If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The aim of this study is to present novel ex-vivo models in the study of complex haemodynamical changes in Stanford type B aortic dissection (TBAD).
Materials and methods
Fifteen fresh porcine aortas were harvested and preserved with 4 °C saline. Ex-vivo models were developed to simulate TBAD in three different situations: model A with patent false lumen, model B with distal re-entry only and model C with proximal primary entry only. These models were connected to standardised pulsatile pumps and the pressure waveforms were monitored and compared. The aortas were scanned with ultrasonography and subjected to post-experiment autopsy.
Results
The three different models were successfully created (n = 13). Pulsatile flow testing was successful and the shapes of the pressure waveforms were similar to those taken from human aorta. Post-testing gross examination confirmed the success of modelling.
Conclusion
Porcine aortas may prove to be useful ex-vivo models in the study of aortic dissection haemodynamics. These models are reproducible and may be used in the study of complex haemodynamic forces during the development and propagation of TBAD. Our three porcine models give a potential possibility in helping clinicians isolate and analyse complex haemodynamical factors in the development, propagation and prognosis of TBAD.
The precise mechanisms responsible for the aortic dissection remain unclear. In addition to inflammatory protein and genetic analysis, many current theories have focussed on the examination of haemodynamics in aortic dissection. Our three porcine models give a potential possibility in helping clinicians isolate and analyse complex haemodynamical factors in the development, propagation and prognosis of the type B aortic dissections.
Acute aortic dissection is a vascular emergency characterised by a primary intimal tear in the wall of the aorta that causes blood to penetrate the intima and enters the media layer.
The high intra-aortic pressure splits the media in a laminar fashion, and spreads in a corresponding intra-medial plane along the length of the aorta creating a false lumen. Acute aortic dissection may be associated with considerable mortality and morbidity if the diagnosis and treatment are delayed.
The precise pathophysiological mechanisms responsible for the development of aortic dissection remain unclear. Aortas with dissections are often heavily atherosclerotic and therefore vulnerable. The aortic wall is usually a highly dynamic and tightly regulated structure.
Considerable haemodynamic stress must be involved in order to create a tear in the intima on diseased aorta. The majority of current theories concentrated on the examination of haemodynamic forces in aortic dissection, most of which were delivered through idealised computational fluid dynamic (CFD) models or finite-element analysis (FEA) of type B aortic dissections.
However, these studies invariably assumed that aortic wall and the dissecting flaps are rigid and non-elastic components, and, therefore, their conclusions were less convincing and cannot be extrapolated to the human model.
There are several in-vivo and ex-vivo models being developed recently (Table 1), all with their inherent benefits and shortcomings. Several studies successfully established live animal type B aortic models;
however, the dissection flaps created were often unpredictable and inconsistent: either with very short length or with undefined propagation and low reproducibility in subsequent experiments. Besides, most of the existed modelling techniques on live animals required invasive surgical procedures such as cutting and suturing on the outer layers of aortic wall, which would inevitably destroy the inherent aortic wall integrity. Some other researchers applied ex-vivo phantoms to study fluid dynamics on type B aortic dissection,
using polytetrafluoroethylene (PTFE) grafts and polymer-tubing systems to mimic aortic wall and dissecting flaps, which were non-pliable and of different physiological characteristics comparing to the real animal or human tissues. Dziodzio et al.
used isolated swine aorta to study the propagation of dissecting flaps in models with different sites of primary tears; however, the elasticity and biomechanical behaviour of the isolated aorta were manipulated in the process of modelling as the author performed a transverse aortotomy to expose the contralateral intima.
Table 1A brief summary of type B aortic dissection models established in recent studies.
We present here with three novel ex-vivo models in the study of aortic dissection haemodynamics.
Materials and Methods
Materials
Fifteen fresh porcine aortas were harvested at a slaughterhouse under permission from local Food and Environmental Hygiene Department. The aortas were procured from the level of aortic valve to the level of renal arteries, preserved in 4 °C saline, and experiments were performed within 48 h after harvesting.
Successfully made porcine TBAD models were connected to a pulsatile pump (1423A, Harvard Apparatus, MA, USA), and pressure waveforms were taken by a set of data recorder (ML866P, Powerlab, NSW, Australia; ML221, Bridge Amp, NSW, Australia) with aortic wall punctuation, using a 23-GA syringe needle (Terumo, NJ, USA). Dissecting flap motions were monitored with a broadband compact linear array transducer (L15-7io, Philip, WA, USA) on an ultrasound duplex machine (IU22, Philips, WA, USA).
Methods
Fifteen raw porcine aortas harvested from the slaughterhouse were dissected, cleaned and preserved in 4 °C saline. They were divided into three groups randomly (five aortas in each group) to simulate three different pathological scenarios: 1, model A with patent false lumen; 2, model B with distal re-entry only; and 3, model C with proximal primary entry only (Fig. 2). In these three models, model A is clinically similar to the untreated acute TBAD case. Model B can simulate the TBAD case treated with stent graft, the proximal tear of which is sealed whereas leaving an intact distal re-entry. Model C may resemble the initially onset cases attacked by TBAD, which is characterised with a fresh primary tear proximally, a short dissecting flap and absence of a distal re-entry. Model C can also be used to simulate the situation in which thrombus partially formed in distal false lumen, obstructing the outflow within the false lumen.
Figure 1Aortas before modelling. A: raw aorta harvested from slaughterhouse. B & C: prepared aorta with tied and sutured. Black arrow: coeliac trunk.
Figure 2Three different pathological scenarios: A, model A with patent false lumen; B, model B with distal re-entry only; and C, model C with proximal primary entry only. *: false lumen. Black arrow: Proximal primary entry. Red arrow: Distal re-entry.
To create a TBAD model with patent false lumen, the aorta was everted with the help of a clamp. Before starting, orifices of small branches on the intimal surface were observed, and the route of creating the dissection flap was carefully planned, trying to avoid passing above the orifices when modelling. A 20-GA I.V. catheter (Insyte Autoguard, BD, UT, USA) was used to puncture the intima till mid-portion of media, and 6–10 ml bloodstained saline was injected, creating an intramural haematoma. The next step was to channel the dissection flap carefully, by both injecting saline and pressing the bleb, without rupturing dissection flap. After creating a satisfactory morphology of dissection flap to a length of 20–25 cm, scalpel and scissors were applied to cut through the flap creating primary entry and re-entries of different sizes at demanded locations (Fig. 3). Finally, the aorta was everted back to finish the procedure.
Figure 3Procedure of creating model A. A: using a 20-GA I.V. catheter puncture intima on a turned over aorta. B: creating a bleb with blood-stained saline. C: channelling the dissection flap. D: cutting the dissecting flap to create a primary entry in a proximal location with scissors. E: creating a re-entry on the distal portion of dissecting flap with scissors.
After everting the aorta, a 20-GA I.V. catheter (Insyte Autoguard, BD, UT, USA) was used to puncture the intima at the distal portion of aorta to create an intramural haematoma at 3 cm above the coeliac artery. Then, likewise, the dissection flap was channelled proximally, lengthening 20–25 cm, without involving orifices of small branches. Finally, the dissection flap was fashioned distally by scalpel and scissors with a tearing of 12-mm length, leaving the proximal flap intact. After everting the aorta outside-in, the model was preserved back into the cold saline.
Model C
The procedure of creating a TBAD model without re-entry was similar to creating model A. A 20-GA I.V. catheter (Insyte Autoguard, BD, UT, USA) was used to puncture the intima, and an intramural haematoma was created by injecting 6–10 ml bloodstained saline. The dissection flap was channelled carefully by both injecting saline and pressing the bleb. After creating a satisfactory morphology of the dissection flap to a length of 20–25 cm, the only difference was that the distal flaps were left intact, while choosing a demanded proximal site to create the primary entry (Fig. 3). Finally, the aorta was everted back to finish the procedure.
All of the small branches of dissection models, including lumbar arteries and visceral branches, were sutured surgically with 6/0 Prolene® (Ethicon, NJ, USA), and both brachiocephalic arteries were double-ligated with 3/0 Mersilk® (Ethicon, Edinburgh, Scotland) (Fig. 1). Three models were connected to the pulsatile pump. The pump rate was set to 60–70 circles per minute. Stroke volume was fixed to 50 ml per stroke and the ratio of systolic phase over diastolic phase was limited to 35:65. Systolic blood pressure in the true lumen ranged from 90 to 150 mmHg, and was regulated by constraint on outflow with 4/0 Prolene sutures (Ethicon, NJ, USA). Simultaneous pressure waveforms of flow in both false and true lumens were taken in all three models at 10 cm distal to the left brachiocephalic artery (BCA) with a pressure sensor (Transpac IV Monitoring Kit, Hospira, IL, USA) connected to the data recording system. Three models from each group were monitored with an intra-operative ultrasound probe for flow status and velocity. At the end, three specimens of each model were cut open from the outer wall of the false lumen to examine the patency of the false lumen and the dissecting flap.
Results
The lengths (from left brachiocephalic artery to renal artery) and outer diameters of aortas under no blood pressure ranged from 34 to 44 cm and from 21 to 27 mm, respectively. In model A, all five aortas were successfully created, and four out of five aortas were successfully modelled in both model B and C.
Detection of the flow status in the models showed the crescent-shaped false lumen under ultrasonography. In all three models, the false lumen expanded and compressed the true lumen during systolic phases and retracted partially during diastolic phases, which is even more obvious on model A and model C. During the cardiac cycle, the dissection flap moved back and forth between true and false lumens at the change of fluid pressure. Pulsatile flow velocities were acquired in both true and false lumens (Fig. 4). We have found that the dissecting flap stirred over a span of distance within the aorta during cardiac cycles as a result of difference of blood pressure in true and false lumens. The velocity of flow in the true lumen ranged from approximately 50 cm s−1 to 200 cm s−1 (Fig. 4). The diameter of proximal descending aorta cyclically ranged from 2.5 cm to 3.5 cm during cardiac cycles. Therefore, the Reynolds number for this saline system was between 1253 and 7018 under room temperature.
Figure 4Ultrasonography of successful models. A: pulsatile flow in model A. B: pulsatile flow in model B. C: pulsatile flow in model C. A1, B1, C1: transverse sections of three models. A2, B2, C2: longitudinal sections of three models. A3, B3, C3: Doppler velocity measurements in three true lumens. *: false lumen.
In the examination of pressure waveforms, both true and false lumens demonstrated ideal peaks and valleys of pressure waveforms (Fig. 5).
Figure 5Pressure waveforms are taken on three models. A: model A. B: model B. C: model C. Red: waveforms taken from true lumens. Blue: waveforms taken from false lumens.
All the false lumens of models A and B remained 100% stable, that is, no obvious extra extension of flaps was observed (Fig. 6). However, autopsy showed that two out of four models of model C developed both extension of dissecting flaps and newly formed distal propagation.
Figure 6Post-modelling gross examination of the three models. A: model A. B: model B. C: model C. *: false lumen. Black arrow: Proximal primary entry. Red arrow: Distal re-entry.
Prognosis of aortic dissection is largely related to the location of the most proximal initial intimal tear, and the Stanford classification is most applicable to clinical practice.
In Stanford type A aortic dissection, the primary intimal tear may propagate to a different extent to involve the ascending aorta, aortic arch and the descending aorta. In Stanford type B aortic dissection, the ascending aorta and the proximal aortic arch are spared, and the primary dissection flap starts distal to left subclavian artery origin and may propagate distally to involve the descending aorta, abdominal aorta or even the iliac arteries.
The pulsatile movement of the dissecting flap suggests that dissecting flaps play a pivotal role in balancing blood pressure between true lumen and false lumen. In this assumption, blood pressure of true lumen and false lumen should not be significantly different, which is in contradiction with the conclusion of several previous published articles based on rigid polymer tubing system. However, this assumption requires further experiments to validate.
We also found that after connecting the three models in our circulation system, dissection flaps in model A and model B remained stable, whereas half of model C's flaps propagated or ruptured distally, which suggested that a false lumen without distal re-entry bears a relatively higher pressure. This is consistent with the findings of Tsai et al.,
which showed that patients with partial thrombosis in the false lumen have a higher risk of mortality due to a relatively higher mean pressure in the false lumen, compared with the patients with patent false lumen.
Ex-vivo models are useful for studying haemodynamics of aortic pathologies, as individual factors in a complicated circulation system can be isolated and analysed, including tearing size and locations, heart rate, blood pressure, stroke volume, etc. In this study, we creatively apply fresh porcine aorta as ‘modelled’ aorta, replacing the synthetic polymer or silicon tubing used in previous ex-vivo studies.
As a result, the shapes of the detected waveforms were very similar to that taken from real patients, and much better than existing studies.
There are some drawbacks and criticisms in our three models. The branches of porcine aortas were sutured and tied off when tested with the pulsatile flow. Tying-off branches would inevitably alter the reality of flow characteristics, but such constraints were limited by the laboratory settings. Dissecting flaps of the three models were freshly torn, soft and thin compared to the chronic human dissecting aortas. Therefore, the three models were more suitable to simulate acute type B aortic dissections, as flaps in chronic situations are relatively thicker and less soft. Given this, these three models are not suited to the study of chronic TBAD and the progressive effect of aneurismal formation of TBAD.
Another drawback was the choice of non-thrombotic blood analogue in this study. Previous ex-vivo aortic haemodynamic adapted water,
Intrasac pressure waveforms after endovascular aneurysm repair (EVAR) are a reliable marker of type I endoleaks, but not type II or combined types: an experimental study.
In the pulsatile flow model study, the blood analogue can hardly be a thrombotic medium as the spontaneous thrombosis could either block the tubing system or disturb the function of the pulsatile pump. As we all know, it is because of the function of epithelial cells lining in the human blood vessel that the blood components do not form a clot easily. But if we expose those clotting factors into air or have them contacted with the polymers on surfaces of mechanical pump and tubing system, those blood components will thrombose quickly, resulting in the damage or blockage of the ex-vivo circuit.
The three models are of much potential in the study of TBAD. Flow patterns and volumetric changes during cardiac cycle within true lumen and false lumen on these three models can be assessed by exploring time-resolved four-dimensional magnetic resonance imaging (MRI) techniques.
The vibration effect of the aortic wall, dissecting flap motions and drag forces on deployed stent grafts can also be studied on these three models. Further, the propagation of the dissecting flap under different flow and pathological conditions can be observed, helping clinicians better understand the aetiology and progress of TBAD. In addition, these valuable parameters can be further used as boundary conditions to establish more realistic computational fluid dynamic models and/or in the study of finite element analysis on type B aortic dissection. Several extreme anatomical variations of type B aortic dissections, circumferential involvement of medial wall
Intrasac pressure waveforms after endovascular aneurysm repair (EVAR) are a reliable marker of type I endoleaks, but not type II or combined types: an experimental study.
To submit a comment for a journal article, please use the space above and note the following:
We will review submitted comments as soon as possible, striving for within two business days.
This forum is intended for constructive dialogue. Comments that are commercial or promotional in nature, pertain to specific medical cases, are not relevant to the article for which they have been submitted, or are otherwise inappropriate will not be posted.
We require that commenters identify themselves with names and affiliations.
00415-7&id=gr1.jpg){kind=link}
00415-7&id=gr2.jpg){kind=link}
00415-7&id=gr3.jpg){kind=link}
00415-7&id=gr4.jpg){kind=link}
00415-7&id=gr5.jpg){kind=link}
00415-7&id=gr6.jpg){kind=link}