Calcification of Medial Elastic Fibers and Aortic Elasticity
Abstract We tested the hypothesis that a simple change in wall composition (medial calcium overload of elastic fibers) can decrease aortic elasticity. Calcium overload was produced by hypervitaminosis D plus nicotine (VDN) in the young rat. Two months later, measurement of central aortic mean blood pressure in the unanesthetized, unrestrained rat showed that the VDN rat suffered from isolated systolic hypertension but that mean blood pressure was normal. Wall thickness and internal diameter determined after in situ pressurized fixation were unchanged, as was calculated wall stress. Wall stiffness was estimated from (1) elastic modulus (determined with the Moens-Korteweg equation and values for aortic pulse wave velocity in the unanesthetized, unrestrained rat and arterial dimensions) and (2) isobaric elasticity (=slope relating pulse wave velocity to mean intraluminal pressure in the phenylephrine-infused, pithed rat preparation). Both increased after VDN, and both were significantly correlated to the wall content of calcium and the elastin-specific amino acids desmosine and isodesmosine. Left ventricular hypertrophy occurred in the VDN model, and left ventricular mass was related to isobaric elasticity. In conclusion, elastocalcinosis induces destruction of elastic fibers, which leads to arterial stiffness, and the latter may be involved in the development of left ventricular hypertrophy in a normotensive model.
Human aging produces many vascular changes, one of the most important being a progressive rise in arterial stiffness.1 2 Simultaneously, the geometry and composition of the aortic wall change, and generally, transmural pressure and wall stress increase. One of the major age-linked changes in wall composition is accumulation of calcium in the media.3 As the major site of calcium deposition is on the elastic fibers,4 calcium accumulation could be responsible for the progressive destruction of the integrity of the elastic retaining network observed with age.5 6 7 This may be a major cause of an age-linked pressure-independent increase in arterial stiffness.8
To study this phenomenon, we developed a rat model of calcium overload (model VDN) produced by treating young rats with vitamin D3 and nicotine.9 Such treatment rapidly produces a lasting 10- to 40-fold increase in the aortic calcium content, the major site of calcium deposition being on the medial elastic fibers near the lumen.10 Arterial elasticity is diminished, as shown by increased characteristic impedance and decreased systemic arterial compliance in anesthetized rats,11 increased pulse wave velocity and aortic elastic modulus in pithed rats,12 and impairment of the in vitro damping capacity of the common carotid artery.13
There is, therefore, a substantial body of evidence suggesting that in this model a marked increase in aortic stiffness can be produced by a simple change in wall composition. In this article, we investigated whether the other factors mentioned above (wall thickness and transmural pressure) might be involved in an ancillary fashion. First, we determined aortic geometry after in situ fixation under pressure. Second, although measurements of mean aortic pressure under anesthesia suggested that the VDN model was normotensive,14 in the present article we measured central aortic mean blood pressure in the unanesthetized, unrestrained rat. We evaluated the combined effects of arterial geometry and transmural distending pressure by calculating wall stress using the Lamé equation.
We attempted to evaluate the relative effect of aortic geometry and composition compared with that of transmural pressure by determining the isobaric elasticity, which we define as follows. Aortic pressure pulse wave velocity (an index of wall stiffness1 ) varies linearly as a function of distending pressure over a wide pressure range in vitro15 and in vivo.16 17 When aortic pulse wave velocity is measured in the pithed rat whose blood pressure is progressively raised from a low initial value (40 mm Hg) to normotension (120 mm Hg) by infusion of the α1-adrenoceptor agonist phenylephrine, a linear relationship between pulse wave velocity and arterial blood pressure is obtained.12 17 The slope of this regression is taken as the isobaric elasticity. Any change in this parameter reflects a change in the aortic geometry and/or composition of the aortic wall, which is independent of the instantaneous transmural distending pressure. Compared with echo wall tracking methods in anesthetized or conscious animals,18 19 this technique has the advantage that pulse wave velocity measurements are evenly distributed over the pressure range and a very wide pressure range is considered.
Having investigated the possible effect of ancillary factors, we then reconsidered our original hypothesis that calcium accumulation on elastic fibers causes their rupture and thus an increase in arterial stiffness. We have previously shown that VDN treatment produces a decrease in the relative percentage of the medial area occupied by fibers stained by Weigert’s solution, which is specific for elastin.12 This result is equivocal, however, as it could reflect an absolute decrease in elastin content or a relative increase in some other component of the media. To overcome this difficulty, we have recently developed a capillary electrophoretic method for measurement of desmosine and isodesmosine.20 These two amino acids are formed during the cross-linking of the alpha helices of the tropoelastin strands under the effect of lysyloxidase during elastin formation.21 In mammals, such cross-linking amino acids are found only in elastin.22 A decrease in the content of desmosine and isodesmosine in the media reflects a decrease in the amount of elastin, in the number of functional cross-linked elastic fibers, or in both.
Finally, as it has been suggested that an increase in aortic stiffness will increase afterload and so produce cardiac hypertrophy,23 we determined the relationship between ventricular mass and indexes of aortic stiffness (isobaric elasticity or elastic modulus calculated with the Moens-Korteweg equation and values for pulse wave velocity and arterial geometry).
Animals and Vitamin D3–Nicotine Treatment
Two-month-old (206±2 g, n=35) male outbred Wistar rats (Ico: WI, IOPS AF/Han) were purchased from Iffa Credo (L’Arbresle, France) and housed under standard conditions (temperature: 21±1°C; humidity: 60±10%; lights on 6 am to 6 pm). They were given standard rodent chow (UAR) and water ad libitum.
On day 1, 21 rats (group VDN) were injected with vitamin D3 (300 000 IU·kg−1 IM; cholecalciferol dissolved in an aqueous solution of benzyl alcohol, citric acid, propylene glycol, and vegetable oil; Duphafral D3 1000; Duphar BV) and nicotine (25 mg·kg−1, 5 mL·kg−1 PO; nicotine hydrogen tartrate, Sigma Chemical Co) at 9 am. The nicotine administration was repeated at 6 pm on the same day. Control rats (n=14) received an injection of 0.15 mol/L NaCl IM and two gavages of distilled water. Rats were then allowed to recover for 2 months, during which time food intake, water consumption, and body weight were measured at regular intervals. All experiments were performed in conformity with the legislation of the European Union.
Tail Systolic Arterial Pressure and Heart Rate
Tail systolic arterial pressure was measured by the tail-cuff method.24 Rats were placed in restraining cages and warmed at 38°C for 15 minutes. A cuff was then placed around the base of the tail and inflated, and a microphone was placed over the ventral tail artery distal to the cuff. The cuff was deflated, and tail systolic arterial pressure was defined as the pressure at which the pulsatile pressure signal reappeared. Heart rate was calculated from a fast-speed chart recording. For each rat, three measurements were taken and averaged.
Tail systolic arterial pressure was determined before vitamin D3–nicotine treatment and 2 months later before hemodynamic measurements.
Aortic Blood Pressure, Pulse Wave Velocity, Pulse Amplification, and Wave Reflections in Unanesthetized, Unrestrained Rats
Two months after the vitamin D3–nicotine treatment, a polyethylene cannula (0.96 mm OD, 0.58 mm ID, Portex SA) was introduced, with rats under halothane (2%)/oxygen anesthesia, through the left common carotid artery into the descending aorta (2 mm downstream of the ostium of the left common carotid artery) for measurement of central aortic blood pressure. A second cannula was placed into the abdominal aorta (via the right femoral artery, 5 mm upstream of the iliac bifurcation) for determination of peripheral aortic blood pressure. Cannulas were filled with heparinized (5 IU·mL−1) 0.15 mol/L NaCl and passed under the skin of the back to the level of the shoulder blades. Rats were then placed in individual cages and allowed 24 hours to recover from surgery. During this period, rats lost 7±1% (controls) and 7±1% (VDN rats) of their body weight. There were no deaths.
Twenty-four hours later, the cannulas were flushed and filled with gas-free, heparinized (5 IU·mL−1) 0.15 mol/L NaCl. Thirty minutes later, cannulas were connected to low-volume pressure transducers (Baxter, Bentley Laboratories Europe) with a 15-cm length of the same polyethylene cannula (0.96/0.58 mm) with two joints. The entire length of polyethylene tubing from the pressure transducer to the intravascular site was 35 cm. Signals were amplified, converted into digital form, and recorded on-line at a sampling rate of 256 Hz on a microcomputer (PC 4-486-66/L, Dell Computers). This gives a sampling rate of 36 per heart beat.
An algorithm detected the minimal (diastolic) and maximal (systolic) values of each pressure signal and calculated mean and pulse aortic blood pressures. Transit time (milliseconds) between the two pressure signals was calculated by another algorithm using least-squares analysis of the differences in the amplitudes of the central (C1 to j) and peripheral (P1 to j) signals at a given point in time. This was repeated keeping the same central amplitudes and using peripheral amplitudes of P1±x to j±x, where x=1 to 10 sampling points. After the determination of the best fit for x, intermediate points for the peripheral signal were created by interpolation between two sampling points. Least-squares analysis was repeated, keeping the same central amplitudes and using peripheral amplitudes of P1±x±k to j±x±k, where k=1 to 10 intermediate points. This calculation was performed for the entire waveform.
As the sampling rate was 1/3.9 milliseconds and 10 intermediate points were created, the final resolution of the calculation was ±0.39 millisecond, which gives a ±3.5% error in the transit time of a wave traveling at 754 cm·s−1 (as measured in VDN rats) over 8 cm (=11 milliseconds). Values were determined beat-to-beat over a period of 4 seconds and averaged (n=approximately 25 heart beats). Heart rate (beats per minute [bpm]) was determined by counting the number of pressure cycles over a period of 4 seconds. Measurements of aortic pressures, transit times, and heart rates were repeated every minute for 1 hour, and all values were averaged.
The dynamic frequency response of the whole recording system (cannula filled with gas-free saline solution plus transducer plus amplifier) was evaluated at room temperature with a sinusoidal wave generator.25 Variations in amplitude and phase lag were evaluated up to 30 Hz. Three independent experiments were performed and values averaged. The system was flat up to 15 Hz and then slightly underdamped. Phase lag in both central and peripheral pressure recording systems was not significantly different from zero at low frequencies but was slightly and significantly less than zero from 12 Hz onward (Fig 1⇓). Damping and phase lag were similar in both central and peripheral pressure recording systems, apart from a slightly larger phase lag in the peripheral pressure recording system at low frequencies.
Pulse wave velocity (centimeters per second) was calculated as the distance between the two cannula tips divided by the transit time. The distance between the two cannula tips was measured after in situ fixation (see below). A damp cotton thread was stuck onto the aorta between the tips of the cannulas, which were marked on the thread. It was then removed and laid straight for measurement of the distance between the two marks (8.5±0.2 and 8.0±0.1 cm in control and VDN rats, respectively; P<.05). When this distance was corrected for body weight (1.9±0.03 and 2.0±0.04 cm·100 g−1 in control and VDN rats, respectively; P>.05), there was no difference between the two groups.
Aortic pulse amplification was calculated as the peripheral aortic pulse pressure divided by the central aortic pulse pressure. This parameter is also called “transmission ratio” and reflects the normal progressive increase in aortic pulse pressure from central to more distal sites.26 It usually decreases with arterial stiffening.26
Wave reflections were measured on the central pressure signal. They were indistinguishable from the systolic peak of the peripheral pressure signal. All central pressure waveforms were of the type A as defined by Murgo et al.27 The following parameters were calculated (Fig 2⇓): (1) the height from the shoulder of the reflected wave to the systolic peak (ΔP, millimeters of mercury), and (2) the augmentation index, or the ratio of ΔP to pulse pressure (percent). The latter is considered to be a good estimate of the effect of wave reflection from peripheral sites on central aortic pressure waveforms. The travel time of the reflected wave (=time from the foot of the pressure wave to the shoulder, Δt, milliseconds) and left ventricular ejection time (=time from the foot of the pressure wave to the diastolic incisura, milliseconds) were also measured. For each rat, measurements were performed over 1 second (approximately six heart beats) and averaged.
Aortic Blood Pressure, Pulse Wave Velocity, and Pulse Amplification in Pithed Rats
After hemodynamic measurements in awake rats, animals were anesthetized again. A femoral venous catheter was implanted for continuous infusion of the α1-adrenoceptor agonist phenylephrine. Rats were then fitted with an endotracheal tube and pithed via the orbit as previously described28 except that atropine was not administered, the jugular veins were not tied off, and the vagi were not cut. Rats were ventilated with a rodent respirator (1.2 mL·100 g−1, 50 strokes per minute; rodent respirator 601, Harvard Apparatus).
The two arterial cannula were connected to the recording system described above, and after a 30-minute stabilization period, baseline hemodynamic parameters were measured. Rats were then infused with phenylephrine17 (40 to 80 nmol·kg−1·min−1 IV) so as to progressively raise aortic mean blood pressure from baseline to the level measured in the same unanesthetized, unrestrained rat. During this gradual increase in blood pressure, central and peripheral aortic blood pressures and transit times were measured every 30 seconds (approximately 40 observations) and at the end of the phenylephrine infusion; when central aortic mean blood pressure reached the level recorded before anesthesia and pithing, pulse wave velocity and pulse amplification were measured. These will be referred to as “normotensive isobaric” values.
Values for pulse wave velocity and pulse amplification (dependent variables) were expressed as a linear function of central aortic mean blood pressure (independent variable). Regression ANOVA was performed on the results of each individual rat for determination of slopes and intercepts, which were then treated as independent parametric variables and averaged. The slope relating pulse wave velocity to central aortic mean blood pressure was taken as the “isobaric elasticity.”
Aortic Wall Composition, Aortic Structure, Elastic Modulus, Wall Stress, and Cardiac Mass
At the end of the experiment, rats were perfused for 30 minutes at their unanesthetized, unrestrained central aortic mean blood pressure (measured before anesthesia and pithing) with 10% formol containing phosphate-buffered saline (120 mmol·L−1 NaCl, 2.7 mmol·L−1 KCl in a phosphate buffer [10 mmol·L−1], pH 7.4, at 25°C). A 0.5-cm sample of the thoracic descending aorta was removed and weighed. Tissue calcium content (micromoles per gram aortic dry weight) was determined by atomic absorption spectrophotometry (AA10, Varian Ltd) following dessication-mineralization and acid digestion (14 mol/L nitric acid) of the tissue.10
Another 1-cm sample of the thoracic descending aorta was excised and immersed in 10% formol. Samples were dehydrated and embedded in paraffin. Sections (20 μm thick) were cut and stained with hematoxylin-eosin for determination of aortic geometry (internal diameter and wall thickness) with the Saisam algorithm (Microvision Instruments). Sixteen sections of each aorta were prepared and placed 4 per slide and then independently examined by two research workers. ANOVA showed no significant effect of slide, section, or research worker (results not shown).
Elastic modulus (E, dynes per centimeter squared) was calculated according to the Moens-Korteweg equation: E=(PWV2·Di·ρ)/h, where PWV is pulse wave velocity (centimeter per second), Di is internal diameter (centimeters), ρ is 1.05 (density of blood), and h is wall thickness (centimeters).
Wall stress (dynes per centimeter squared) was calculated according to the Lamé equation: Wall Stress= (CAMBP·Di)/h, where CAMBP is central aortic mean blood pressure (dynes), Di is internal diameter (centimeters), and h is wall thickness (centimeters).
A 0.1-cm sample of the abdominal aorta was removed and weighed (1 to 2 mg). Aortic wall content of the elastin-specific cross-linking amino acids desmosine and isodesmosine was determined by capillary zone electrophoresis and UV detection after hydrochloric acid hydrolysis.20 Results are expressed as micrograms per gram aortic dry weight. This analysis was performed in 15 of 21 rats.
The heart was removed and the left ventricle dissected free and weighed. As body weights were very slightly but significantly different, results for left ventricular mass were expressed as the left ventricular weight (grams) divided by body weight (kilograms).
Results are expressed as mean±SEM. Linear regression was performed with standard techniques, and results are expressed as slope and intercept. For any given parameter, missing values per group were less than or equal to 2. Differences between groups were evaluated by ANOVA plus Bonferroni’s test. The null hypothesis was rejected at a probability level of 95% (P<.05).
Food Intake, Water Consumption, Body Weight, Tail Systolic Arterial Pressure, and Heart Rate
Starting body weights were not different (211±2 and 202±2 g in control and VDN rats, respectively) (Fig 3⇓). After vitamin D3–nicotine treatment, VDN rats lost weight during the first 8 days, and then growth rate was similar in both groups. After 2 months, body weight in the VDN group (403±9 g) was slightly (−10%) lower than that of the control group (449±13 g, P<.05).
Food intake and water consumption were lower for the first 10 days after vitamin D3–nicotine treatment in the VDN group (Fig 3⇑) and then became slightly greater. From day 30 to day 60, food intake and water consumption did not differ significantly between the groups.
Starting tail systolic arterial pressures (controls: 128±4 mm Hg; VDN: 125±3 mm Hg) and heart rates (controls: 440±17 bpm; VDN: 412±8 bpm) were similar. Two months after vitamin D3–nicotine treatment, tail systolic arterial pressure increased in VDN rats (151±5 mm Hg) compared with controls (136±5 mm Hg, P<.05); heart rate was unchanged (controls: 446±17 bpm; VDN: 445±15 bpm).
Aortic Calcium and Desmosine Plus Isodesmosine Contents and Cardiac Mass
Vitamin D3–nicotine treatment produced a 15-fold increase in thoracic aortic calcium content in VDN rats compared with controls (controls: 24±2 μmol·g−1; VDN: 318±85 μmol·g−1, P<.05). Vitamin D3–nicotine treatment decreased aortic desmosine plus isodesmosine content (−55%) (controls: 1243±83 μg·g−1; VDN: 555±36 μg·g−1, P<.05). Aortic desmosine plus isodesmosine content (dependent variable) was negatively correlated to aortic calcium content (independent variable, Fig 4⇓) within the VDN group.
Left ventricular mass significantly increased after vitamin D3–nicotine treatment (1.73±0.06 and 1.99±0.08 g·kg−1 in control and VDN rats, respectively; P<.05). There was no significant correlation between left ventricular mass (dependent variable) and aortic elastic modulus (independent variable) within the VDN group. Values for the linear regression were as follows: slope=0.004±0.013; intercept=1.94±0.19; and P=.7846. However, left ventricular mass (dependent variable) was significantly correlated to isobaric elasticity (independent variable) in the VDN rats (Fig 5⇓).
Aortic Blood Pressure, Pulse Wave Velocity, Pulse Amplification, and Wave Reflections in Unanesthetized, Unrestrained Rats
VDN rats were not “hypertensive”: aortic mean blood pressures (central and peripheral) were not increased (Table 1⇓). There were significant increases in central and peripheral systolic pressures (+11% to 14%, respectively) and decreases in central and peripheral diastolic pressures (−12% to 11%, respectively) in VDN rats compared with controls. In control rats, peripheral aortic pulse pressure was larger than central aortic pulse pressure because of an increase in systolic pressure and decrease in diastolic pressure. This gave a 37% increase in pulse pressure from the central to the peripheral site. In VDN rats, the increase in central pulse pressure (69%) was greater than that in peripheral pulse pressure (57%); this resulted in a 10% decrease in pulse amplification compared with controls (Table 1⇓, Fig 6⇓, bottom).
Aortic Blood Pressure, Pulse Wave Velocity, and Pulse Amplification in Pithed Rats
Continuous infusion of phenylephrine produced a gradual increase in arterial blood pressure and pulse wave velocity. An example of the linear relationship between central aortic mean blood pressure (independent variable) and pulse wave velocity (dependent variable) in both groups is shown in Fig 7⇓. In VDN rats, the slope (=isobaric elasticity) was steeper than in controls (+92%, Table 2⇓), but the intercept was similar.
Within the VDN group, isobaric elasticity (dependent variable) was significantly correlated to aortic calcium content (independent variable, Fig 8⇓), but linear regression with aortic desmosine plus isodesmosine content failed to reach significance (Fig 9⇓). Normotensive isobaric pulse wave velocity in the phenylephrine-infused pithed rat was not significantly different from the value measured in the awake rat (Fig 6⇑, top).
There was a linear relationship between central aortic mean blood pressure (independent variable) and pulse amplification (dependent variable, Fig 7⇑). The slope of this linear regression was lower in VDN rats (−38%, Table 2⇑), whereas the intercept was tripled. Normotensive isobaric pulse amplification in the phenylephrine-infused pithed rat was similar to that measured in the awake rat (Fig 6⇑, bottom).
Aortic Geometry, Elastic Modulus, and Wall Stress
Thin wall conditions applied, as the global value for wall thickness/external radius was 0.088±0.02 (n=35). Aortic internal diameter, the ratio of wall thickness to internal diameter, and wall stress were similar in both groups (Table 3⇓).
Elastic modulus was doubled in VDN (+98%, Table 3⇑) compared with control rats. There was a significant linear relationship between elastic modulus (dependent variable) and aortic calcium content and aortic desmosine plus isodesmosine content (independent variables, Figs 8⇑ and 9⇑).
Among the many factors that modify arterial elasticity and are mentioned above, transmural distending pressure per se is probably not a major determinant in the VDN model, as aortic mean blood pressures (both central and peripheral) were not different from those of controls. Furthermore, increased wall thickness relative to diameter can also be eliminated, as aortic geometry was similar in VDN and controls.
As wall thickness is unchanged, intrinsic changes in wall composition and/or increased smooth muscle cell activation can be invoked as explanations of the decreased elasticity in the VDN model. Increased smooth muscle tone seems to be of minor importance in this model. The increase in in situ common carotid artery compliance after inactivation of smooth muscle cells with cyanide was negligible.11
Elimination of the above factors leads us to the conclusion that the decrease in wall elasticity in the VDN model stems from a change in the extracellular matrix, with a relative decrease in the amount of wall elements with a low elastic modulus (elastin and smooth muscle) and/or a relative increase in the amount of wall elements with a high elastic modulus (collagen). This article deals with elastin and, more specifically, the process of elastocalcinosis. It should be borne in mind, however, that VDN may produce smooth muscle cell necrosis29 30 and that the decrease in elastin content observed during vascular aging is accompanied by fibrosis.31 32
In VDN rats, although the number of animals was not large (n=21), we found a significant correlation between aortic calcium content (independent variable) and indexes of aortic elasticity (elastic modulus and isobaric elasticity). It has been known since the early work of Lansing and others4 10 33 34 that calcium will fix preferentially on elastic fibers. Such calcification could make the elastic fibers stiffer without causing their destruction. However, using quantitative, two-dimensional histomorphometry, we have previously reported a decrease in the percentage of elastin in the (thoracic) aortic wall.12 Our present results, using a new capillary electrophoretic method, showed a lower content of elastin-specific cross-linking amino acids, desmosine and isodesmosine, in the (abdominal) aortic wall of the VDN rats. Such evidence suggests that although elastin measurements were made at different levels of the aorta, there is a decrease in the aortic elastin content in VDN rats. Moreover, we found a significant inverse correlation between aortic calcium and desmosine plus isodesmosine contents. The facts that (1) calcium and desmosine plus isodesmosine contents were measured at different levels of the aorta (thoracic and abdominal, respectively) and (2) elastin content is lower in the abdominal aorta than in the thoracic aorta may explain why a rather weak correlation was obtained. It may be assumed that a closer relationship would have been observed had the desmosine plus isodesmosine been measured in the thoracic aorta. All these observations suggest that in this rat model, arterial calcium overload is at least partly responsible for the increased aortic stiffness via progressive destruction of the elastic network followed by a change in the functional properties of the elastic fibers.
The increase in aortic stiffness in the awake VDN rat was confirmed by results in pithed VDN rats: The slope relating pulse wave velocity to central aortic mean blood pressure (isobaric elasticity) was steeper than in controls, but the intercept was not modified. A direct effect of phenylephrine on the aorta can be eliminated as isobaric pulse wave velocity calculated in the phenylephrine-infused pithed rat was not different from that measured at the same central aortic mean blood pressure before pithing. These results suggest that the increased slope observed in VDN rats was not related to a direct effect of phenylephrine on the aorta but reflects structural changes of the aortic wall. A given increase in intravascular pressure produced a proportionally greater increase in pulse wave velocity in VDN rats. Interestingly, Dobrin and coworkers35 36 showed that in an in vitro elastase-treated common carotid artery preparation, the slope relating strain to stress was steeper after enzymatic elastolysis, whereas the intercept was not changed. The similarity between these results in vitro and ours in vivo suggests that the main factor responsible for the increased isobaric elasticity in VDN rats is destruction of the elastic network. The possibility that calcium loading of elastic lamellae can produce a structural change is currently being investigated with the use of confocal microscopy.
Although aortic mean blood pressure was not increased, VDN rats exhibited isolated systolic hypertension. Stroke volume is not increased in this model.11 Furthermore, pulse amplification was lower in VDN rats. These changes could arise from a decrease in arterial compliance and/or a change in the timing and intensity of wave reflections.26 37 The latter does not appear to be involved, as parameters linked to wave reflections (augmentation index and Δt) were unchanged in VDN rats. It is possible that as the central blood pressure waveform of the rat is basically of the A type,27 a change in waveform following arterial stiffening (apart from an increase in pulse pressure and ΔP) will be hard to detect. One can also not exclude the possibility that the pressure recording system used does not allow accurate detection of the inflection point. The accurate detection of the inflection point defining the start of the augmented pressure in systole requires the presence of at least the seventh or eighth harmonic, ie, a frequency component of some 50 Hz. It is not clear what the frequency response of the manometer might be at 50 Hz. However, if a second-order underdamped response is assumed, the components at 50 Hz may well be severely attenuated. Thus, if the inflection point cannot be determined accurately, the comments regarding wave reflection based on the augmentation index calculation may not be totally valid.
In pithed rats, there was a linear relationship between pulse amplification (dependent variable) and central aortic mean blood pressure (independent variable), the slope of which was lower in VDN rats, whereas the intercept was greater. This relationship between pulse amplification and central aortic mean blood pressure reveals that pulse amplification was lower in VDN rats compared with controls at normotension. At low pressures (<80 mm Hg), however, amplification was greater in VDN than control rats. In addition, pulse amplification became less than 1 at values of central aortic mean blood pressure less than 70 to 75 mm Hg in controls and values less than 60 to 65 mm Hg in VDN rats. This suggests that at low pressures the pressure pulse is damped as it travels from the central to more distal sites. Such damping is impaired in VDN rats at low pressures when strain is borne mainly by elastin. This is in line with our hypothesis that the major factor responsible for altered pulse amplification in this rat model is decreased distensibility after impairment of elastic properties.
Left ventricular mass was 16% greater in VDN rats than controls. No significant correlation was found between left ventricular mass (dependent variable) and aortic elastic modulus (independent variable). However, left ventricular mass was significantly related to isobaric elasticity. According to our working hypothesis, calcium deposition on the elastic fibers leads to an increase in aortic stiffness (and cardiac afterload), which in turn would produce compensatory cardiac hypertrophy. Thus, cardiac mass may be related to elasticity when the latter is measured at pressures lower than normotension, where elastin will play a predominant role (isobaric elasticity), and not to elasticity measured at normotension (elastic modulus). The relationship between cardiac mass and different indexes of arterial elasticity and/or compliance may evolve with time, as we previously found that cardiac mass was linked to elastic modulus in “young” (30 days old) but not in “older” (3 months old) VDN rats,12 comparable to the situation in this article.
In conclusion, calcium overload of medial elastic fibers leads to destruction of the elastic network, which decreases aortic elasticity. This in turn produces cardiac hypertrophy, with no change in aortic mean blood pressure. We suggest that the VDN model could provide data on one aspect of the age-linked vascular (and cardiac) dysfunction in human beings, namely, calcium overload of compliance arteries.
This laboratory is a member of the EHC Biomed project “Eureca” BMHI-CP94-1375.
- Received August 2, 1996.
- Revision received September 25, 1996.
- Accepted October 18, 1996.
Milnor WR. Hemodynamics. Baltimore, Md: Williams & Wilkins;1989:77-79.
Milnor WR. Cardiovascular Physiology. New York, NY: Oxford University Press; 1990:52-54.
Fleckenstein A, Frey M, Zorn J, Fleckenstein-Grün G. Calcium, a neglected key factor in hypertension and arteriosclerosis. In: Laragh JH, Brenner BM, eds. Hypertension: Pathophysiology, Diagnosis and Management. New York, NY: Raven Press; 1990:471-509.
Lakatta EG. Cardiovascular aging. In: Masoro EJ, ed. Handbook of Physiology, Section 11: Aging. New York, NY: Oxford University Press; 1995:413-474.
Robert L, Jacob MP, Fülöp T. Elastin in blood vessels. In: Ciba Foundation, ed. The Molecular Biology and Pathology of Elastic Tissues. Chichester, UK: John Wiley & Sons; 1995:286-299.
Ooyama T, Sakamoto H. Elastase in the prevention of arterial ageing and the treatment of atherosclerosis. In: Ciba Foundation, ed. The Molecular Biology and Pathology of Elastic Tissues. Chichester, UK: John Wiley & Sons; 1995:307-317.
Atkinson J. Vascular calcium overload: physiological and pharmacological consequences. Drugs. 1992;44(suppl 1):111-118.
Atkinson J, Poitevin P, Chillon JM, Lartaud I, Levy B. Vascular calcium overload produced by vitamin D3 plus nicotine treatment diminishes arterial distensibility in rats. Am J Physiol. 1994;266:H540-H547.
Tatchum-Talom R, Niederhoffer N, Amin F, Makki T, Tankosic P, Atkinson J. Aortic stiffness and left ventricular mass in a rat model of isolated systolic hypertension. Hypertension. 1995;26:963-970.
Boutinet S, Giummelly P, Capdeville-Atkinson C, Atkinson J. Role of calcium in the alterations of damping capacity in a rat model of elastocalcinosis. Br J Pharmacol. 1995;116:276P. Abstract.
Bazett HC, Dreyer NB. Measurements of pulse wave velocity. Am J Physiol. 1922/1923;63:94-116.
Hayoz D, Nussberger J, Weber B, Brunner HR. The renin-angiotensin system and arterial wall behavior. J Cardiovasc Pharmacol. 1993;22(suppl 5):S48-S52.
Van Gorp AD, van Ingen Schenau DS, Willigers J, Hoeks APG, De Mey JGR, Struyker Boudier HAJ, Reneman RS. A technique to assess aortic distensibility and compliance in anesthetized and awake rats. Am J Physiol. 1996;270:H780-H786.
Peslin R, Fredberg JJ. Oscillation mechanics of the respiratory system. In: Macklem PT, Meads J, eds. Handbook of Physiology, Section 3: Mechanics of Breathing. Bethesda, Md: American Physiology Society; 1986:145-177.
Nichols WW, O’Rourke MF. McDonald’s Blood Flow in Arteries. London, UK: Edward Arnold; 1990:216-250.
Murgo JP, Westerhof N, Giolma JP, Altobelli SA. Aortic input impedance in normal man: relationship to pressure wave forms. Circulation. 1980;62:105-116.
Shipley RE, Tilden JH. A pithed rat preparation suitable for assaying pressor substances. Proc Soc Exp Biol N Y. 1947;64:453-455.
Fleckenstein A, Frey M, Fleckenstein-Grün G. Protection by calcium antagonists against experimental arterial calcinosis. In: Pyörälä K, ed. Secondary Prevention of Coronary Heart Disease, Workshop of the International Society and Federation of Cardiology. New York, NY: Georg Thieme Verlag; 1983:109-122.
Fleckenstein-Grün G, Frey M, Thimm F, Hofgärtner W, Fleckenstein A. Calcium overload: an important cellular mechanism in hypertension and arteriosclerosis. Drugs. 1992;44(suppl 1):23-30.
Cox RH. Effects of age on the mechanical properties of rat carotid artery. Am J Physiol. 1977;2:H256-H263.
Michel JB, Heudes D, Michel O, Poitevin P, Philippe M, Scalbert E, Corman B, Levy B. Effect of chronic ANG I-converting enzyme inhibition on aging processes, II: large arteries. Am J Physiol. 1994;267:R124-R135.
Lansing AI, Rosenthal T, Alex M. Significance of medial agechanges in the human pulmonary artery. J Gerontol. 1950;5:211-215.
Dobrin P, Canfield T. Identification of smooth muscle cell series elastic component in intact carotid artery. Am J Physiol. 1977;232:H122-H130.
Avolio A. Ageing and wave reflection. J Hypertens. 1992;10(suppl 6):S83-S86.