronary ligation (n = 10), or subjected to a combination of both MI and VDN treatment (VDN/MI, n = 14). A sham-treated group served as control (Ctrl, n = 10). Transthoracic echocardiography was performed every 2 wk, whereas invasive indexes were obtained at week 8 before death. Calcium, collagen, and protein contents were measured in the heart and the aorta. Systolic blood pressure, pulse pressure, thoracic aortic calcium, and end-systolic elastance as an index of myocardial contractility were highest in the aging model group compared with MI and Ctrl groups (PVDN < 0.05, 2-way ANOVA). Left ventricular wall stress and brain natriuretic peptide (PVDN×MI = not significant) were highest, while ejection fraction, stroke volume, and cardiac output were lowest in the combined group versus all other groups (PVDN×MI < 0.05). The combination of ISH due to this aging model and MI demonstrates significant alterations in cardiac function. This model mimics several clinical phenomena of cardiovascular aging and may thus serve to further study novel therapies.

  • calcium
  • echocardiography
  • hypertrophy
  • hypertension
  • infarction

during aging, alterations in aortic structure and function occur, leading to a decrease in aortic compliance. In particular, the arterial extracellular matrix undergoes many profound age-related changes responsible for wall stiffening. Age-dependent medial degeneration, including phenomena such as elastocalcinosis (calcification followed by degeneration and fragmentation of elastic fibers), is probably the most important element in increased arterial stiffness, progressively leading to a loss of compliance (3) and an increase in pulse wave velocity (PWV). This leads to early return of reflected waves from the periphery and elevated central systolic and pulse pressure, eventually yielding isolated systolic hypertension (ISH) (4). More than six decades ago, Blumenthal et al. (6) noted that the time course of the decrease of elasticity with age closely paralleled the curve of the intensity of medial calcification with age.

Today, the subsequent chronic early return of a reflected pressure wave to the heart is accepted as an independent cardiac risk factor (8, 38). This phenomenon can be reproduced and studied in dogs after replacement of the thoracic aorta by a plastic tube (22). In small animals, calcinosis and stiffening of the aorta, typically observed during aging, can be induced by administration of vitamin D3 and nicotine (VDN) (16, 31) and leads progressively to the development of ISH and compensatory left ventricular (LV) hypertrophy (21, 2527).

Epidemiologic studies have also shown that coronary heart disease (CHD) and arterial hypertension are independent predictors of congestive heart failure, the occurrence being 40% when both pathologies are concomitant (18). However, most experimental models of heart failure are based on a single cause. Therefore, the development of a model combining myocardial infarction (MI) and ISH due to aging would be of particular interest.

MI models are relatively well developed, and there are several possibilities for inducing arterial hypertension in small animals. However, most of these are typical models of essential hypertension: spontaneously hypertensive rats (SHR) (13), renal hypertensive models (2, 29), and Dahl salt-sensitive rats (19). Models mimicking ISH are rare, except for the above-mentioned VDN model.

To date, however, no data have been reported on the effects of VDN on diseased (MI) hearts. We therefore hypothesized that such a model would show aggravated alterations of cardiac structure and function and an accelerated evolution toward compromised cardiac function compared with either of the two pathologies isolated, and would therefore represent a novel small-animal model of combined MI and aging, better mimicking the pathophysiological alterations found in the clinical environment.



Male Wistar rats (2 mo old; Charles River, Lyon, France) were maintained in temperature- and humidity-controlled rooms with a typical light-dark cycle and given standard chow and mineral water (Mont Roucous) ad libitum. The investigation conforms to the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health (NIH Pub. No. 85-23, revised 1996). The protocol was approved by our local ethical committee.

Animals (n = 64) were randomized into four age- and weight-matched groups as follows.

Control group (n = 10, 226 ± 7 g).

On day 1, the control (Ctrl) group animals received saline intramuscularly (volume equivalent to the VDN group) and underwent two 5 ml/kg gavages of distilled water at 9 AM and 5 PM. At week 3, animals were anesthetized with isoflurane (Forene, Abbott, Baar, Switzerland), intubated, ventilated (100% oxygen, 60 cycles/min, tidal volume of 2 ml; model 683, Harvard Apparatus, Holliston, MA), and placed on a heating pad before a sham operation was performed (left thoracotomy in the 3rd intercostal space and opening of the pericardium). The thorax was closed. Analgesia was then administered intraperitoneally (40 mg/kg) and again 24 h after surgery (Pro-Dafalgan, Upsamedica, Baar, Switzerland).

Aging model (n = 15, 226 ± 8 g).

The procedure was described in detail previously (25). Briefly, on day 1 at 9 AM, 300,000 IU/kg of vitamin D3 (Duphafral D3 1000, Duphar, Weesp, The Netherlands) was injected in the hind leg muscle and 25 mg/kg nicotine (nicotine hydrogen tartrate, Sigma) in 5 ml/kg sterile water was given orally by gavaging. Gavaging was repeated once at 5 PM the same day. Three weeks later, the animals underwent a sham operation (see Ctrl group for description).

MI group (n = 10, 243 ± 7 g).

On day 1, the animals were treated similarly as animals in the Ctrl group. At week 3, a MI was induced by ligation of the left anterior descending coronary.

Combined aging and MI models (n = 29, 213 ± 24 g).

The VDN/MI animals received the VDN treatment on day 1 and underwent a MI at week 3 according to the procedures detailed above.

Invasive Cardiac Measurements (Conductance Catheter)

At week 8, the animals were reanesthetized and intubated, and the right jugular vein and carotid artery were isolated. A 2-Fr conductance catheter (CC; SPR 838 Aria, Millar Instruments) was inserted into the ascending aorta via the right carotid artery. Systolic (SBP), diastolic (DBP), mean (MBP), and pulse pressures (PP) were obtained. Thereafter, the CC was advanced into the LV. Parallel conductance was measured after injection of 20 μl of 10% saline into the jugular vein in accordance with the method of Baan et al. (5). An occlusion analysis was performed by temporarily occluding the inferior vena cava below the diaphragm via a minilaparotomy. For each animal, the CC calibration correction factor α was assessed with the use of an ultrasonic flowmeter (Transonic, Ithaca NY) placed around the ascending aorta.

The following parameters were derived: stroke volume (SV), end-systolic volume (ESV), end-diastolic volume (EDV), ejection fraction (EF = SV/EDV), peak negative value of the time derivative of LV pressure (dP/dtmin), time constant of relaxation defined as the time constant of the pressure drop from the time of peak negative dP/dt (τ, using monoexponential decay method with zero asymptote), heart rate (HR), and LV end-diastolic pressure (LVEDP). Cardiac output (CO) was calculated as SV multiplied by HR.

Pressure-volume loops were obtained by occlusion of the inferior vena cava and used to calculate the slope (end-systolic elastance, Ees), preload recruitable stroke work (PRSW) (with stroke work assessed from the area enclosed by the pressure-volume loop), preload-adjusted dP/dtmax-EDV (slope of the relation between dP/dtmax and EDV; their relationship is linear, and slope represents an inotrope-sensitive, load-independent index of contractility), potential energy (PE), and pressure-volume area (PVA). The ratio of stroke work and PVA is a measure of efficiency.

Echocardiographic Measurements

Transthoracic echocardiography (TTE) was performed as previously described (21). In brief, TTE (C256 Sequoia echocardiographic system, Acuson, Mountain View, CA) was performed at baseline in 15 arbitrarily selected rats before they were randomized into the 4 groups. Thereafter, a new TTE was performed in each rat every second week until week 8. A 15-MHz linear array transducer (15L8, Acuson) with bidimensional and color Doppler imaging was used. From the long-axis view, an M-mode trace of the LV was obtained, and LV end-diastolic diameter (LVDed), LV systolic diameter (LVDes), and posterior and septal wall diastolic thickness (PWth and SWth) were measured. LV ejection time (ET) was measured as the time from the beginning to the end of the aortic flow wave. Isovolumetric relaxation time (IRT) was measured as the interval between the aortic closure click and the start of mitral flow, while isovolumetric contraction time (ICT) was obtained as the time delay between the cessation of mitral inflow and the onset of aortic ejection. Additionally, the mitral valve closure time (MCO) was measured. The myocardial performance index (MPI) was then defined as (MCO − ET)/ET = (ICT + IRT)/ET. LV fractional shortening (LVFS) was calculated as (LVDed − LVDes)/LVDed × 100. Velocity of circumferential fiber shortening (Vcf) was calculated with the following formula: (LVDed − LVDes)/(ET × LVDed). EDV and ESV were calculated with Simpson's method, and SV was calculated as EDV − ESV. LV mass was calculated as {[(LVDed + SWth + PWth)3 − LVDed3] × 1.04} × 0.8 + 0.14 (in g, with LV dimensions expressed in cm). Relative wall thickness was assessed as RWT = (PWth + SWth)/LVDed, which is equivalent to the ratio h/r (h being LV wall thickness and r being its radius). Values >0.40 reflect signs of concentric hypertrophy (15). The meridional end-systolic wall stress (ESS) was calculated only at death as [0.334(SBP)(LVDes)]/[PWth(1 + PWth/LVDes)] in 103 dyn/cm2 (33). Finally, diastolic function was assessed with the deceleration time (DT) of the pulmonary venous diastolic flow, considered to be more accurate than pulmonary artery occlusion pressure in predicting left atrial pressure (23).

Cardiac and Vascular Tissue Calcium Content

A 10-mm sample of the descending thoracic aorta and the cardiac apex were removed, and tissue calcium content (μmol/g dry wt) was determined by atomic absorption spectrophotometry (AA10, Varian) after mineralization and acid digestion of the tissue (17).

Collagen Analysis

Collagen content was measured from hydroxyproline content by protein hydrolysis followed by colorimetric spectrophotometry (30).

Cardiac Infarct Size and Histology

After completion of these measurements, the LV was excised and weighed. The infarct size area (IS) was determined as a percentage of the entire LV area, as reported previously (10). For histology, heart tissue samples were fixed in 4% formaldehyde and mounted in paraffin block, and slices were obtained with a microtome and treated with either a periodic acid Schiff staining (Sigma-Aldrich Chimie SARL, Epalinges, Switzerland) protocol (to discriminate cell borders) or 0.1% picrosirius red (for collagen; Sigma). The means of the cardiomyocyte cross-sectional area and diameter were calculated by photomicrographs of 100 cells/specimen with a computer-assisted image analysis system (Metamorph analysis).


The following primary antibodies were incubated for 2 h at room temperature. Mouse antibody against connexin 43 (BD Biosciences Pharmingen, Basel, Switzerland) and rabbit antibody against laminin (to discriminate the cell border; Sigma) used as primary antibodies were diluted (1:100) in PBS containing 4% normal goat serum, whereas goat Alexa 488 anti-mouse and goat Alexa 555 anti-rabbit were used as secondary antibodies.

RNA Extraction and Quantitative RT-PCR

Total tissue RNA was extracted from the apex of the heart with an RNeasy Mini Kit (Qiagen, Hilden, Germany). Reverse transcriptase reaction was performed starting from 1 μg of total RNA in the presence of 0.5 μmol/l dNTP (Amersham, Otelfingen, Switzerland), 1 μg of oligo(dT) (Promega, Mannheim, Germany), and 200 U/μl Superscript II (Invitrogen, Basel, Switzerland) at 42°C for 1 h. For each quantitative RT-PCR reaction 1/10th of the reverse transcriptase product was used. RT-PCR was performed on an ABI PRISM 7700 (BD Biosciences, Heidelberg, Germany) with Absolute QPCR ROX mix (ABgene-Axon, Baden-Dättwil, Switzerland). PCR primers and probe sequences were synthesized by MWG Biotech (Ebersberg, Germany). Sequences were as follows: 1) rat brain natriuretic peptide precursor (BNP) forward 5′-CAGCTCTCAAAGGACCAAGG-3′ and reverse 5′-AGAGCTGGGGAAAGAAGAGC-3′ with an internal probe of 5′-FAM-cgccttccggatccaggagagacttcg-3′-TAMRA; 2) rat GAPDH forward 5′-CCATCACTGCCACTCAGAAGAC-3′ and reverse 5′-TCATACTTGGCAGGTTTCTCCA-3′ with an internal probe of GAPDH rat FAM-5′-CGTGTTCCTACCCCCAATGTATCCGT-3′-TAMRA. Expression level of mRNA was determined by gene-specific standard curves. Taqman results were expressed as a ratio of BNP to GAPDH mRNA level for each sample and normalized to the sham-operated controls.

Statistical Analysis

Values are reported as means ± SD. Each TTE analysis was repeated three times and averaged. Comparison of TTE evolution was analyzed by repeated-measures analysis of variance. If significance was found for a variable, pairwise comparisons were performed with Bonferroni's adjustment for multiple tests. P values <0.05 were considered significant. All analysis was performed with SPSS software (SPSS 11.5, SPSS, Chicago, IL). Concerning all other parameters, comparisons were carried out with two-way ANOVA (PVDN, PMI, PVDN×MI) plus the Bonferroni post hoc test in order to establish an interaction of the aging model on top of the MI heart. P values <0.05 were considered significant.


Early and Late Mortality, General Characteristics, Hemodynamics, and Aortic Parameters

There were no deaths in the Ctrl and MI groups, whereas in the aging model, 4 of 15 rats died within the first 2 wk (early mortality rate of 27%) and 2 died at week 5 (total mortality of 40%). In the combined group, 15 of 29 rats died within the first 2 wk (early mortality of 52%) and none died later.

At week 8, body weight was lowest and HR slightly increased in the VDN groups (PVDN < 0.05; Table 1). All VDN rats exhibited significantly elevated SBP and PP compared with Ctrl and MI rats (PVDN < 0.05; Table 1), with no change in MBP. As expected, the aortic calcium content was found to be highly elevated in the VDN rats compared with Ctrl and MI rats (PVDN < 0.05; Table 1).

View this table:
Table 1.

General characteristics, hemodynamics, and aortic parameters

Left Ventricular Pressures, Volume, and Stress

EDV and ESV slightly decreased in VDN rats (PVDN < 0.05; Table 2) and increased after MI (PMI < 0.05)(Figs. 1 and 2). LVEDP increased only in MI rats (PMI < 0.05, but the post hoc test did not give differences between groups because of dispersion); therefore, ESS, which increased after VDN and infarction (PVDN, PMI < 0.05; Table 2), reached the highest value (×2.5) in the VDN/MI group (PVDN×MI < 0.05). SV and CO evolved differently within the four groups (PVDN×MI < 0.05; Table 2), as they did not change or tended to be maintained in VDN- or MI-alone rats compared with Ctrl rats but fell significantly in VDN/MI rats (Fig. 1).

Fig. 1.

Pressure-volume loops of the 4 groups showing relative ejection fractions for each group. Ctrl, control; MI, myocardial infarction; VDN, vitamin D3 and nicotine; VDN/MI, combined VDN and MI.

Fig. 2.

Representative M-mode transthoracic echocardiography (TTE) of the 4 groups at week 8. Septal wall motion dyskinesia is visible in the MI and VDN/MI groups, whereas wall thickening is seen in the VDN group.

View this table:
Table 2.

Left ventricular pressures, volumes, and stress measured by conductance catheter or transthoracic echocardiography

Left Ventricular Contractility and Relaxation Indexes, Systolic and Diastolic Functions

Adjusted dP/dtmax versus EDV was elevated in VDN rats (PVDN < 0.05, but no statistical significance could be reached after the post hoc test because of large deviations; Table 3). Ees increased in VDN rats and decreased after infarction (PVDN and PMI < 0.05; Table 3). Consequently, in VDN rats PE and PVA (Fig. 1) were reduced, and therefore efficiency and stroke work were significantly higher (PVDN < 0.05; Table 3), whereas in MI rats PE and PVA did not change (PMI = not significant) and efficiency decreased (PMI < 0.05; Fig. 1). EF evolved differently according to VDN and MI effects: it tended to be maintained in VDN-alone rats (PVDN = not significant), fell after infarction (PMI < 0.05), and fell to an even lower value in VDN/MI rats (PVDN×MI < 0.05).

View this table:
Table 3.

Contractility and relaxation indexes measured by conductance catheter or transthoracic echocardiography

Systolic function evaluated by TTE (LVFS, Vcf) and combined with diastolic function (MPI) was conserved in the aging model (VDN alone) and most compromised in the two MI groups (Fig. 3, AC).

Fig. 3.

Serial TTE of the 4 groups. Values are means ± SD. LVFS, left ventricular (LV) fractional shortening (A); Vcf, velocity of circumferential fiber shortening (B); MPI, myocardial performance index (C); DT, deceleration time (D); RWT, relative wall thickness (E); BL, baseline (F); W, week. P < 0.05: *vs. Ctrl, †vs. MI, and §vs. VDN.

DT, an indicator of diastolic function, was reduced in all groups versus the Ctrl group but most compromised in the combined model (Fig. 3D). Time constant of relaxation τ increased and dP/dtmin decreased after infarction (PMI < 0.05), with no impact of VDN (Table 3).

Left Ventricular Structure and Composition

The aging model (VDN alone) showed signs of concentric hypertrophy, because h/r (represented as RWT) were >0.40 compared with the Ctrl group while significantly larger compared with the MI group (Figs. 2 and 3E). RWT was thinnest in the infarcted groups, with the MI group being significantly smaller versus the Ctrl group (Figs. 2 and 3E). LV mass in all three treatment groups increased with respect to the Ctrl group and were all significantly greater at week 8 (Fig. 3F). Myocardial fibrosis (Fig. 4B), myocyte hypertrophy (Fig. 4A), and increased expression of BNP occurred after infarction and VDN (PMI and PVDN < 0.05; Table 4), with no impact of VDN/MI (PVDN×MI = not significant). IS was not significantly different between MI alone (31 ± 6%) and VDN/MI (37 ± 7%, P = 0.332; Fig. 4B).

Fig. 4.

A: cardiomyocyte hypertrophy was estimated from LV section for the 4 groups. Top: representative cardiomyocyte cross-sectional area and diameter as revealed by histology with periodic acid Schiff. Bottom: cardiomyocyte length as visualized by immunohistochemistry using double staining with antibodies against laminin (red) and connexin 43 (green). B: extent of myocardial fibrosis was qualitatively visualized by picrosirius red histological staining. Top: entire cross-sectional area of the heart. Bottom: same sections at ×40 magnification. Top panel shows isolated examples from the basal cardiac position and might not be representative of actual infarct size.

View this table:
Table 4.

Cardiac composition


We present a novel experimental model of age-mimicking ISH combined with MI in rats. This model confirms the presence of critical signs of compromised cardiovascular function but also demonstrates structural and biochemical alterations the majority of which are altered compared with the isolated MI or VDN models. When an interaction analysis was performed, in order to establish the effects of combining raised afterload with MI, LV ESS, SV, EF, and CO were found to be significantly more influenced by the two pathologies together than by one alone.

Effect of Aging and Aging Model on Normal Heart

The aging model indicates that contractility is globally increased, which is an adaptation in the face of the increased aortic impedance and PWV, as previously reported (20, 25). It was previously reported that cardiac performance is maintained or improved until 23 mo of age in rats and afterwards drops to significant levels (34). In the resting aging heart, there are largely no alterations of systolic function, with preserved EF and SV; because resting HR is unchanged or only minimally reduced with aging, CO is also preserved (36). This is supported by our TTE data (Fig. 3). Instead, diastolic function does undergo significant age-related changes, with a reduction in early diastolic filling compensated for by increased end-diastolic filling and a consequent progressive reduction of the echocardiographic early wave-to-atrial wave velocity ratio (E/A) (28). This is supported by our TTE data with respect to the DT of the pulmonary venous flow (Fig. 3). Additionally, the aging model further enhanced the cardiac calcium content both in infarcted animals and in noninfarcted animals, thus confirming previous observations (4). The extent of fibrosis and collagen levels was not influenced in the VDN-aging model. This is contrary to what was reported by Raya et al. during aging (34). Accordingly, our previous study showed that collagen content of the rat hearts was increased after VDN treatment. These discrepancies could be explained by the lower rate of calcification in the present study (LV 1.2-fold) compared with previous experiments (LV 5- to 9-fold). In agreement, the calcification of the thoracic aorta was also lower in this study compared with our previous study (<26-fold vs. 30- to 50-fold). These changes may have led to a reduced aortic wall stiffening and a lower increase in LV afterload, which in turn affected the adaptative cardiac fibrosis of the present study (25).

BNP is an attractive parameter with which to follow the evolution of cardiac function because it is an independent predictor of long-term survival (32), it reflects the remodeling process in hypertension (39, 40), and it is correlated with EF (11). In our study, we support these findings by showing that the ESS in the aged group is maintained after acute increase in ESS causes wall thickening that returns ESS to normal, thus acting as a feedback inhibition (Table 2).

Effect of Aging and Aging Model on Infarcted Heart

In clinical situations, the prevalence of eccentric hypertrophy is also typically observed in patients suffering from hypertension and having CHD (43) or having had a MI (42, 35). In our study, the aging model showed a significant elevation of h/r measured by TTE (Fig. 3). Similar findings were reported by Grossman et al. (15) with respect to concentric hypertrophy observed in pressure-overloaded patients indicated by increased h/r and eccentric hypertrophy in volume-overloaded patients with a larger LV diameter but normal h/r.

It appears that in the combined group the initial application of the VDN protocol caused LV hypertrophy to develop, which could have protected and thus prevented the LV from dilating after the inception of the MI in the combined group (7% rise in EDV vs. Ctrl), whereas without VDN MI led to substantial increase (31%) in EDV versus Ctrl (Fig. 1 and Table 2). Furthermore, the rise in ESV in the MI and combined groups by 100% and 72%, respectively, reflects reduced contractility, evoking a compromised EF. The EF in the combined group is lower than in the MI-alone group, probably due to the fact that the preexisting concentric hypertrophy following the VDN treatment provoked a drop in SV (Table 2). Both groups of infarcted animals showed the highest calcium content. The extent of fibrosis and collagen levels was elevated in both infarcted groups. This was also observed in a model combining MI and SHR (44). Finally, cellular hypertrophy appeared to be significantly increased when the infarcted animals also had VDN-induced ISH. This was also observed by Raya et al. (34).

As opposed to the MI model, in which both EDV and ESV were increased, only ESV was significantly augmented in the combined model. Consequently, the ejection phase indexes were dramatically compromised in this group (Table 2). This is in accordance with results obtained with the combination of MI with either renal hypertension (2) or aortic banding (1). Anversa et al. (2) showed lowest values for SV and CO in their combined group, whereas Anthonio et al. (1) expressed the biggest compromise in heart weight-to-body weight ratio. Some discrepancies were reported compared with our data because of the hypertensive models' ontogeny as well as the delay before certain structural and functional parameters were measured. Nass et al. (29) used a renal hypertensive model and the animals were killed after 4-wk observation, whereas Kass et al. (22) performed an acute experiment after replacement of the thoracic aorta by a plastic tube. In the present series of VDN rats diastolic pressure was maintained, while it is normally decreased after 60 yr of age. This decrease was captured correctly, however, when compliance was decreased by banding or by stiff tube replacement of the aorta. In our study, the expression of BNP was highest in the MI groups, but significance with respect to the interaction analysis was not reached because of high standard deviations in the combined group. These findings were also reported in clinical studies (35), confirming therefore the interest of the combined model for translational applications to clinical implications. Pressure-stretch release coupling mechanisms have been identified as the principal stimuli of BNP secretion. Briguori et al. (9) showed that BNP levels are related to LV outflow tract gradient, LV dysfunction, and severity of LV hypertrophy. BNP thus reflects the degree of LV afterload, being synthesized and released when the ventricular wall stress/tension is increased (9). In our study, we support these findings by showing an increase in ESS in the MI groups. Furthermore, the BNP levels follow the same trend as the ESS. Also, a direct significant correlation could be established between ESS versus BNP levels when all data were pooled (r = 0.77).

Additionally, it has been hypothesized that ESS potentiates increased ischemic injury (34). This is feasible, because we have shown that ISH and changes in morphology augment ESS (Table 2), thereby releasing BNP (Table 4). Also, these morphological changes alter efficiency, and the already damaged heart ends up working against a larger afterload.

Finally, it may be hypothesized that the above-mentioned mechanisms are the cause of the greater (while not significantly different) IS in the combined group (37% vs. 31% in MI alone). This may explain the most compromised cardiac function, reflected by the drop in EF, among all the other groups studied (Table 2). This is supported by the significant negative interaction of VDN and MI on EF (P < 0.001) and SV (P < 0.04) while positive and significant for wall stress (P < 0.03) (Table 2). However, the influence of IS can be excluded. Indeed, when performed in a separate analysis after selection of rats with closer IS (9/10 MI rats with IS of 32.5 ± 4.3% vs. 11/14 rats of the combined group with IS of 34.4 ± 4.5%), two-way ANOVA gives a similar interpretation of results with respect to the interaction analysis, i.e., SV, CO, and EF were lowest, and ESS was highest, in the combined group.

It has indeed been proposed that the presence of LV hypertrophy is a necessary cofactor in the exacerbation of ischemic damage by chronic hypertension (7). Hence, hypertension would augment the post-MI remodeling. Epidemiological studies have emphasized the close relationship between elevated blood pressure and the incidence of cardiovascular disease (41), and SBP and PP, in particular, have been demonstrated to be strong independent predictors of cardiovascular mortality (14, 37).


Combining two methods of inducing altered cardiac and vascular function also means combining the limitations of the two techniques. However, in the present study, the standard deviation observed in the combined group was in the same range as in the other groups. Differing results might in fact have been obtained if the MI was performed before the VDN treatment (19) or even performed at a later date to allow for compensatory changes in cardiac structure and function to appear, the latter being more representative of human pathology, where arterial stiffening due to aging usually precedes MI events. The higher mortality rate in the combined group is due exclusively to the VDN treatment at the start of the experiment, and its reported toxic effect is confined to the initial 10 days of the study, after which the growth rate is similar to controls and thus should not influence the heart weight-to-body weight ratio (26). Nicotine is probably ganglioplegic at the dose used, and thus unlikely to be the cause of any rise in blood pressure or HR. Being a parasympathetic agent, nicotine's effects on the cardiovascular system are muscarinic. It has a short plasma half-life of 2.5 h, so it would be cleared rapidly from the organism without perturbation. Therefore, there seems to be no apparent rise in mortality in the combined group, which could differ if the study lasted >2 mo or if the IS was larger in the MI groups.

In conclusion, the addition of MI to a raised afterload induced by ISH is confirmed in this animal model of aging to substantially augment the severity of compromised cardiac function represented by SV, CO, EF, and ESS. Therefore, the combination appropriately mimics several clinical conditions of aging and constitutes an attractive model to study the ontogeny of altered cardiovascular function and eventually heart failure, and subsequently the potential of novel therapeutic strategies.


This work was supported by the Swiss National Research Foundation (3100AO-104257/1), Cardiovascular Scientific Foundation (Fonds Scientifique Cardiovasculaire), Swiss Life Anniversary Foundation for Public Health and Medical Research, Novartis Foundation for Medico-Biological Research, the Regional Development Committee (CPER 2000–2006, “Bioingéniérie Appliqué aux Régulations Cellulaires,” Metz, France), the greater Nancy Urban Council (Nancy, France), and the Pharmacolor Association (Nancy, France).


The authors acknowledge the assistance from the Bioimaging and Optics (BIOP), Image Processing and Analysis Office of the Ecole Polytechnique Fédérale de Lausanne and thank Patrick Liminaña (Pharmacology Laboratory, Pharmacy Faculty, UHP Nancy 1, France) for excellent technical help with histomorphometric analysis preparation.


  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


View Abstract