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
Department of Vascular Surgery, Drum Tower Hospital Affiliated to Nanjing University School of Medicine, Nanjing University, Nanjing, ChinaDepartment of Thoracic and Cardiovascular Surgery, Nanjing First Hospital, Nanjing Medical University, Nanjing, China
Address for reprints: Min Zhou, MD, PhD, Department of Vascular Surgery, Drum Tower Hospital Affiliated to Nanjing University School of Medicine, #321 Zhong Shan Rd, Nanjing 210008, China.
Although physician-modified fenestrated and branched endografts (PMEGs) were proposed as an alternative to thoracoabdominal aortic aneurysms (TAAAs) repair in 2012, PMEG use is still limited by the lack of long-term data in large series. We seek to compare the midterm outcomes of PMEGs in patients with postdissection (PD) and degenerative (DG) TAAAs.
Methods
Data were analyzed for 126 patients (age 68 ± 13 years; 101 men [80.2%]) with TAAAs treated by PMEGs from 2017 to 2020, including 72 PD-TAAAs and 54 DG-TAAAs. Early and late outcomes were compared between patients with PD-TAAAs and DG-TAAAs, including survival, branch instability, and freedom from endoleak and reintervention.
Results
Hypertension and coronary artery disease were present in 109 (86.5%) and 12 (9.5%) patients. PD-TAAA patients were younger (63 ± 10 vs 75 ± 12 years; P < .001), and more likely to have diabetes (26.4 vs 11.1; P = .03), history of previous aortic repair (76.4% vs 22.2%; P < .001), and smaller aneurysm size (52 vs 65 mm; P < .001). TAAAs were extent I in 16 (12.7%), II in 63 (50%), III in 14 (11.1%), and IV in 33 (26.2%). Procedural success was 98.6% (71 out of 72) and 96.3% (52 out of 54) for PD-TAAAs and DG-TAAAs (P = .4). The DG-TAAAs group sustained more nonaortic complications than PD-TAAAs (23.7% vs 12.5%; P = .03) in adjusted analysis. Operative mortality was 3.2% (4 out of 126), which didn't differ between the groups (1.4% vs 5.6%; P = .19). The mean follow-up was 3.01 ± 0.96 years. There were 2 (1.6%) late deaths (from retrograde type A dissection and gastrointestinal bleeding [n = 1 each]), 16 (13.1%) endoleaks, and 12 (9.8%) instances of branch vessel instability. Reintervention was performed in 15 (12.3%) patients. At 3 years, survival, freedom from any branch instability, and freedom from endoleak and reintervention were 97.2%, 97.3%, 86.9%, and 85.8% in the PD-TAAAs group, respectively, which did not differ significantly from DG-TAAAs patients (92.6%, 97.4%, 90.2%, and 92.3% all P values > .05).
Conclusions
Despite the difference in age, diabetes, prior history of aortic repair, and aneurysm size preoperatively, PMEGs achieved similar early and midterm outcomes in PD-TAAAs and DG-TAAAs. Patients with DG-TAAAs were more prone to early nonaortic complications, which represents an aspect for improvement to optimize outcomes and warrants further study.
PMEG achieved similar midterm outcomes in postdissection and degenerative TAAAs. However, patients with degenerative TAAAs were more prone to early complications.
Physician-modified fenestrated and branched stent grafts had acceptable midterm outcomes and may be a reasonable approach for complex TAAAs. The procedure achieved similar early and midterm outcomes in postdissection and degenerative TAAAs; however, patients with degenerative TAAAs were more prone to early complications.
Fenestrated and branched endovascular aortic repair (F/B-EVAR) is an established technique to treat degenerative thoracoabdominal aneurysms (TAAAs) with satisfactory early and midterm results.
However, PD-TAAAs present with different characteristics than degenerative TAAAs (DG-TAAAs). There are many differences in endovascular technique requirements and planning design strategies in the treatment of PD- vs DG-TAAAs. The unique characteristic of PD-TAAAs, namely narrow true lumen (TL), stiff chronic dissection flap, and TL and false lumen (FL) origins of visceral arteries, pose increased technical challenges for repair.
Although physician-modified fenestrated and branched endografts (PMEGs) were proposed as an alternative approach to TAAAs in 2012, its use is still limited by the lack of long-term data in large series. The outcomes of both pathologies had been reported only by a few centers and clinical experience is lacking. We sought to compare the midterm outcomes of PMEGs in patients with PD- and DG-TAAAs.
Materials and Methods
Study Design and Patient Population
A retrospective cohort study was undertaken that included all consecutive patients treated with PMEGs between January 2017 and December 2020 at Nanjing Drum Tower Hospital. All patients were considered unfit to undergo open surgical aortic repair due to advanced age or severe comorbidities (American Society of Anesthesiologists classification ≥III). Indications for surgery were as follows
: aortic aneurysm size >5.5 cm or rapid growth (5 mm/year) and persistent back pain, a frank or impending rupture, organ/lower limb malperfusion, or other aneurysm-related complications; patients who had previous heart surgery or history of median sternotomy and are unfit for open surgical repair or the anatomic exclusion criteria for PMEGs, including small diameter (iliac artery <6 mm or target vessel [TV] <3 mm); excessive angulation or heavily calcified access; and aberrant or early branching. The study was approved by the local institution review board (No. 2017-015-05; December 2016).
Operative Planning and Device Design
Measurements were based on high-resolution computed tomography angiography (CTA) datasets. Procedure planning and device sizing were performed using a dedicated 3-dimensional vascular imaging workstation and Osirix MD software (Pixmeo SARL). The device design was based on analysis of the centerline of the flow measurements to determine accurate estimates of length, axial clock position, arc length, and angles. In dissected anatomies where the TL partially collapsed in a crescent shape, the diameter of the endograft was planned based on the maximum length of the crescent from 1 extremity to the other. In some cases, with a dissected common iliac artery, an oversized limb, or an iliac bifurcation device was positioned to push the dissection flap against the wall and achieve late-period FL thrombosis.
All patients were treated using PMEGs based on the Zenith TX2 platform (Cook Medical) or the Ankura endograft platform (Lifetech). Based on the extent of the aneurysm, vessel angulation, and the aortic diameter at the level of each target artery, PMEGs were designed with fenestrations, scallops, or cuffed branches. A diameter-reducing wire (ie, a constraining wire) was used to facilitate TV catheterization. Because balloon-expanding stent grafts (eg, Advanta [Atrium Medical Corp], Begraft [Bentley InnoMed], or VBX [W.L. Gore & Associates]) were unavailable in mainland China, fenestrations/branches were secured with a self-expandable covered Fluency stent graft (CR Bard) and a Viabahn stent graft (W.L. Gore & Associates).
The PMEGs were constructed under strict sterile conditions in the operating room on the table using basic surgical instruments and preoperative measurements obtained from reconstructed CTA images using centerline flow analysis software used for centerline flow analysis, such as Endosizes, Osiri MD, 3 Mensions software. The use of a constraining wire allowed partial deployment of the graft and facilitated cannulation of the branches through the fenestrations while preserving the ability to rotate the graft, which was essential to achieve perfect alignment and optimal outcomes. Once all the target vessels were accessed and the bridging stent grafts were ready for deployment, the main stent graft was fully unconstrained.
To address the narrow TL, a long proximal stent graft could be deployed in the descending thoracic aorta, landing 2 to 3 cm above the celiac trunk to re-expand the TL somewhat (Figure 1). Catheterization of TVs originating from the FL could be performed via an existing re-entry tear. In cases in which the re-entry tear was not located at the level of a perivisceral artery, several methods could be used. These methods comprised using wires with tips that could be stiffened, the back of a wire, a wire with the support of a guiding sheath, or even a needle to perforate the dissection flap.
Figure 1Patient presented with type A dissection treated with ascending and arch replacement and frozen elephant trunk. A, Preoperative computed tomography (CT) scan 3-dimensional volume rendering and multiplanar reconstructions. B, The physician-modified fenestrated and branched endografts (PMEG) with 4 fenestrations. C-E, Repair of dissection aorta aneurysm with PMEGs and good perfusion of all targeted vessels. F, Staged repair of abdominal aorta and CT angiography showed aneurysm exclusion and good perfusion of all targeted vessels.
Staging was indicated in patients with Crawfor extent I or extent II TAAAs at low risk of rupture. Among these patients, the most common approach was sequential aortic coverage wherein the thoracic stent grafts were placed first. The second-stage procedure involved placement of the fenestrated and branched devices or an additional bifurcated graft and contralateral limb.
Follow-up
Patients underwent clinical and laboratory examinations before discharge. Follow-up comprised CTA 1 month, 6 months, and 12 months after the procedure and yearly thereafter. Early follow-up was defined as the first 30 postoperative days. Branch instability was defined as branch-related endoleaks, disconnections, branch occlusion, device migration influencing a branch, and branch-related growth. Major adverse events (MAEs) analyzed were myocardial infarction and cardiac death, respiratory and renal failure, lower limb/bowel ischemia, spinal cord ischemia, and stroke. A deterioration in renal function was defined as a >30% decline in baseline estimated glomerular filtration rate. Technical success was defined as successful stent graft deployment with revascularization of all intended vessels and aneurysm exclusion without type I or type III endoleak, target occlusion, or conversion to open repair. End points included 30-day mortality, technical success rate, MAEs, TV instability, reinterventions, and endoleak.
Statistical Analysis
Statistical analyses were performed using SPSS version 20.0 (IBM Corp). Continuous variables were reported as number with mean ± SD or median with range, whereas categorical variables were expressed as frequencies. Data for the pooled analysis are given as proportion and 95% CI. All P values are 2-sided. Survival, freedom from endoleak, branch instability, and reintervention were assessed by Kaplan-Meier analyses.
Results
Baseline Characteristics
A total of 126 patients (aged 68 ± 13 years; 101 men [80.2%]) with TAAAs were treated by PMEGs from 2017 to 2020, including 72 PD-TAAAs and 54 DG-TAAAs. Hypertension and coronary artery disease were present in 109 (86.5%) and 12 (9.5%). PD-TAAAs patients were younger (aged 63 ± 10 vs 75 ± 12 years; P < .001) and more likely to have diabetes (26.4 vs 11.1; P = .03), history of previous aortic repair (76.4% vs 22.2%; P < .001), and smaller aneurysm size (52 vs 65 mm; P < .001) (see Table 1). There was no significant difference in terms of preoperative risk factors and comorbidities except diabetes mellitus (P = .03), which was more frequent in the PD-TAAA patients.
Table 1Demographic and clinical characteristics of 126 patients treated by physician-modified fenestrated and branched endografts for postdissection and degenerative aneurysm
Variable
Overall (n = 126)
Postdissection (n = 72)
Degenerative (n = 54)
P value
Demographic characteristic
Age (y)
68 (13)
63 (10)
75 (12)
.001
Sex
Male
101 (80.2)
59 (81.9)
42 (77.7)
.56
Female
25 (19.8)
13 (18.0)
12 (22.2)
.56
Cardiovascular risk factors
Cigarette smoking
88 (69.8)
52 (72.2)
36 (66.7)
.50
Hypertension
109 (86.5)
63 (87.5)
46 (85.2)
.70
Coronary artery disease
12 (9.5)
6 (8.3)
6 (11.1)
.80
Chronic pulmonary disease
24 (19.0)
17 (23.6)
7 (13.0)
.13
Diabetes mellitus
25 (19.8)
19 (26.4)
6 (11.1)
.03
eGFR<90 mL/min/1.73 m2
36 (28.6)
12 (16.7)
24 (44.4)
.12
Stroke
20 (15.9)
8 (11.1)
12 (22.2)
.09
Other history
Immunological diseases
4 (3.2)
2 (2.8)
2 (3.7)
.77
Prior open aortic repair
16 (12.7)
16 (29.2)
0
.001
Prior endovascular aortic repair
51 (40.5)
39 (54.2)
12 (22.2)
.001
ASA class III
62 (49.2)
40 (55.6)
22 (40.7)
.10
ASA class IV
47 (37.3)
23 (32.0)
24 (44.4)
.15
Aneurysm diameter (mm)
58 (19)
52 (16)
65 (12)
.001
Aneurysm type
Extent I TAAAs
16 (12.7)
12 (16.7)
4 (7.4)
.063
Extent II TAAAs
63 (50.0)
46 (52.8)
17 (31.5)
.001
Extent III TAAAs
14 (11.1)
5 (8.1)
9 (16.7)
.087
Extent IV TAAAs
33 (26.2)
9 (14.5)
24 (37.5)
.800
Type of pathology
.001
Post-TAAD
16
16
0
.001
Post-TEVAR TBAD
39
39
0
.001
Post-TBAD
17
17
0
.001
Categorical variables are presented as n (%), whereas continuous variables are presented as mean ± SD. P < .05 represent significant difference. eGFR, Estimated glomerular filtration rate; ASA, American Society of Anesthesiologists; TAAA, thoracoabdominal aortic aneurysms; TAAD, type A aortic dissection; TEVAR, thoracic endovascular aortic repair; TBAD, type B aortic dissection.
The time required to modify a stent graft to create 2 to 4 fenestrations, constrain the device, and resheath it varied in our experience from 30 to 65 minutes, with an average of 50 minutes. This time interval often coincided with the preparation of the patient by the anesthesiology team and the time needed for bilateral femoral artery exposure and upper extremity artery access. Forty-two (1 out of 3) cases used a preload wire technique. A total of 405 TVs were treated with PMEGs altered with 335 reinforced fenestrations, 2 scallops, and 68 branches. The vessels comprised 66 celiac artery (CA), 113 superior mesenteric arteries, and 226 renal arteries. Embolization and unreinforced fenestrations were performed in 18 and 33 celiac arteries, respectively; The remainder of 9 CA vessels were compromised at the early stage of the learning curve. Twelve (16.7%) neofenestrations were created by penetration and angioplasty of the dissection membrane opposite the TV. A total of 56 (13.8%) cuffed branches were used for longer overlap for TVs. The mean number of TVs per patient was 3.2 ± 0.9. Cerebrospinal fluid drainage was used in 1 patient (0.8%) undergoing Crawford extent type II DG-TAAA. There were no differences in the volume of contrast material used, mean fluoroscopy time, and total radiation dose. Patients with DG- vs PD-TAAA had higher estimated blood loss volumes (502 mL vs 342 mL; P = .01), packed red blood cell transfusion rates (53.7% vs 30.6%; P = .009), and temporary iliofemoral conduits (29.2% vs 6.9%; P = .001). No significant differences were found in the technical success rate (98.6% vs 96.3%; P = .40), or the length of hospital stay (17 ± 7 days vs 17 ± 10 days; P = .90) between the groups. Seventeen patients (13.5%) received a 1-stage internal iliac artery branch device to treat an iliac artery aneurysm. A staged approach was used in 48 (38.1%) elective patients (Table 2).
Table 2Procedural details of 126 patients treated physician-modified fenestrated and branched endografts for postdissection and degenerative aneurysm
Variable
Overall (n = 126)
Postdissection (n = 72)
Degenerative (n = 54)
P value
Urgent procedure (%)
64 (50.8)
40 (80.6)
24 (44.4)
.21
CSFD (%)
1 (0.8)
0
1(1.9)
.24
Upper extremity approach (%)
Left side
117 (92.9)
72 (100)
45 (83.3)
.001
Right side
5 (4.0)
0
5 (9.3)
.008
Iliac branch device (%)
17 (13.5)
13 (18.1)
4 (7.4)
.08
Staged repair (%)
48 (38.1)
27 (37.5)
21 (38.9)
.87
Contrast material volume (mL)
236 ± 69
223 ± 24
252 ± 49
.20
Fluoroscopy time (min)
69 ± 12
63 ± 9
74 ± 19
.68
Total radiation dose (mGy)
2107 ± 500
1812 ± 318
2201 ± 504
.70
Temporary iliofemoral conduits (%)
21(16.7)
5 (6.9)
16 (29.2)
.001
preload wire technique (%)
42 (33.3)
20 (27.8)
22 (40.7)
.127
Total operating time (min)
332 ± 115
326 ± 108
339 ± 126
.52
EBL (mL)
410 ± 324
342 ± 199
502 ± 425
.01
Transfusion PRBC (%)
51 (40.5)
22 (30.6)
29 (53.7)
.009
Paraplegia
0
0
0
NA
Paralysis
1
1
0
.39
Hospital stay (d)
17 ± 8
17 ± 7
17 ± 10
.90
Vessels per patient
3.2 ± 0.9
3.3 ± 0.9
3.2 ± 1.0
.38
Technical success（%）
97.6
98.6
96.3
.40
Categorical variables are presented as n (%) and continuous variables are presented as mean ± SD. P < .05 represent significant difference. CSFD, Cerebrospinal fluid drain; EBL, estimated blood loss; PRBC, packed red blood cell; NA, not applicable.
There were 4 deaths (3.2%) within the first 30 days, and the DG-TAAAs group had a greater than 4-fold higher rate of perioperative deaths. However, there was no statistically significant difference in the death rate between the groups (1.4% vs 5.6%; P = .19). One (1.4%) patient undergoing dissection aneurysm died on the seventh postoperative day because of rupture of a distal stent-induced new entry. The remaining 3 patients who died were all in the DG-TAAA group. One patient with a symptomatic type III TAAA died on day 3, and postoperative massive peripheral embolism (ie, shaggy aorta syndrome) was considered the cause of death. At admission, CTA of the descending aorta in this patient showed severe arteriosclerosis and a massive mural thrombus. One patient died following multiple organ failure after successful aneurysm repair with a PMEG. One patient developed a renal capsular hematoma and died on postoperative day 2 owing to hypovolemic shock. Twenty-four patients (19%) experienced MAEs, with more MAEs in the DG-TAAAs (23.7%) than in the PD-TAAAs group (12.5%; P = .03). The most common MAEs were acute kidney injury in 9 patients (7.1%), with 4 (3.1%) requiring new-onset dialysis; respiratory failure in 7 patients (5.5%); bowel ischemia in 2 patients (1.6%); major stroke in 5 patients (3.9%); myocardial infarction in 5 patients (3.9%); and lower limb ischemia in 3 patients (2.4%) (Table 3). One Crawford extent type II dissection aneurysm patient developed transient spinal cord injury (SCI) with no clinical signs or symptoms of permanent paraplegia.
Table 3Adverse events of 126 patients treated by physician-modified fenestrated and branched endografts for postdissection and degenerative thoracoabdominal aortic aneurysm
Variable
Overall (n = 126)
Postdissection TAAA (n = 72)
Degenerative TAAA (n = 54)
P value
30-d Mortality
4 (3.2)
1 (1.4)
3 (5.6)
.19
Major adverse events
23 (18.3)
9 (12.5%)
14 (23.7)
.03
Acute kidney injury
9 (7.1)
3 (4.2)
6 (11.1)
.13
Dialysis
4 (3.1)
1 (1.4)
3 (5.6)
.19
Respiratory failure
7 (5.5)
2 (2.8)
5 (9.3)
.11
Bowel ischemia
2 (1.6)
1 (1.4)
1 (1.9)
.84
Major stroke
5 (3.9)
1 (1.4)
4 (7.4)
.08
Myocardial infraction
5 (3.9)
1 (1.4)
4 (7.4)
.08
Lower limb ischemia
3 (2.4)
1 (1.4)
2 (3.7)
.40
Follow-up (mo)
37.5 ± 8.8
39.2 ± 8.2
35.8 ± 9.3
.90
Mortality
2 (1.6)
1 (1.4)
1 (1.9)
.81
Branch vessels
392
234
158
–
Branch occlusion, stenosis
9 (2.3)
5 (2.1)
4 (2.5)
.67
Bridging stent migration
6 (1.5)
2 (1.3)
4 (2.5)
.17
Values are presented as n (%), mean ± SD, or n. TAAA, Thoracoabdominal aortic aneurysm.
The mean follow-up duration was 3.21 ± 0.67 years for those with PD-TAAAs and 3.01 ± 1.04 years for those with DG-TAAAs. All-cause mortality data were reported for 2 (1.6%) patients during midterm follow-up. Aortic-related mortality was 0.8% (1 out of 122). One Crawford extent type II PD-TAAA patient who had a previous thoracic endovascular aortic repair died of retrograde type A dissection 7 months after discharge. One DG-TAAA patient died in the postoperative third month because of upper gastrointestinal bleeding. The 3-year cumulative survival probability was 97.2% and 92.6% for the PD- and DG-TAAAs, respectively (P = .23) (Figure 2, A). The freedom from any type endoleak was 86.9% and 90.2% (P = .81) (Figure 2, B) for PD- and DG-TAAAs groups, respectively. The estimated freedom from branch instability at 3 years was 97.3% vs 97.4% for the PD- and DG-TAAAs (P = .99) (Figure 2, C). The estimated freedom from reintervention for the PD- and DG-TAAAs at 3 years were 85.8% and 92.3%, respectively (P = .26) (Figure 2, D).
Figure 2Kaplan-Meier curves stratified by dissection thoracoabdominal aortic aneurysm (TAAA) and degenerative TAAA demonstrating estimates of survival (A) during follow-up, freedom from any type endoleak (B), freedom from the branch instability (C), and reintervention (D). Thick lines depict the point estimates. The dashed thin line represents the 95% confidence limits. PD, Postdissection; DG, degenerative.
Although fenestrated endograft technology had existed worldwide for more than a decade and was associated with encouraging outcomes in the treatment of complex aortic pathologies, unfortunately, custom-made devices or off-the-shelf devices were unavailable in many countries, including China. Facing this issue, we turned to PMEGs as an alternative solution. We herein described our early and midterm outcomes of PD- and DG-TAAAs submitted to PMEGs. In our series, the 30-day mortality was 3.2% and the pooled technical success rate was 97.6%. The 3-year cumulative survival probability was 97.2% and 92.6% for the PD- and DG-TAAAs, respectively (P = .23). A report from a high-volume aortic center in the United States reviewed outcomes of 145 TAAA patients treated with custom-made devices and PMEGs.
The authors reported that the 30-day mortality rate was 5.5% for PMEGs (pararenal aortic aneurism, 1.2%; Crawford extent IV, 3.8%; and Crawford extent I-III, 17.1%) and 0.0% for custom-made devices (P = .002). However, at 3 years, no significant difference was observed between the PMEG and custom-made device groups with respect to the mean rate of patient survival (68% ± 4% vs 67% ± 8%; P = .11), freedom from reintervention (68% ± 4% vs 68% ± 8%; P = .17), or secondary TV patency (98% ± 1% vs 98% ± 1%; P = .89). Considering that patients treated with PMEGs had more comorbidities than those treated with custom-made devices, they concluded that PMEGs were likely to have an even greater influence on fit patients. Following the satisfactory results of F/B-EVAR in DG-TAAA repair, some centers proposed to approach PD-TAAAs with this technology.
Outcomes of endovascular repair of chronic postdissection compared with degenerative thoracoabdominal aortic aneurysms using fenestrated-branched stent grafts.
This feature was present in our cohort, within whom the mean age was 63 years (vs 75 years) (P < .001). Patients with PD-TAAAs could have a chronic type B aortic dissection or a previous aortic surgical/endovascular repair of type A aortic dissection. In our series, 39 (51.2%) of cases were previously submitted to endovascular aortic procedures and 16 (22.2%) had a previous open aortic arch repair. The latter might increase the technical difficulty of arterial access from above; for example, in cases of acute angle reimplantations or frozen elephant trunk. Open repair of PD-TAAAs was usually more complex than that of DG-TAAAs with significant postoperative mortality and morbidity.
PD remodeling was exceptionally difficult to treat in patients with a narrow TL that could not accommodate the size of branched stent grafts. Another technical challenge in PD aneurysm was the presence of aortic branches originating from the FL or from both a TL and FL. In our study, in cases without an intimal tear identified on preoperative CTA, a neofenestration was created by penetration and angioplasty of the dissection membrane opposite the TV. When working in the narrow TL, we explored and applied some special techniques to overcome such difficulties, which are as follows:
•
Staged approach with stenting of the thoracic aorta up to the level of the CA. This allowed further expansion of the TL and gradual thrombosis of the FL in the thoracic aorta.
•
Balloon molding of the renal-mesenteric segment. This usually helped to expand the TL somewhat, allowing endograft devices to completely deploy. However, the intimal flap in the chronic phase was rigid and might cause incomplete endograft deployment, difficult repositioning, and require catheterization. Those factors might contribute to or be indicators of the unfavorable postoperative outcomes. Patients with a rigid intimal flap might not be good candidates for total endovascular aortic repair in our center's experience.
•
Use of fenestrations or inner branches for narrow TLs in the renal-mesenteric segment that originate from a small diameter stent-graft using preloaded catheters or a guidewire for access using a brachial approach.
•
Upper extremity access and staggered deployment of PMEG.
•
Double diameter-reducing ties to achieve a further diameter reduction and facilitate the rotation and adjusting after initial deployment.
For more extensive PD aneurysm, a more complex endovascular repair was required to exclude all entry and re-entry to prevent further rupture. If FL perfusion persists with a large retrograde flow after F/B-EVAR was performed, embolization procedures were required. Our series included FL embolization of the branch vessel that originated from a FL (usually a celiac artery), partly from the FL (superior mesenteric artery or renal artery). Communication tears located in visceral arteries should be addressed before or at the time of the F/B-EVAR procedure because later access could be difficult. There were many re-entry tears in the dissection segment, and retrograde flow continued to fill the FL into the dissection. The FL was then cannulated through a known tear previously reviewed on a CT scan. The FL was subsequently embolized with coils. Large coils were used, and these were packed with smaller coils. The aortic dissection usually extends to iliac arteries. Landing in a dissected common iliac artery might provide adequate sealing over time with complete re-expansion of the TL and obliteration of the retrograde FL flow during follow-up.
Angulation at the level of the perivisceral aorta also represented a formidable challenge for both measurement and alignment of branches to the fenestrations. The use of arc length measurements and centerline adjustments for angulated aortic anatomy/narrow TL were of great importance in preoperative planning strategies. Another issue that was especially relevant for patients with a thrombosis-loaded aneurysmal sac was the distance between the stent graft and the aortic branches. In aneurysms with larger diameters, the gap that the branched stents had to bridge was longer, and the stents were more susceptible to the cranial and caudal forces that occur during the cardiac cycle. This could potentially result in higher rates of fracture, kinking, and branch occlusion in cases where the branches did not oppose the aortic wall. To avoid this issue, we preferred cuffed branches over fenestrations in the treatment of patients with large aneurysmal sacs whenever possible. The use of cuffed branches has been suggested to provide a better seal between the main stent graft and the bridging branch artery. Cuffed branches can provide at least a 2 cm overlapping seal zone between the branch artery stent and the main stent graft (Figure 3). But branches might be compressed in PD aneurysms with narrow TL, so inner branches or fenestration were more suitable for PD aneurysm. Another option to further reinforce a branch stent bridging a longer gap and to add stability was to add another self-expandable bare stent to prevent kinking at the distal landing point. Generally, the increased anatomical complexity and modifications of any fenestrated/branched stent grafts introduced the potential for additional modes of failure regarding branch vessel compromise. Stent grafts did not always conform to angulated anatomy and tended to straighten along with the aorta when deployed. Severe aortic angulation might not be suitable for PMEG deployment and might be associated with less favorable outcomes.
Figure 3A, Preoperative computed tomography (CT) scan demonstrating type III thoracoabdominal aortic aneurysm (TAAA). B, The physician-modified fenestrated and branched endografts with 3 branches. C, An intraoperative angiogram demonstrating type III TAAA. D, Completion angiogram after successful repair of type III TAAA. E and F, Follow-up CT angiography showing aneurysm exclusion. G, Follow-up CT angiography showing good perfusion of all targeted vessels.
The sizing of TVs bridging stents requires particular attention. Patients with TVs originating from the FL or a larger-diameter aneurysmal sac could have longer distance between fenestrations/branches and the native vessels. Both these aspects could increase the risk of TV instability and could lead to a higher incidence of reinterventions. The European collaborators on stent/graft techniques for the Aortic Aneurysm Repair Registry presented midterm data that revealed that reintervention rates after F/B-EVAR differ significantly in reported series, with rates varying between 10% and 57%.
Need for secondary interventions after endovascular repair of abdominal aortic aneurysms. Intermediate-term follow-up results of a European collaborative registry (EUROSTAR).
The causes of reintervention were heterogeneous and seemed to be associated with TV instability and endoleaks. In accordance with data regarding reinterventions following F/B-EVAR, endoleaks (23%), persistent FL perfusion (18%), and proximal/distal disease progression (36%) were the most common indications for secondary intervention.
Need for secondary interventions after endovascular repair of abdominal aortic aneurysms. Intermediate-term follow-up results of a European collaborative registry (EUROSTAR).
For PD-TAAAs, the cause of reintervention seemed to be associated with high back flow in the FL from distal entry tear. In our study, the reintervention rate during follow-up was 10.7%. The higher reintervention rate might be partly attributable to higher numbers of type Ic and III endoleaks and lesion progression in aortic pathology (Table 4). For these reasons, particular attention could be required in the follow-up to timely detection and management of possible problems. In cases of aneurysmal evolution of the iliac artery or dissection at this level, an iliac branch device should be implanted to close the last distal tear or exclude the aneurysmal sac and prevent the risk of reintervention.
Table 4Description of endoleaks and secondary intervention in 126 patients treated by physician-modified fenestrated and branched endografts for postdissection and degenerative thoracoabdominal aortic aneurysm
and without significant overall improvement in recent years. The etiology of SCI is multifactorial and incompletely understood. In an analysis performed by Dias and colleagues,
the factors independently associated with SCI were Crawford extent type II TAAAs that had a higher contrast volume. Although the optimal approach to spinal cord protection was debated, the core principles were minimizing ischemia and reperfusion injury. No permanent SCI was observed in our cohort. Spinal cord protection strategies used in our center include staging, maintaining a mean aortic pressure >90 mm Hg, performing therapeutic cerebrospinal fluid drainage, ensuring early limb reperfusion, and performing left subclavian artery revascularization. We did not perform routine cerebrospinal fluid drainage because of the risk of hemorrhagic cerebral and spinal complications.
To achieve satisfactory outcomes and minimize the incidence of adverse events, adjunctive approaches were necessary. The preload wire design led to a statistically significant reduction in the sheath-to-TV time and, accordingly, in the total duration of the procedure.
This also meant shorter general anesthesia and fluoroscopy times, as well as shorter lower extremity ischemic times. Temporary iliofemoral conduits were needed selectively to reduce sheath-related complications, especially for patients with calcified and tortuous access vessels or diseased iliac and femoral arteries.
These experiences and practices, combined with our results, reflect that PMEGs played a significant role in the real-world practice of complex EVARs in China, short- and midterm results achieved by PMEGs were similar to custom-made devices or off-the-shelf devices, and PMEGs had favorable outcomes for PD-TAAAs.
The results of this analysis should be interpreted within the scope of the limitations of this study. This was a single-center, small sample size, and predominantly single-surgeon study that was performed retrospectively. We cannot make a comparison of PMEG with custom-made or off-the-shelf devices because they are not available in our country. The clinical follow-up time was relatively short because most of the procedures were performed during the past 4 years. There was also a learning curve associated with evolution of the technology that undoubtedly influenced our threshold for reintervention. Comparisons between groups was not adjusted by possible confounders. The selection of patients referred to our center for PMEGs also restricts our ability to draw conclusions about the applicability of the devices for use in the general population.
Conclusions
PMEGs were effective to treat PG-TAAAs and PD-TAAAs and there was no significant difference between groups. However, data suggest a higher morbidity of DG-TAAAs compared with PD-TAAAs.
Conflict of Interest Statement
The authors reported no conflicts of interest.
The Journal policy requires editors and reviewers to disclose conflicts of interest and to decline handling or reviewing manuscripts for which they may have a conflict of interest. The editors and reviewers of this article have no conflicts of interest.
The authors thank the clinical research coordinators in the participating centers and Min Zhou, MD, PhD, who provided technical help and writing assistance.
References
Verhoeven E.L.
Katsargyris A.
Bekkema F.
Oikonomou K.
Zeebregts C.J.
Ritter W.
et al.
Ten-year experience with endovascular repair of thoracoabdominal aortic aneurysms: results from 166 consecutive patients.
Outcomes of endovascular repair of chronic postdissection compared with degenerative thoracoabdominal aortic aneurysms using fenestrated-branched stent grafts.
Need for secondary interventions after endovascular repair of abdominal aortic aneurysms. Intermediate-term follow-up results of a European collaborative registry (EUROSTAR).
This work was supported by grants from the National Natural Science Foundation of China (81,771,952), Natural Science Foundation of Jiangsu Province (BK20171115).
00013-5&id=fx1.jpg){kind=link}
00013-5&id=gr1.jpg){kind=link}
00013-5&id=gr2.jpg){kind=link}
00013-5&id=gr3.jpg){kind=link}