Volume 32, Issue 2 , Pages 129-135, August 2006
Endografting of the Descending Thoracic Aorta Increases Ascending Aortic Input Impedance and Attenuates Pressure Transmission in Dogs
Article Outline
- Abstract
- 1. Introduction
- 2. Methods
- 3. Results
- 4. Discussion
- 5. Conclusions
- Acknowledgements
- References
- Copyright
Abstract
Objectives
Endografting is being used to manage aneurysms, dissections and acute traumatic disruptions of the thoracic aorta. The acute effects of such interventions on ventricular afterload and on pressure wave transmission characteristics are not well known.
Methods
In five dogs, a 55
mm endograft was introduced into the descending aorta, just distal to the left subclavian artery, with oversizing of 20%. Following formaldehyde induced complete heart block, the hearts were paced (30–120
bpm). The ascending aortic pressures and flows were recorded using Millar micro-tip manometers and ultrasonic flowmeters, respectively. Arterial pressures proximal and distal to the stent site were also recorded. For each heart rate, parameters of a modified Windkessel (SVR: systemic vascular resistance, Z0: characteristic impedance, C: total arterial compliance) were estimated. The pulse wave velocity (PWV) and reflection coefficient (Γ) were calculated from the pressure wave transfer functions.
Results
The Z0 (0.25±0.05 vs 0.41±0.06
mmHg/ml
s−1, P<.05) was increased and C was decreased (0.45±0.07 vs 0.28±0.04
ml/mmHg, P<0.001) following endograft placement. SVR tended to increase (P=.06) and ascending aortic Γ was unchanged. The PWV increased (418±67 vs 755±135
cm/s, P<.05) and the distal Γ decreased (0.09±0.10 vs −0.49±0.07, P<.05).
Conclusions
Endografting in the proximal descending aorta cause unfavorable changes in the ascending aortic input impedance and an increase in the PWV through the grafted segment, consistent with an increase in the modulus of elasticity. The grafts produce a negative Γ at the distal end, an uncommon occurrence in the systemic circulation. Whether this change is of sufficient magnitude to result in post-graft dilation is unknown.
Keywords: Stents, Pressure wave transmission, Arterial impedance, Pulse wave velocity
1. Introduction
Endografting of the descending thoracic aorta has been successfully utilized in the management of a number of pathologies including aneurysms, dissections, and traumatic disruption.1 The outcome is largely dependent upon the interaction of the endovascular device and the aortic wall at the zones of fixation. A tight, durable seal to isolate the aortic pathology from pressure and flow is the ultimate goal of these techniques. Failure to achieve this initial technical success, or subsequent dilatation of the aorta at the landing zones2 leads to endoleak, that compromises the integrity of the endovascular reconstruction. The extent to which the compliance mismatch3 induced by the insertion of the stent might contribute to subsequent dilation is unknown, although cases of proximal aortic dilatation after open insertion of a Dacron graft in the ascending aorta have been described.4 The altered shear stresses at the compliance-mismatched graft–aortic interface could contribute to the development of neo-intimal hyperplasia as observed with lower extremity bypass, or to local damage at the aortic wall, with subsequent dilation, ulceration, or pseudoaneurysm formation.5
To date, the in vivo characterization of the compliance mismatch subsequent to endovascular stenting of large arteries has been qualitative.6, 7 A simple means of quantifying the effect of stent insertion on compliance mismatch through the measurement of pulse wave velocity (PWV) and the reflection coefficient (Γ) has been described.8 This approach relies on principles similar to those used in the mathematical modeling of the local hemodynamics induced through stent insertion.9
Similarly, the effect of endografting of the descending aorta on ascending aortic hemodynamics is unknown. The insertion of Dacron prosthesis in the descending aorta during the repair of traumatic disruption has been shown to increase characteristic impedance (stiffness) of the ascending aorta.10 This increase in stiffness can potentially reduce cardiac efficiency11 and may predispose to the development of left ventricular hypertrophy.12
In this study, the effects of stenting of the descending thoracic aorta on the pressure–flow relationship in the ascending aorta are described and the pressure wave transmission across the stented segment of the aorta is characterized.
2. Methods
2.1. Animal preparation
Five dogs (20–27
kg; either sex) were anesthetized with 30
μg
kg−1
h−1 of fentanyl citrate while ventilated with a 2:1 nitrous oxide-to-oxygen mixture using a constant-volume respirator set to deliver a tidal volume of 15
ml
kg−1
min−1. Temperature was maintained at 37
°C by a circulating-water warming blanket. A mid-line sternotomy was performed with the dog in the supine position. The pericardium was opened sufficiently to instrument the heart. Complete heart block was induced through injection of formaldehyde into the atrio-ventricular node. The heart was then paced at 90
bpm. Pressure in the ascending aorta (PASC), the descending aorta immediately distal to the left subclavian artery and above the proximal landing zone (PPROX) and the descending aorta 6.0
cm farther, just distal to distal landing zone (PDIST) were measured using catheters introduced through the right carotid, left carotid and left femoral arteries, respectively. PASC and PDIST were measured with 8-Fr micromanometer-tipped catheters with reference lumens (PC-480 Millar Instruments) while PPROX was measured with a 3-Fr micromanometer-tipped catheter (SPR-524 Millar). Flow through the ascending aorta (QASC) was measured using an ultrasonic flow meter (Transonic Systems, Ithaca, NY, USA). A laparotomy was performed and the aortic bifurcation was cannulated with a 6-Fr sheath to allow passage of the amplatz extra-stiff deployment guidewire. A single limb ECG was also recorded. Conditioned signals (model VR16, Electronics for Medicine/Honeywell, Pleasantville, NY, USA) were recorded by means of a computer using data-acquisition software (Sonometrics). The analog signals were passed through anti-aliasing, low-pass filters with a cut-off frequency of 100
Hz and were then sampled at a frequency of 500
Hz. Digitized data were subsequently analyzed with specialized software (CVSOFT, Odessa Computer Systems, Calgary, Alberta, Canada).
All animal experiments complied with the ‘Guide for Care and Use of Laboratory Animals’, Institute of Laboratory Animal Resources, Commission on Life Sciences, National Research Council and had been approved by the Animal Care Committee of the University of Calgary.
2.2. Calibration
The micromanometers in the aorta were referenced during diastole to the fluid-referenced micromanometer in the ascending aorta. The difference in timing between the flow wave and pressure wave measured in the ascending aorta were corrected by aligning maximum dP/dt with maximum dQ/dt.13 This compensated for the spatial separation between transducers and the phase delay introduced by the flow meter.
2.3. Experimental protocol
All hemodynamic measurements were made with the respirator stopped in the end-expiratory position for not more than 20
s. The control measurements were a series of 10
s recordings with the heart paced at rates varying, by increments of 10, from 30 to 120
bpm. The heart was paced at 90
bpm between recordings. Following the control measurements the distal micromanometer was withdrawn to the femoral artery and a Zenith-Cook, Gianturco stent (ESLE 10-55 or 12-55) with 20% oversizing was introduced over the wire through the aortic bifurcation. The stent was advanced into the descending aorta, just distal to the left subclavian artery, and deployed under fluoroscopic guidance. Complete apposition of the endograft to the aortic wall was confirmed with angiography. The distal micromanometer was advanced to its previous location (confirmed with fluoroscopy) and the aorta was ligated at the bifurcation. Following a 10
min stabilization period, the post-stent data series was recorded.
2.4. Analysis
2.4.1. Ascending aortaThe ascending aortic input impedance (AAII) was calculated from the discrete Fourier transforms of the simultaneously recorded PASC and QASC waves. The modulus and the phase of the impedance spectrum for each series were collated into 1
Hz frequency bins for statistical analysis.
The AAII was also modeled as a lumped three-element windkessel. The windkessel parameters were characteristic impedance (Z0), systemic vascular resistance (SVR) and total arterial compliance (C). Z0 was calculated as the average of the moduli from 5 to 12
Hz. Systemic vascular resistance (SVR) was calculated as mean pressure divided by mean flow. Total arterial compliance (C) was calculated from Z0 and the corner frequency (the frequency at which the impedance modulus is 3
dB greater than the modulus of Z0).14
The pressure wave transfer functions (PWTF) across the descending aorta were calculated from the discrete Fourier transforms of the simultaneously recorded PPROX and PDIST (Fig. 1). The PWTF was then used to calculate the apparent phase velocities (Cph) and the pulse wave velocity (PWV) was taken as the average value of Cph between 5 and 12
Hz.15 An estimate of the global reflection coefficient (Γ) was calculated from the low frequency phase velocities of the PWTF and the PWV.8 The apparent attenuation coefficient (AAC=ln(modulus)) of the PWTF for each series were collated into 1
Hz frequency bins for statistical analysis.16

Fig. 1.
Pressure recordings across the descending aorta at control (a) and following endovascular stenting (b). There is a visible decrease in the transit time and attenuation of the distal pressure post-stenting.
2.5. Statistics
Using a two-tailed paired Student's t-test where P<.05 was considered significant, a number of comparisons were made, before and after stent insertion. The phase and moduli of the AIII and the AAC of the PWTF were compared by frequency bins. The windkessel parameters (SVR, Z0 and C), PWV and Γ were also compared.
3. Results
The data for the ascending aorta analysis were complete for all five dogs. There were technical problems with the PDIS for one dog and it was not included in the analysis of the PWTF.
The mean aortic pressure (SD) was 83 (14) mmHg before, and 81 (22) mmHg after implantation of the stent (P=.69). The mean flow (SD) was 0.89 (.18) l/m before, and 0.66 (.23) l/m following implantation (P=.09).
Stent implantation increased the modulus of the AAII at all non-zero frequency bins below 5
Hz (Fig. 2). There was no statistically significant change in the phase shift of the AAII.
The lumped windkessel parameters of the AAII were altered following stent implantation (Table 1). Z0 was increased by 65% and C was decreased by 40%. SVR increased following stent implantation, though the difference did not reach statistical significance (P=.06).
Table 1. Hemodynamic parameters of the ascending aorta
| Control | Post-stent | |
|---|---|---|
| Z0 (mmHg/ml | .250 (.052) | 414 (.061)* |
| Compliance (ml/mmHg) | .446 (.072) | 276 (.041)* |
| SVR (mmHg/ml | 5.8 (1.2) | 7.7 (1.7) |
*P<.05 vs control. |
Implantation increased the PWV across the stented segment of the aorta and generated a negative reflection at the distal end of the stent (Table 2). The stent produced a frequency-dependent reduction in the AAC of the PWTF (Fig. 3) which reached statistical significance at 4.5
Hz.
Table 2. Transmission characteristics of the descending aorta
*P<.05 vs control. |

Fig. 3.
AAC of the pressure wave transfer function of the descending aorta. Data are means (SE). *P<.05 vs control.
4. Discussion
4.1. Ascending aortic input impedance
This is the first report in which the acute effect of endografting on ascending aortic input impedance has been quantified. Endografting of the descending aorta in these dogs results in a 30% increase in the modulus of the AAII at frequencies below 5
Hz, a 65% increase in Z0, and a 40% decrease in aortic compliance. An increase in Z0 is usually attributed to an increase in local aortic stiffness,17 suggesting that the stenting resulted in an increase in the modulus of elasticity of the ascending aorta. The increase in Z0 following stent placement is similar in magnitude to that reported in patients following an aortic interposition graft of the descending thoracic aorta,10 and less than the 90% increase in Z0 seen following acute interposition grafting of the descending thoracic aorta in swine.18 The authors of the latter paper suggested that sensors in the transverse arch may modulate the heart's compensatory response, resulting in a decrease in cardiac output, as compared with a total arch reconstruction. The increase in Z0 may, therefore, be a consequence of smooth muscle activation, as has been previously described and has been attributed to stimulation of the sympathetic nervous system.17
A similar effect of endografting on the pressure–flow relationship of arteries remote from the stented artery has been described by Rolland et al.7 While their method of analysis precludes direct comparison, analysis of the published pressure–flow loops13 reveals that the arterial segment proximal to the stent (Cragg and Palmaz-Schatz) demonstrated an increase in stiffness relative to the arterial segment at the distal end. In addition, the pressure–flow relationship of the iliac arteries contralateral to these iliac stents was altered, suggesting the hemodynamic disturbance is not restricted to the stented segment.
The stent-induced decrease in compliance was smaller than that reported following Teflon banding of the entire aortic arch, including the proximal descending aorta (40 vs 75%).19 The Teflon banding also resulted in a greater increase in Z0 and SVR than did the stent. The acute changes following banding were attenuated at 2 days post-op. Whether the post-stenting changes in AAII persist is unknown, though the changes in the stent-artery mechanics progress up to 4 weeks post-implantation due to ongoing fibrosis of the arterial wall,3 and the subsequent increase in the stiffness of the stented arteries persists to at least 3 months.20
In a chronic dog model, larger and isolated increases in Z0 have resulted in left ventricular hypertrophy.12 The reduction in compliance following aortic stenting would be expected to have greater impact on stroke volume,14 myocardial oxygen consumption11 and left ventricular load than that seen with an isolated increase in Z0.
4.2. Pressure wave transfer function
Endografting of the descending thoracic aorta resulted in a 80% increase in the PWV across the stented segment, the development of a negative Γ at the distal end of the stent, and a decrease in the AAC across the stent. These results are due to the increased stiffness of the stent–aortic unit and the impedance mismatch at the distal end.
The change in PWV implies an increase in the modulus of elasticity of the stented aorta, according to the Moens–Korteweg equation. A decrease in the coefficient of distensibility following stenting of the rabbit aorta has previously been reported.21 This change would be expected to increase PWV22 across the stented segment by 50% on average, with a stent dependent range of 30–70%.
The use of a covered-stent has a profound effect on the blood supply to the media of the stented aorta.23 Interruption of the vaso vasorum has an effect of similar magnitude on the blood supply to the media and produces an acute decrease in the coefficient of distensibility of the aorta.24 These changes in distensibility are not of sufficient magnitude (36%) to explain the increase in PWV. The increase in PWV is probably a consequence of multiple factors including medial ischemia, dilation of the aorta due to over-sizing and the inherent mechanical properties of the stent. The relative contribution of each of these factors may be important in determining the ultimate success of compliance matching stents25 in attenuating the local hemodynamic effect.
The change in Γ across the stent was not expected because the positively reflected waves at the proximal end should be cancelled by the negatively reflected waves returning from the distal end.9 With the increase in stiffness of the ascending aorta subsequent to stenting, the proximal reflection was reduced, unmasking the distal negative impedance mismatch. Whether the negative reflection at the distal end is of sufficient magnitude to induce damaging changes in shear stress is unknown, though in vitro assessment of coronary artery stents demonstrated flow instabilities 1
cm downstream of the stent, with turbulent intensities of 5–20%, depending upon the state of the stented vessel.26 Also, a study of carotid artery stenting demonstrated that stenosis of the native carotid arteries was prevented through the addition of compliance matching cuffs at the distal ends of the stents.27
The reduction in the AAC at the mid-frequencies can be explained through the altered mechanical properties of the stent–aorta unit9 and the presence of a negative Γ at the distal end of the stent. An in vitro model of endografting of an abdominal aortic aneurysm demonstrated loss of laminar flow with propagating vortices along the length of the stent.28 The loss of laminar flow, if of sufficient magnitude, would also account for some of the attenuation of the AAC. These changes to the AAC occurred in spite of the occlusion of the aorta at the bifurcation. The occlusion would be expected to increase the reflection coefficient in the thoracic aorta. This will affect the transfer function across the stent, reducing the described changes.
As younger patients are increasingly exposed to this technology (e.g. endograft repair for blunt traumatic aortic arch injury), the long-term implications of these hemodynamic and aortic wall changes assume even greater importance. While the changes seen with endografting are similar in magnitude to those described for open interposition grafts, the ‘tethering’ effect of the sutured anastamosis following open graft repair may limit the effect of local graft–arterial wall interactions and compliance mismatch on post-repair aortic remodeling. This is not the case with endograft reconstructions.
4.3. Stent characteristics
The Gianturco stent used in this study (Cook ELSE 10-55) is one of over 100 stents currently available for vascular or non-vascular use. Its classification, by design and engineering characteristics,29 is that of a self-expanding, full hard stainless steel, wire, individual ring stent. While there is yet no biological assessment of stent design,30 it has been suggested that under physiological conditions stent stiffness can be added linearly to the stiffness of the overlying artery.31 Stent stiffness could, therefore, be used to approximate changes in the modulus of elasticity of both normal and diseased arteries following stenting.
Within a group of commercially available stents, the Gianturco stent's response to circumferential loading, measured by the slope of the stress–length relationship, demonstrated the least resistance to deformation and was completely elastic.32 The Gianturco stent used in our study has been previously shown to have a coefficient of stiffness similar to that of the Nitinol stents, though the methodology was different.33 Regardless, the nominal stiffness34 of these stents is in the range of 10–30% of that described for the thoracic aorta of the dog,15 and is of insufficient magnitude to explain the changes seen in our study and in others.21 This suggests the stent's effect on aortic stiffness is not simply additive.
5. Conclusions
The implantation of endovascular devices in the proximal descending aorta results in an increase in left ventricular afterload. The described changes following endografting may have long term consequences in younger patients undergoing stenting for traumatic disruption of the aorta, in patients with coronary artery disease and in patients in whom left ventricular systolic reserve is reduced. Any significant changes in cardiac function induced through endografting should be demonstrable through the use of cardiac magnetic resonance imaging35 or multi-detector row CT imaging.36 These modalities could be used in human studies to characterize the hemodynamic consequences of stent implantation in diseased aortas.
The change in the local hemodynamics of the stented aorta may not be reliably predicted through mathematical or in vitro modeling unless the response of the adjacent aorta to the intervention is included. Careful quantification of the interaction in vivo is required to characterize this response and to fully determine the potential deleterious effects of stenting. It is possible that these effects could be demonstrated in humans using echocardiography37 or magnetic resonance imaging.38 Information regarding aortic wall–endograft interaction could be used to develop ‘compliance-matched’ stents that may result in better long-term durability and more effective endovascular repair.
Acknowledgements
We acknowledge the excellent technical support received from Ms Rozsa Sas and Ms Cheryl Meek.
References
- Endovascular repair of thoracic aortic lesions with the Zenith TX1 and TX2 thoracic grafts: intermediate-term results. J Vasc Surg. 2005;41(4):589–596
- Complications of endovascular repair of high-risk and emergent descending thoracic aortic aneurysms and dissections. J Vasc Surg. 2004;40(2):228–234
- . Changes in arterial wall compliance after endovascular stenting. J Vasc Surg. 1919;905–911[see comment]
- Prosthetic replacement of the aorta is a risk factor for aortic root aneurysm development. Ann Thorac Surg. 2002;73(2):455–459
- . Is compliance mismatch the major cause of anastomotic arterial aneurysms? Analysis of 42 cases. J Cardiovasc Surg. 1985;26(2):147–150
- . Gosling's Doppler pulsatility index revisited. Ultrasound Med Biol. 1998;24(4):597–599
- Hemodynamics and wall mechanics after stent placement in swine iliac arteries: comparative results from six stent designs. Radiology. 1999;213(1):229–246
- . A method for estimation of intensity of wave reflection in an elastic tube. Cardiovasc Eng. 2004;4:229–236
- . Effects of stent stiffness on local haemodynamics with particular reference to wave reflections. J Biomech. 2004;37(3):339–348
- . Effect of an inelastic aortic synthetic vascular graft on exercise hemodynamics. Ann Thorac Surg. 1995;59(4):981–999
- . Effect of reduced aortic compliance on cardiac efficiency and contractile function of in situ canine left ventricle. Circ Res. 1992;71(3):490–502
- . Inelastic vascular prosthesis for proximal aorta increases pulsatile arterial load and causes left ventricular hypertrophy in dogs. J Thorac Cardiovasc Surg. 2002;124(4):768–774
- . Local and regional wave speed in the aorta: effects of arterial occlusion. Med Eng Phys. 2004;26(1):23–29
- . Stroke volume effect of changing arterial input impedance over selected frequency ranges. Am J Physiol. 1985;248(4 Pt 2):H477–H484
- . Hemodynamics. 2nd ed. Baltimore: Williams and Wilkins; 1988;
- . 1st ed. Dynamics of the vascular system. vol. 1. USA: World Scientific Publishing Co., Singapore; 2004;
- . Pressure dependence of the canine aortic characteristic impedance and the effects of alterations in smooth muscle activity. Med Biol Eng Comput. 1985;23(4):324–338
- . Pressure–flow loops and instantaneous input impedance in the thoracic aorta: another way to assess the effect of aortic bypass graft implantation on myocardial, brain, and subdiaphragmatic perfusion. J Thorac Cardiovasc Surg. 2003;125(3):699–710
- Hemodynamics induced after acute reduction of proximal thoracic aorta compliance. Eur J Vasc Endovasc Surg. 2003;26(2):195–204
- Wall mechanics of the stented rabbit aorta: long-term study and correlation with histological findings. J Endovasc Ther. 2003;10(3):577–584
- . Changes in wall mechanics after endovascular stenting in the rabbit aorta: comparison of three stent designs. AJR American Journal of Roentgenology. 2001;176(3):803–807
- . Non-invasive measurement of mechanical properties of arteries in health and disease. Proceedings of the Institution of Mechanical Engineers Part H. J Eng Med. 1999;213(3):195–202[review; 41 refs]
- . An experimental study of endovascular stenting with special reference to the effects on the aortic vasa vasorum. Cardiovasc Intervent Radiol. 1998;21(1):45–49
- Effect of vasa vasorum flow on structure and function of the aorta in experimental animals. Circulation. 1995;91(10):2669–2678
- . Hemodynamics and wall mechanics of a compliance matching stent: in vitro and in vivo analysis. J Vasc Intervent Radiol. 2002;13(1):97–105
- . Flow instabilities induced by coronary artery stents: assessment with an in vitro pulse duplicator. J Biomech. 1995;28(1):17–26
- Compliance matching stent placement in the carotid artery of the swine promotes optimal blood flow and attenuates restenosis. Eur J Vasc Endovasc Surg. 2004;28(4):431–448
- . Flow patterns in an endovascular stent-graft for abdominal aortic aneurysm repair. J Biomech. 2004;37(1):89–97
- . A survey of stent designs. Minimally invasive therapy and allied technologies: Mitat. vol. 11. 2002;p. 137–147
- Physical properties of endovascular stents: an experimental comparison. J Vasc Intervent Radiol. 2000;11(5):645–654
- . Evaluation of the compressive mechanical properties of endoluminal metal stents. Catheter Cardiovasc Diagn. 1998;44(2):179–187
- . Strength, elasticity, and plasticity of expandable metal stents: in vitro studies with three types of stress. J Vasc Intervent Radiol. 1994;5(5):745–750
- . Elastic characteristics of the self-expanding metallic stents. Invest Radiol. 1988;23(5):370–376
- . Short-term effects of a new intravascular nitinol stent in canine arteries. Invest Radiol. 1999;34(5):367–373
- . MR imaging assessment of cardiac function. J Magn Reson Imaging. 1919;789–799[review; 83 refs]
- Multi-detector row CT of left ventricular function with dedicated analysis software versus MR imaging: initial experience. Radiology. 2004;230(2):403–410
- A novel Doppler echocardiographic method of measuring the biophysical properties of the aorta in pediatric patients. J Am Soc Echocardiogr. 2003;16(7):745–750
- . Estimation of aortic compliance using magnetic resonance pulse wave velocity measurement. Phys Med Biol. 2000;45(6):1703–1713
PII: S1078-5884(06)00093-1
doi:10.1016/j.ejvs.2006.01.020
© 2006 Elsevier Ltd. All rights reserved.
Volume 32, Issue 2 , Pages 129-135, August 2006

