Hybrid approach to thoracoabdominal aortic aneurysms in patients with prior aortic surgery

Hybrid group 

Beginning in 2001, 13 high-risk patients (12 men) with a median age of 69.6 years (range, 35 to 82 years) underwent one-stage TAAA stent graft repair with visceral aortic debranching and prior retrograde revascularization of visceral arteries (Fig 1). The criteria we used to define these patients as being at high-risk for conventional TAAA repair were American Society of Anesthesiologists (ASA) class 3 or 4 associated with a preoperative forced expiratory volume in 1 second (FEV1) <50%. These patients’ data were prospectively collected in a computerized database.

View Large Image Download to PowerPoint

Fig 1. A, Preoperative computed tomography angiography (CTA) scan shows a type II thoracoabdominal aortic aneurysm (TAAA) in a previous unsuccessful attempt of open and endovascular thoracic aneurysm repair. B, Intraoperative photograph shows the creation of an infrarenal neck with grafting of aneurysmatic aorta and two customized Y grafts directed to the renal arteries, the celiac trunk and the superior mesenteric artery. C, Postoperative CTA and drawing D, shows patency of the Y grafts and correct deployment of the thoracoabdominal endograft with complete aneurysm exclusion of the TAAA.

The mean aneurysm size was 7.3 cm (range, 6.4 to 12 cm). One patient underwent emergency surgery because of TAAA rupture. Anatomic inclusion criteria for the hybrid repair procedure were a minimum proximal aortic neck length of 20 mm in the descending aorta, the possibility of an adjunctive visceral open reconstruction to be effective in lengthening the distal aortic neck in the visceral or infrarenal aorta to a minimum of 15 mm, and an aortic neck diameter allowing 15% to 20% oversizing in the absence of circumferential thrombus or calcifications.

In all cases, 16-row multislice computed tomography (CT) with 1-mm intervals and multiplanar reconstructions (Aquilion 16; Toshiba Medical Systems, Tokyo, Japan) was performed for planning and sizing. The extent of the TAAA, the proximal and distal landing zones,16 revascularized visceral arteries, and stent graft used are summarized in Table I.

Table I. Thoracoabdominal aortic aneurysm extent, proximal and distal stent graft landing zones, revascularized visceral arteries and deployed endografts in the 11 patients underwent hybrid repair

TAAA, Thoracoabdominal aortic aneurysm; CT, celiac trunk; SMA, superior mesenteric artery; RRA, right renal artery; LRA, left renal artery, VAP, visceral aortic patch. The proximal landing zones according to the Ishimaru classification16 are shown in the drawing below the table. Zone 0: overstenting of all the supra-aortic trunks; zone 1:over-stenting of left sublavian and left common carotid arteries; zone 2: overstenting of the left subclavian artery; zone 3 and 4 distal to the left subclavian artery. Regarding the distal landing zones classification we considered, as shown in the drawing below the table, as zone 0, the overstenting of all visceral aortic branches; as zone 1, the overstenting of the celiac trunk, of the superior mesenteric artery, and of the higher renal artery; as zone 2, the overstenting of the celiac trunk and of the superior mesenteric artery; as zone 3, the overstenting of the sole celiac trunk; and as zone 4 the stent graft repair proximal to the celiac trunk.

All of the patients underwent a one-stage procedure performed in the operating room. Preoperative cerebral spinal fluid (CSF) drainage was performed in five selected patients in whom a long segment of native descending aorta was planned to be covered by the stent graft (1 type II TAAA with prior abdominal aortic repair, 1 type II TAAA with prior ascending aortic repair, and 3 type I TAAA with prior abdominal aortic repair). CSF drainage was not performed preoperatively in the other patients who had a long segment of already grafted thoracic aorta (4 type I TAAA with previous descending aortic repair and 1 aneurysm of visceral aortic patch in a prior type II TAAA repair) or had a type IV TAAA (2 with prior abdominal aortic repair). CSF drainage was instituted postoperatively in one patient with an aneurysm of the visceral aortic patch (prior type III TAAA repair) when delayed symptoms of spinal cord ischemia developed.

A portable digital C-arm image intensifier with road-mapping capabilities (series 9600, OEC Medical System or Moonray, Simad Medical Technology, Modena, Italy) was used. All the patients were operated on under general and epidural anesthesia. Cell salvage, rapid infusers, and transfusion were used. The abdominal aorta, common iliac arteries, and the first 2 cm from the origin of the common hepatic artery, superior mesenteric artery, and renal arteries were exposed, when required, through a transperitoneal mid-line approach with the patient in a supine position.

The choice of inflow site for retrograde visceral artery bypass grafting was based on the extent of the TAAA, the presence of prior abdominal aortic repair, and the quality of the walls of the native aorta and iliac arteries. In seven of eight cases with pre-existing abdominal aortic grafts for prior infrarenal aortic repair (n = 5) or TAAA repair (n = 2) the retrograde grafts were anastomosed to the previous aortic graft to avoid clamping and an anastomosis on atherosclerotic arteries. Only in one case was a pre-existing abdominal aortic graft for a prior infrarenal aortic repair not used as the inflow site owing to the good quality and favorable anatomic features of the right common iliac artery that was used as the donor vessel (Fig 2).

View Large Image Download to PowerPoint

Fig 2. A,Preoperative computed tomography angiography (CTA) shows a type I thoracoabdominal aortic aneurysm (TAAA) in a previous AAA repair. B, Intraoperative photograph shows the detail of the customized Y graft directed to the celiac trunk and superior mesenteric artery from the right common iliac artery. Postoperative CTA shows patency of the Y graft and correct deployment just above the renal arteries with aneurysm exclusion of the TAAA of the thoracoabdominal endograft.

Among the five patients who had not undergone prior abdominal aortic grafting, the native abdominal aortic wall was of good quality in two cases, enabling us to use it as the donor vessel. In another two patients, we preferred to use the right common iliac artery as the donor vessel because of the better quality of the wall of this artery and its favorable anatomy. A one-stage procedure was done in another patient in which an AAA was first repaired with a tube graft, and then visceral bypass grafts were anastomosed to the distal part of the infrarenal graft, allowing an adequate in-graft distal endograft landing zone. All the anastomoses to the donor vessels were created in an end-to-side fashion.

Overall, 32 retrograde visceral bypasses were performed (13 to the celiac trunk, 13 to the superior mesenteric artery, and 6 to renal arteries). Synthetic grafts (Dacron or expanded polytetrafluoroethylene) with diameters of 6 mm and 8 mm were used in all patients. We preferably used customized Y grafts or separated bypass grafts, or both, for each recipient vessel. We used a single bypass with a sequential graft technique in one patient, and an inverted bifurcated graft in another. For celiac trunk revascularization, the graft was routed in front of the renal vein behind the pancreas, an arteriotomy was made in the common hepatic artery, and an end-to-side anastomosis was made. An end-to-end anastomosis was usually preferred for the superior mesenteric artery, and this type of anastomosis was always used for the renal arteries.

Before and during renal artery cross-clamping, diuresis was forced by systemic infusions of furosemide and mannitol (0.5 g/kg). During the cross-clamping, a bolus of 300 mL of cold (4°C) crystalloid lactated Ringer solution was perfused into the orifice of the renal artery with a 9F Pruitt-Inahara (LeMaitre Vascular, St. Petersburg, Fla) occlusion perfusion balloon catheter. A continuous perfusion of approximately 500 mL was then administered to reduce the temperature of the kidney to 15°C to 18°C.

In all reconstructions, the grafted vessels were ligated proximally to prevent retrograde perfusion of the sac after endovascular exclusion of the aneurysm. The grafts were then covered with retroperitoneum or omental flap whenever possible.

The access vessel for endograft insertion was the common femoral artery (exposed through an inguinal incision) in four patients, an iliac approach was used in three patients, and the device was inserted through the infrarenal aorta in 6 patients. No conduit was used for inserting the endograft, and the access sites were repaired by direct suture or by a synthetic patch chosen according to the size and the quality of the artery.

A variety of stent grafts were used, including Zenith (Cook, Bloomington, Ind), Excluder (W.L. Gore & Assoc, Flagstaff, Ariz), and Talent (Medtronic, Minneapolis, Minn), and a median of 1.5 stent grafts (range, 1 to 3) were deployed in each patient.

One patient undergoing long-term dialysis had intentional over-stenting of both the renal arteries, and two patients with a right solitary functioning kidney had intentional over-stenting of the left renal arteries (higher) without revascularization. A completion aortograph was done after deployment of the endografts. Stent graft ballooning was performed selectively.

Patients were evaluated with postprocedure contrast CT scans at scheduled follow-up imaging at 1, 6, and 12 months, and yearly thereafter. Clinical follow-up was also done at regular intervals of 6 months.

Conventionally treated group 

From the entire series of patients who had undergone TAAA repair since 1988, we selected 29 patients (25 men), with a median age of 65.3 years (range, 58 to 79 years), who had had an open repair of TAAA (10 type I, 5 type II, 4 type III, 7 type IV, and 3 aneurysms of the visceral aortic patch) after prior aortic surgery (22% ascending, 38% descending, 42% abdominal aortic repair, and 10.3% redo TAAA), at ASA 3 or 4, and preoperative FEV1 <50%.

These patients’ data had been prospectively collected in a computerized database and are reported in Table II. The median aneurysm size was 8.1 cm (range, 5.9 to 9.3 cm). All the procedures were performed electively.

Table II. Preoperative characteristics of the 29 patients with prior aortic surgery who underwent conventional repair of thoracoabdominal aortic aneurysm

	Patients	ASA	Prior surgery sites	TAAA Crawford extent	Visceral arteries management
	1	3	TAAA	VAPaneurysm	3vesselVAP
	2	3	TAAA	VAPaneurysm	Vesselselectivereimplant
	3	3	TAAA	VAPaneurysm	4vesselVAP
	4	4	Ascending aorta, AAA	TypeI	2vesselVAP
	5	3	Ascending aorta, TAA	TypeIV	4vesselVAP
	6	4	Ascending aorta, AAA	TypeI	4vesselVAP
	7	3	TAA	TypeIV	4vesselVAP
	8	3	TAA	TypeII	4vesselVAP
	9	4	TAA	TypeI	2vesselVAP
	10	3	TAA	TypeIII	3vesselVAP
	11	3	TAA	TypeIV	4vesselVAP
	12	3	TAA	TypeIII	4vesselVAP
	13	4	TAA	TypeIV	3vesselVAP
	14	4	TAA	TypeIII	4vesselVAP
	15	3	TAA	TypeII	Vesselselectivereimplant
	16	3	TAA	TypeIII	3vesselVAP
	17	3	Ascending aorta	TypeIV	4vesselVAP
	18	3	Ascending aorta	TypeI	2vesselVAP
	19	4	Ascending aorta	TypeI	2vesselVAP
	20	3	AAA	TypeI	4vesselVAP
	21	4	AAA	TypeII	4vesselVAP
	22	4	AAA	TypeIV	Vesselselectivereimplant
	23	3	AAA	TypeI	2vesselVAP
	24	3	AAA	TypeI	2vesselVAP
	25	3	AAA	TypeIV	4vesselVAP
	26	4	AAA	TypeII	3vesselVAP
	27	4	AAA	TypeII	Vesselselectivereimplant
	28	3	AAA	TypeI	4vesselVAP
	29	4	AAA	TypeI	2vesselVAP

TAAA, Thoracoabdominal aortic aneurysm; AAA, abdominal aortic aneurysm; VAP, visceral aortic patch.

For type I to III TAAA, distal aortic perfusion with left atriofemoral bypass and CSF drainage were used routinely. For type IV TAAA, hypothermic visceral perfusion was used routinely and CSF drainage was not performed.

Patients were evaluated with postprocedure contrast CT scans or magnetic resonance imaging at scheduled follow-up imaging at 3 and 12 months and yearly thereafter. Clinical follow-up was also conducted at regular intervals of 6 months.

We analyzed the outcomes in our patients, reporting the results and methods in accordance with the current reporting standards for endovascular aortic repair prepared and revised by the Ad Hoc Committee for Standardized Reporting Practices in Vascular Surgery (Society for Vascular Surgery/American Association for Vascular Surgery).18

Hybrid group 

Hybrid repair of TAAA was technically successful in all 13 patients. No intraoperative deaths occurred. No additional intraoperative cuff placement was required. The median procedure time was 239 minutes (range, 187 to 465 minutes). A median of 4.7 U packed red blood cells was transfused for a median blood loss of 480 mL (range, 380 to 2900 mL). All patients were admitted to the intensive care unit where they stayed for a median of 3 days (range, 1 to 12 days). The median hospital length of stay was 8 days (range, 7 to 35 days).

The hybrid repair was clinically successful in 10 (76.9%) of 13 patients. Three patients died postoperatively for a mortality rate of 23%: one on day 35 related to respiratory failure; one on day 3 from postoperative coagulopathy, and one on day 25 from pancreatitis.

Delayed onset paraplegia at 2 days postoperatively was observed in one (7.7%) patient with visceral aortic patch aneurysm, which resolved after CSF drainage. Other postoperative complications included acute renal failure in two patients (15.4%) that resolved without dialysis, and respiratory failure in one (7.7%).

At a median follow-up of 14.9 months (range, 11 days to 57.4 months), the mid-term clinical success rate was 76.9%. One patient died after 2 years of follow-up from causes unrelated to the aneurysm. We did not observe any aneurysm-related deaths, endoleak, stent graft migration, or visceral revascularization–related morbidity.

Conventionally treated group 

No intraoperative deaths occurred. The mean procedure time was 169 minutes. A mean of 8.7 U of packed red blood cells was transfused for a median blood loss of 1350 mL (range, 850 to 6500 mL). Perioperative mortality was 17.2% and morbidity was 44.8%. Perioperative complications included 3 patients with permanent paraplegia, 1 with coagulopathy, 2 with renal failure, and 7 with respiratory failure. All patients were admitted to the intensive care unit, where they stayed for a median of 2 days (range, 1 to 23 days).

At a median follow-up of 5.4 years (range, 1.7 to 7.9 years) no significant complications related to the aortic repair had been reported except for three (10.3%) patients with asymptomatic dilatation of visceral aortic patch <5 cm undergoing radiologic surveillance.

Preoperative comorbidities, risk factors, TAAA extent, age, and distribution of sites of prior aortic surgery were not statistically significantly different between the hybrid group and the conventionally treated group (χ2 and Student’s t test). Mortality and morbidity rates, intraoperative blood loss, operative time, median intensive care unit stay, and median duration of hospital stay were not notably different between the hybrid group and the conventionally treated group.