Introduction
Endovascular fenestration is used to manage complicated aortic dissection in a manner analogous to open surgical septectomy. The intent to create a large communication between the aortic true and false lumina providing free interluminal flow and equalization of pressure differentials. The technique depends on the clinical situation. The two most commonly reported include endovascular balloon septectomy and longitudinal guidewire-mediated septectomy.
Balloon fenestration of the dissection septum is most frequently performed in acute dissection complicated by branch vessel ischemia. The purpose is to create a hole or tear within the septum separating the true and false lumina. This produces a more balanced admixture of blood flow in the aortic lumina and functionally reduces or relieves malperfusion of tissue beds supplied by branch arteries originating from a previously compromised aortic lumen.
Guidewire-mediated longitudinal septal fenestration is typically used in cases of chronic dissection associated with false lumen aneurysm formation within the thoracic and/or abdominal aorta. In the setting of chronic dissection, branch vessel malperfusion is rarely critical, and the role of fenestration therefore has another focus. In selected patients at high risk of open aneurysm repair, endograft placement may offer an alternative therapy. However, the presence of the dissection prevents the graft from obtaining an adequate seal and graft expansion. Consequently, a Type I endoleak commonly occurs.
The intent of a longitudinal fenestration is to divide the septum and to create a uniluminal segment including an appropriate transaortic diameter neck suitable for endograft anchoring around the entire outer circumference of the aorta across both lumina. Now, with the endograft channeling all the blood flow, the aneurysm sac is isolated, thus mitigating the risk of progressive aneurysm enlargement.
Fenestration and Acute Aortic Dissection
Branch vessel malperfusion, caused by acute aortic dissection, is a potentially life-threatening condition if untreated. Mechanisms of branch vessel obstruction secondary to aortic dissection are classified as static, dynamic, or a combination of both. Static involvement is caused by direct extension of the dissection process and the aortic septum into a branch artery resulting in mechanical obstruction. If associated with evidence of critical ischemia, static branch vessel compromise is typically treated by placement of stents within the true lumen of the affected branch.
Dynamic obstruction is caused by prolapse of the dissection septum over branch vessel origins. In cases of acute dissection complicated by dynamic branch vessel malperfusion, compromised flow to critical abdominal and/or leg arteries is frequently a result of false lumen compression of the aortic true lumen (true lumen collapse) that supplies these ischemic branches. When aortic true lumen obliteration exists, there is typically an imbalance in flow and pressure within the two aortic lumina. Often there is a very large primary intimal entry tear, and the resultant false lumen in-flow is frequently greater than its outflow capacity. Consequently, the aortic false lumen diastolic blood pressure is commonly greater than that in the aortic true lumen. In this clinical situation, placement of an aortic stent graft in the true lumen over the primary entry tear results in a dramatic decrease in false lumen in-flow and an almost instantaneous relief of aortic true lumen compromise and the ischemia involving branch arteries originating from it. This is the standard current therapy for so-called “dynamic branch vessel involvement.” However, in cases with unfavorable or unsuitable anatomy for endograft placement or when the appropriate device is unavailable, endovascular septal fenestration is an option. The procedure is performed within the infrarenal aorta where a balloon is inflated across the septum to create a hole or tear. In the setting of acute dissection, this tear is predictably a linear transverse rent in the lamella oriented perpendicular to the longitudinal course of the aorta. The immediate result is a decrease in the resistance to false lumen outflow, increasing flow from the false lumen. This effectively decompresses the aortic false lumen and relieves the upstream compromise affecting branch vessels originating from the true lumen.
The technique is dependent on successfully transgressing the septum—usually from the diminutive, narrowed aortic true lumen to the much larger, dominant false lumen—using a 21-gauge needle directed through a curved, relatively rigid cannula or guide catheter oriented perpendicular to the lamella. Correlation between live fluoroscopy and preprocedural computed tomography (CT) imaging helps to orient the guiding cannula so that it is pointed in the direction of the mid septum. Real-time intravascular ultrasound imaging is invaluable in confirming direction of the needle puncture. Once the needle is advanced 4 mm to 6 mm to cross the septum, a 0.014′′ or 0.018′′ guidewire is advanced until it is well proximal to the level of septal transgression. A 4-French or 5-French catheter is advanced over the needle and guidewire. After their removal, an injection of contrast media is used to confirm successful crossing. A 0.035′′ guidewire is then introduced and the catheter and guiding cannula are removed.
A series of sequentially larger diameter angioplasty balloons are then inflated across the septum. After reaching a diameter of 20–24 mm, aortography is performed to confirm the adequacy of the new communication and the flow to previously compromised branch vessels originating from the aortic true lumen. In most cases, balloon septectomy to the range of diameters previously detailed is sufficient to relieve critical compromise of aortic true lumen branches. This may be difficult to determine definitively intraprocedurally and may require close monitoring of symptoms, blood chemistry studies, and peripheral hemodynamics as appropriate in the immediate postprocedural period before a final determination of whether the fenestration was successful in reversing branch ischemia.
The salutary effect in terms of true lumen enlargement on CT imaging following balloon fenestration is never as dramatic anatomically as that observed following aortic stent graft placement of the primary entry tear, but physiologically, an improvement in true lumen flow is usually evident when the aortic fenestration is performed distal to the involved branches as opposed to more proximally in the distal thoracic segment. A balloon fenestration proximal to the involved abdominal branches should be avoided because it acts as a secondary entry tear and may exaggerate the imbalance in flow between the lumina and do little to alleviate aortic true lumen obliteration in the abdomen.
If the clinical outcome is deemed inadequate, supplemental revision may be undertaken. This may involve placement of a large self-expanding noncovered metallic stent across the fenestration, such as a baffle, guiding flow proximal in the aortic false lumen to distally in the true lumen via the existing fenestration. A large 20–24 mm diameter Wallstent (Boston Scientific Inc., Watertown, Massachusetts) that is between 40 mm and 80 mm long is frequently selected to serve as a frame to buttress open the fenestration and improve the desired hemodynamic effect.
Alternatively, the balloon fenestration may be extended or a new communication created longitudinally using a guidewire-mediated septectomy. This procedure is more commonly used in chronic dissections to create a segmental uniluminal channel. This technique is described later. Both methods are usually successful in reversing symptoms in the majority of acute dissection patients unresponsive to balloon septectomy alone.
In a study by Park et al., 16 patients received aortic fenestration and/or branch stenting as treatment; 30-day mortality was 8.3% and branch vessel malperfusion improved in 90%, although there was also expansion of the outer aortic diameter at 1-year follow-up.
In a study by Midulla et al., 35 patients with aortic dissections were treated with endovascular fenestration. All were technically successful, with 26 patients requiring additional stenting; 30-day mortality was 34%, 8 of 23 survivors suffered from ischemic complications, and all patients had a patent false lumen.
Complications
Serious adverse events related to balloon fenestration are relatively uncommon and tend to be associated with the steps performed prior to the actual balloon inflation. Inadvertent injury to access vessels or the distal aorta during insertion and advancement of the guiding cannula or sheath, which needs to be somewhat rigid to allow accurate directional orientation for needle passage across the flap, is a risk. Similarly, attempts at a needle transgression of the septum from one aortic lumen to the other may be inaccurate with puncture through the outer wall of the aorta into surrounding tissues, the inferior vena cava, or ureter. Depending on the needle size and subsequent passage of any catheters or larger devices, lack of recognition of the unanticipated transaortic tract may lead to bleeding, vessel rupture, arteriovenous fistula, or ureteral laceration. Any of these complications may prove catastrophic and require keen awareness that an unexpected problem exists and prompt attention to treating it appropriately. One way to mitigate a serious adverse event is to use a small-gauge needle, such as a 21-gauge Chiba needle for the septal puncture rather than a larger 18- or 19-gauge needle. Alternately, TourGuide deflectable sheaths may be used in combination with re-entry catheters/devices with intravascular ultrasound to cross the septum more accurately and to avoid puncture extrinsic to the aorta. Although the fenestration should be placed infrarenally, going too low frequently leads to the lack of a successful procedure for restoration of perfusion to the viscera.
Another risk of this procedure exists when thrombus is present at the level of fenestration. This more frequently involves the presence of clot within the false lumen, but it may also exist in the true lumen. The risk of distal embolization may occur from endovascular device manipulation within any partially thrombosed vessel. This is still quite rare. This procedure is relatively safe with concerns usually more related to whether the septectomy has provided clinically successful and complete reversal of branch vessel obstruction and tissue malperfusion.
If the operation is inadequate and extension of the fenestration or its buttressing with a stent as described earlier is undertaken, these techniques become more complicated with additional steps that may be associated with more serious adverse effects. The complications associated with longitudinal guidewire-mediated fenestration will be described later in the chapter, but creation of a baffle using a large self-expanding stent to buttress the septectomy may have unique related risks.
The main risk is the possibility of branch vessel narrowing or obstruction resulting from the presence of the aortic stent overlying its origin. Although a bare, noncovered stent is used, it may have to be placed with a portion of the device within the true lumen of the visceral segment of the abdominal aorta. Depending on its precise position within the aortic true channel, it may abut or cross true lumen branches that may be impaired at their origin. If this is observed, placement of a branch vessel ostial stent delivered through and expanded across the overlying struts of the self-expanding aortic stent is an effective treatment.
In all cases, vigilant postfenestration imaging surveillance at regular intervals is necessary to check for disease progression. In general, the main risk in chronic dissection is false lumen aneurysm formation, but following fenestration there is ongoing risk of branch vessel compromise as its morphology evolves. One example of late recurrent branch vessel ischemia following balloon fenestration was reported by Williams et al. Two months after percutaneous transfemoral fenestration in a Type III aortic dissection, subacute mesenteric ischemia was diagnosed. Continued expansion of the false lumen required aortic resection and graft replacement of the distal thoracic and abdominal aorta.
Fenestration and Chronic Aortic Dissection
False lumen thrombosis is traditionally regarded as a measure of success in thoracic endovascular aortic repair (TEVAR) for acute Type B aortic dissection. However, residual distal septal communications and patent intercostal and lumbar arteries may still provide perfusion and tension to the false lumen, leading to progressive expansion. The effect of this phenomenon on subsequent false lumen aneurysm formation is difficult to predict, but in many cases, complete healing of the aorta is unlikely to be achieved.
A patent false lumen is considered a predictor for late death caused by false lumen rupture and for retreatment of the descending aorta to manage progressive false lumen enlargement. In addition, patients with acute Type B aortic dissection who have a partially thrombosed aortic false lumen exhibit a significantly higher annual growth rate when compared with patients having a patent or completely thrombosed false lumen. Consequently, patients with partial thrombosis require more intense follow-up imaging surveillance and may benefit from prophylactic intervention to avoid false lumen aneurysm formation and rupture.
Treating chronic Type B aortic dissection with TEVAR has several unique concerns. The proximal stent graft landing zone is often compromised because of proximity to arch vessels and lack of a healthy distal neck. In most post-TEVAR cases, the false lumen in the abdominal aorta is not completely excluded, with patency maintained by residual fenestrations distal to the stent graft that perfuse the thoracic false lumen in a retrograde fashion. In this situation, the lower thoracic aortic false lumen remains pressurized and, as a result, is at risk of progressive expansion.
In chronic Type B dissection with false lumen aneurysm, the presence of a septum dividing the aorta into two lumina within proximal and/or distal potential neck segments, in anatomy otherwise suitable for endograft aneurysm exclusion, represents a challenge for endovascular management. In this setting, guidewire-mediated septectomy with longitudinal cutting of the dissection septum may be a useful adjunct to endograft placement by providing a suitable landing zone to anchor and to seal the device.
The rationale for endovascular longitudinal septectomy is to create a uniluminal neck segment either in the distal thoracic aorta to facilitate isolation of a thoracic false lumen aneurysm (proximal device placement in a usual landing zone) or in the infrarenal aorta to create a suitable proximal landing zone to anchor a standard EVAR device to treat an abdominal aneurysm. The restoration of a single channel proximal and/or distal to an aneurysm provides a neck segment that allows circumferential sealing of an endograft against the outer wall of the aorta without an intervening dissection lamella.
The existence of two flow channels in dissection is often associated with pressure and flow discrepancies that cause progressive false lumen dilation and aneurysm formation. Longitudinal septal cutting divides the aortic lamella and has been shown to equalize the flow and pressure in the true and false lumina while it creates a uniluminal channel more amenable to (T)EVAR.
The procedure starts by obtaining bilateral common femoral access, although occasionally it is possible to perform through unilateral groin access. Catheters are introduced into both lumina, guided by CT imaging, arteriography, and/or intravascular ultrasound to identify the orientation of the two channels in the iliofemoral conduit arteries, to detail the termination of the dissection process, and to pinpoint existing natural fenestrations in the septum. Once access is obtained, the septum is punctured at the desired level with a needle directed by a guiding cannula in a manner identical to that described previously. In most cases the puncture is made with the needle advanced from the smaller aortic true lumen across the septum toward a larger false lumen. Once successful passage is confirmed, a guidewire is introduced into the false lumen and snared. The snare and guidewire are then withdrawn and exteriorized such that the wire forms a loop from the aortic true lumen, across the lamella and out of the false lumen. The exact location of the needle puncture to initiate balloon or guidewire septectomy is variable and based on the specific anatomic configuration and orientation of the septum and the related clinical symptoms found in each individual case of dissection.
With both ends of the guidewire outside the patient, the access sheaths in the groins are positioned close (3–5 cm) below the wire segment transgressing the septum to provide support for the fenestration maneuver. Two techniques for initiating the next step of septal division have been described. One involves applying simple downward traction on the ends of the guidewire to create a cleavage tear in the lamella. Once the “break-in” rent is made, the guidewire is pulled distally like a cheese wire to extend the rip in the septum.
Alternatively, the guidewire-mediated fenestration may be initiated by pulling the guidewire back and forth in a sawing motion, like a Gigli wire saw. Once the initial cut is established the sawing motion is continued with a downward force to continue the septal division inferiorly to the desired distal location. Irrespective of the technique used, the starting point of the fenestration must be well proximal (4–7 cm) to the intended landing zone of the proximal margin of an EVAR device. This also applies to creating a suitable distal landing zone for TEVAR isolation of thoracic false lumen aneurysm where a generously long longitudinal fenestration is necessary both proximal and distal to the intended distal aspect of the endograft to facilitate its full expansion and to ensure a circumferential seal to prevent endoleak.
After the septum is longitudinally divided like a curtain, the procedure is completed with standard (T)EVAR graft placement with anchoring of the device within uniluminal neck segment(s). There is no consensus regarding the relative superiority of the techniques in the setting of chronic aortic dissection. There has never been a direct comparison reported. Our experience suggests that less downward traction force is required when a back-and-forth sawing motion is applied during guidewire-mediated longitudinal fenestration.
Several reports in the medical literature have described the technique of guidewire-mediated longitudinal septal fenestration. Those who report the procedure performed in conjunction with standard endograft placement to manage a false lumen aneurysm associated with aortic dissection stress the importance of preprocedural planning to ensure that the proposed landing zone created after septotomy is dimensionally suitable with a transaortic diameter within the treatment range of available devices. In cases where the resultant diameter of the uniluminal neck after fenestration is too large, other options are preferred.
Complications
Arterial tortuosity and variations in septal tissue composition (fibrosis, calcification), dimensions (thickness), and adjacent luminal properties (mural thrombus) may reduce the effectiveness of guidewire-mediated fenestration and influence the risks of the procedure. Unstable arterial ulcerations, dense plaque, and luminal thrombus may result in an increased risk of peripheral embolization during the procedure. Indeed, longitudinal guidewire septotomy irrespective of these variables is associated with potential complications.
There is a risk of aortic wall damage or rupture during the procedure distinct from an incidence of endoleaks (Types Ia and Ib) caused by incomplete sealing following endograft placement with persistent aneurysm perfusion, growth, and rupture. Intraprocedural aortic rupture has not been reported in the literature, but it remains a key concern.
Clearly, the technique of guidewire-mediated septotomy is not precisely controllable in terms of constraining the longitudinal course of lamella tearing, which may result in obstruction of branch arteries. Prolapsing septal segments following guidewire fenestration may cause malperfusion by covering branch vessel origins, resulting in ischemia involving the viscera, kidneys, spinal cord, or lower extremities. Similarly, it is possible that sheared septal tissue may actually prolapse into branch vessel ostia and compromise flow or that a septal flap may narrow an aortic lumen (true or false) proximal to a branch that it supplies.
Fortunately, with prompt recognition, these untoward events may be amenable to endovascular rescue using stents or stent grafts placed within the aorta and/or affected branch. Malperfusion of arteries branching from the aorta may also result from a shift in false lumen thrombus during either guidewire fenestration or as a consequence of stent graft placement. We have experience with one case of renal artery obstruction ( Fig. 27.1 ) where this may be the responsible mechanism. To decrease the risk of inadvertent ischemia, consideration should be given to obtaining wire access to the lowest renal prior to septotomy, to allow for parallel graft rescue should prolapse of the septum occur.
