Discussion
Even if the surgery of an acute Debakey type I aortic dissection is performed with an open distal anastomosis and is mostlikely managed by a hemiarch aortic reconstruction, this approach has a main disadvantage, that is to leave entry tears in the proximal descending thoracic aorta leading to false lumen patency in some patients, increasing the need for re-operation7-10. Total arch replacement with Frozen Elephant Trunk allows for distal extension of a stent graft implant into the true lumen of the descending aorta excluding re-entry tears in the arch and proximal descending thoracic aorta. In the setting of acute aortic dissection and deeply ill patients, this approach can increase the CPB and circulatory arrest times, and exposes patients to additional risks of paralysis, stroke and haemorrhage11,12. It may be more useful in particular situations in the setting of aortic dissection like when there is arch aneurysm, or there is an entry tear within the arch or in the proximal descending thoracic aorta associated with the dissection13.
Between these two approaches, the AMDS represents a novel hybrid solution providing long thoracic coverage alleviating malperfusion and excluding entry tears without significantly increasing the complexity of surgery4-6.
DARTS trial has shown a good rate of aortic remodeling. However, this was a composite criterion; positive remodeling was defined on evidence of false lumen obliteration, complete false lumen thrombosis and favourable changes in aortic dimensions14,15. Our real-life experience shows that these results are more the consequence of lumen’s diameters correction than a complete false lumen thrombosis, being possibly due to the primary entry tear exclusion. There was no need for a redo surgery in our four patients, but we think that the presence of Nitinol in the arch could jeopardize a second procedure at this level.
In the two patients having intestinal malperfusion we noticed clinical improvement. One of them remains having intestinal angina due to dissection of the superior mesenteric artery. Favourable changes in aortic true lumen and false lumen dimensions were found in most of our patients but the AMDS was compressed at the isthmus in one of them. We did not have any mortality in our patients. This shows that AMDS is a reliable and secure device. However, its benefits remain unclear when it comes to a positive remodeling and seems less likelihood comparable to a frozen elephant trunk. The main reason seems to be an insufficient radial force of the AMDS, which tend to lengthen rather to expand.
Many studies addressing the radial force of endovascular stents emphasized on the importance of understanding radial force when selecting a stent for every patient. Radial force of endovascular stents provides effective support for blood vessels, maintains adequate lumen patency, and secures fixation to artery wall16. It varies among stent designs, and differences depend on the type of stents, the site of deployment or layer characteristics of each stent17,18. In vivo, endovascular stents would be affected by the vessel curvature, blood pressure, vascular smooth muscles characteristics and much more dynamic factors17. Surgeons should evaluate the possibility of stent deformities during and after surgery19. It is believed that treatment of dissection with endovascular stent requires fewer radial force compared with the treatment of aneurysm because too much radial force at distal ends may lead to new entry tears. However, the arch geometry for thoracic aorta requires larger radial forces to seal20. In its initial experience on AMDS stent, Montagner et al. concluded that the low radial force of AMDS stent is intended just to readapt the intima against the media and adventitia and it is the subsequent expansion of the true lumen that will drive the resolution of malperfusion21. Furthermore, they emphasized on the importance of low radial force of the AMDS stent which can unlikely damage the intima. They also reported three failures of device deployment, one of them being due to high turtosity of the aorta causing kinking and incomplete AMDS expansion22.
Eventhough we observed improvement of malperfusion, this may be attributed only to primary entry tear resection. Actually, the radial force of the AMDS is not adequate to guarantee distal expansion of the true lumen, especially in case of visceral malperfusion and this was seen in one of our patients. AMDS implantation should be avoided in patients with aortic calcifications or kinking to prevent incomplete stent expansion.
Another feature should be mentioned is that the AMDS is an uncovered stent. In this condition, any entry tear not excluded by this bare stent will remain active. This is why, we should absolutely avoid AMDS implantation in patients with any primary entry tear in the
aortic arch.
The main limitations of this study remain in its small sample size, its retrospective design and the absence of long-term follow-up.