University of Groningen
Endovascular approaches to complex aortic aneurysms
de Niet, Arne
DOI:
10.33612/diss.111895510
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2020
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
de Niet, A. (2020). Endovascular approaches to complex aortic aneurysms. University of Groningen. https://doi.org/10.33612/diss.111895510
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
CHAPTER
5
Endograft
conformability in
fenestrated endovascular
aneurysm repair for
complex abdominal aortic
aneurysm
Arne de Niet Esmé J. Donselaar Suzanne Holewijn Ignace F.J. Tielliu Jan Willem H.P. Lardenoije Clark J. Zeebregts Michel M.P.J. Reijnen
Ch
ap
ter 5
80
Abstract
Endovascular aneurysm repair may change the anatomy, with potential repercussions for graft-related complications. We compared the impact of two commercially available custom-made fenestrated endografts on patient’s anatomy.
From March 2002 until July 2016 all patients treated with a fenestrated endograft in two hospitals were screened for this cohort study. Inclusion criteria were availability of qualified pre- and post-operative computed tomography angiography (CTA) and treatment with an endograft solely containing fenestrations and no branches. Group A included patients treated with the Zenith™ Fenestrated endograft and group B included patients treated with the Fenestrated Anaconda™ endograft. Measured variables were aortic diameter at the level of the superior mesenteric artery (SMA) and renal arteries (RA), target vessel angle, target vessel clock position, and the target vessel tortuosity index (TI). Variables were tested for inter- and intra-observational agreement. Changes within groups and differences between groups were analyzed.
145 patients met the inclusion criteria. Group A contained 110 patients and group B 35 patients. There was a good observational agreement in all tested variables. The native anatomy changed in both groups after implantation of the endograft. In group A changes were seen in the angle of the celiac artery (CA) (P = .012), SMA (P = .022), left renal artery (LRA) (P < .001) and right renal artery (RRA) (P < .001) angles, the aortic diameter at the SMA level (P < .001), and the clock position of the LRA (P < .001) and RRA (P < .001). In group B changes were seen in the angle of the LRA (P = .001) and RRA (P < .001) angle, and in the SMA TI (P = .044). Between groups differences in change were seen for the aortic diameter at the SMA and RA level (P < .001 and P < .001, respectively), and LRA clock position (P = .019). Both custom-made fenestrated endografts alter vascular anatomy. The data suggest a higher conformability of the Fenestrated Anaconda™ endograft compared to the Zenith® Fenestrated.
5
Introduction
The treatment of choice for abdominal aortic aneurysm (AAA) repair has shifted from open repair towards endovascular aneurysm repair (EVAR), mainly because of favorable early results.1-4 The boundaries of EVAR kept being pushed and fenestrated endografts enabled
endovascular treatment of short-neck AAAs, juxtarenal and suprarenal AAAs. The fenestra-tions maintain flow through visceral arteries while the endograft seals in a healthy neck above the aneurysm.5 For optimal case planning pre-operative assessment of aorto-iliac
morphology is essential. These measurements, usually performed by 3D-reconstructions of computed tomography angiography (CTA) have good inter-observational agreement for EVAR planning.6-8 As a software adjunct, the central lumen line (CLL) allows reconstruction
of the aorta and branching arteries, and is used in both standard and complex endovascular AAA repair.9
The Cook Zenith® Fenestrated endograft (Cook Medical Inc. Bloomington, IN, USA) was the first commercially available fenestrated design and consisted of multiple self-expanda-ble stainless steel Z-stents and full-thickness woven polyester fabric, containing fenestrations between the struts of the Z-stent.10,11 The more recently introduced Fenestrated Anaconda™
endograft (Terumo Aortic, Inchinnan, Scotland) consists of multiple independent nitinol rings, a woven polyester graft and an unsupported proximal body containing
fenestrati-ons.12,13 The fenestrations and target vessels are cannulated and stented after deployment of
the main body. The choice of endograft is mostly based on the experience and preference of the clinician.
EVAR may change the native anatomy of the patient.14 Such a conformational change may lead
to a proximal seal zone failure through infrarenal aortic angle change or iliac limb complica-tions through iliac artery tortuosity changes.15,16 Different infrarenal endograft designs have a
different conformability, and therefore, the choice of endograft influences the risk of compli-cations.16,17 After fenestrated endovascular aneurysm repair (FEVAR), the endograft
implan-tation and placement of stents in the target vessels, influence arterial angle and curvature.18
Altered anatomy could potentially kink the stented target vessel, strain the endograft, lead to fatigue or create thrombosis and distal emboli. The exact influence of fenestrated endograft implantation on human aortic anatomy is unknown.19 The differences in design
Ch
ap
ter 5
82
anatomy, which might be of importance for graft related complications, particularly involving the stents in the target vessels. The aim of the present study was to assess the conformability of two commercially available fenestrated endografts and to study the differences in anatomic changes after stent placement.
Methods
All patients treated from March 2002 to July 2016 in two Dutch hospitals were screened for inclusion. Patients who underwent fenestrated endovascular aneurysm repair and had pre- and postoperative CTA assessments with a slice thickness of ≤2 mm were eligible for inclusion. Group A included patients treated with the Zenith® Fenestrated endograft and group B included patients treated with the Fenestrated Anaconda™ endograft. Thoracic or branched EVAR patients were excluded. The study was conducted in accordance with the principles of the Declaration of Helsinki and Good Clinical Practice guidelines. Retrospective “patient’s files” research is not in scope of the Dutch WMO (Wet Mensgebonden Onderzoek: Law human bound research) and a waiver of the Institutional Review Board was obtained that their review was not necessary (reference number M17.207929). As a consequence, no informed consent was obtained. Patients’ data were analyzed anonymously.
Pre-operative patient characteristics were gathered, including ASA-classification.20 Patient
characteristics were classified according to the reporting standards of the Society of Vascular Surgery (SVS), and a higher SVS score related to a higher peri-operative mortality risk.21
Technical success was considered an endograft deployed as planned, including stented fe-nestrations, in the absence of a type I or type III endoleak, and the absence of conversion or death, extending 24 hours post-operatively. Assisted technical success was considered in case an endovascular adjunctive procedure was necessary during the first 24 hours post-operati-vely.22
Analysis of CTA parameters
Measurements were performed using Aquarius iNtuition™ Version 4.4.7 (TeraRecon, Inc. Foster City, CA, USA) and Philips IntelliSpace Portal 8.0 (Philips Healthcare, Eindhoven, The Netherlands). An automatically drawn CLL was manually adjusted when necessary, and 3D-reconstructions were automatically created. Measurements were performed using the pre-
5
and first post-operative CTA scans. Measurements included the maximum aortic diameter between the upper and lower margins of the superior mesenteric artery (SMA) and between the upper margin of the most cranial renal artery (RA) and lowest margin of the most caudal RA. All target vessels were measured for tortuosity index (TI) (Figure 1A)23, clock position
(Figure 1B) and angle relative to the aortic CLL (Figure 1C). On the post-operative CTA, two straight lines were drawn along three points on the CLL of the stented target vessel. The first point was placed at the distal marker of the target vessel stent, the second was placed 1 cm proximal to the distal marker, and the third
was placed 1 cm distal of the distal marker. One straight line was drawn from point one to two and the second straight line from point one to point three. The angle between both straight lines was measured. Variables were truncated towards millimeters, TI towards two decimals and degrees truncated to zero decimals.
Figure 1: Reconstructions of aorta by computed tomography angiography measurements. A: Tortuosity index, measuring the distance over the central lumen line (CLL) from origin of target vessel to the first bifurcation of >50% in diameter of the main branch. B: Clock position, spine orientation is reset on dorsal side and the target vessel is measured relative to a straight line from CLL to 12 o’clock. Time after 12 is labeled positive (+) and before 12 is labeled negative (-). C: target vessel angle, the angle of CLL of the target vessel from origin to 1 cm from origin relative to CLL of the aorta.
Perpendi-cular orientation is 90o and downward orientation counts
Ch
ap
ter 5
84
Statistics
Continuous variables were tested for normal distribution by observation of Q-Q plots and reported in the tables. Normal distributed variables were reported as mean with standard deviation (SD) and skewed variables as median with interquartile range (IQR). Discrete variables were presented with frequencies and percentages. Differences in continuous data between groups of baseline patient and anatomic characteristics were tested with the student T-test or with the Mann-Whitney U test in skewed data. Differences in discrete data between groups were tested with the Fisher’s exact test. Changes in anatomy within groups were tested with paired student T test or the Wilcoxon signed-rank test in skewed data. The difference in anatomic change between balloon expandable (BE) covered stents, self-expandable (SE) bare metal stents or a combination within groups were tested with the ANOVA for repeated measures and Kruskal Wallis for single measures.
To define intra- and interobserver variability, the first observer (AN) did all tested measu-rements in both systems, and did repeated measumeasu-rements (pre- and post-operative CTA) in 20 randomly assigned cases in Aquarius iNtuitionTM. The second observer (ED) measured
variables (pre- and post-operative CTA) in 20 randomly assigned cases in Aquarius iNtuitionTM. The second observer measured variables (pre- and post-operative CTA) in 20
randomly assigned cases in Philips IntelliSpace and did repeated measurements in those same cases (post-operative CTA only). Consistency agreement was tested with the two-way mixed intra-class correlation coefficient (ICC). An ICC below 0.500 indicated a poor, between 0.500 and 0.750 a moderate, between 0.750 and 0.900 a good, and between 0.900 and 1.000 an excellent reliability.24 Observer one repeatedly measured with Aquarius iNtuitionTM and
measured 20 randomly assigned cases with Philips IntelliSpace, and observer two repeatedly measured with Philips IntelliSpace and measured 20 randomly assigned cases with Aquarius iNtuitionTM. Both operating systems were analyzed separately for observer variability. Within
each operating system, methods of measuring aortic diameter and visceral arteries were identical and therefore combined to test observer variability. Combined variables were aortic diameter at SMA with RA, all TIs (both pre- and post-operative vessels), all clock positions, all target vessel angles relative to aortic CLL and all target vessel angles distal of the stent. Analysis of anatomic change within and between groups was performed with separated variables and in stented target vessels only. P-values <0.05 were considered statistically significant. Statistical analysis was done with IBM® SPSS® Version 23.0.0.3 (Armonk, NY, USA).
5
Results
A total of 234 patients were treated with a fenestrated endograft since the technique was first introduced at the two participating centers in 2002. Eighty-nine patients were excluded because the available CTAs did not meet inclusion criteria, consequently resulting in 145 included cases for further analysis. One-hundred and ten patients were assigned to group A and 35 to group B. Baseline patient characteristics are shown in table I.
In group A balloon expandable (BE) covered stents were used in 85 cases (77.3%), self- expandable (SE) bare metal stents in 20 cases (18.2%) and in five cases (4.5%) a combination of them was used for the target vessels. The cases with SE bare metal stents were the first treated FEVAR cases. In group B all target vessels were stented with a BE covered stent. Table II shows procedural results for both groups. At completion angiography overall there were more endoleaks in group B (34 cases (31.0%) in group A vs. 18 cases (51.4%) in group B, P = .027) and included a type Ia endoleak in seven (6.4%) cases in group A, and in six (17.1%) cases in group B (P = .053). On the post-operative CTA, two (1.8%) type Ia endoleaks were diagnosed in group A and one type Ia (2.9%) endoleak in group B (P = .391). No reinterven-tions had been performed to resolve an endoleak until after the first postoperative CTA. Type II endoleaks were seen in 26 cases (23.6%) in group A, and 12 (34.3%) in group B (P = .214). Until the first post-operative visit one type III endoleak was seen in group A (0.9%), none in group B (P = .573).
In group A 37 intraoperative adjunctive procedures were performed in 33 patients. Additional stenting of a target vessel was necessary in 12 cases, because of kinking, stenosis, endoleak, overly short stent or a puncture hole in a target vessel. One patient had an embolus to the right kidney, successfully treated with thrombolysis. In four cases the overlap between main endograft parts was insufficient and an additional cuff was placed. The other adjunctive procedures were treatments of an iliac or femoral artery. A total of 11 adjunctive procedures were performed in 10 patients in group B. Additional angioplasty was done for non-stented visceral arteries in two cases, and reinforcing stenting of a stented target vessel in one case. In one case there was an inability to cannulate the CA. To prevent endoleak a proximal extension cuff was introduced to seal the fenestration. All other adjunctive procedures concerned the iliac or femoral artery.
Ch
ap
ter 5
86
TABLE I: PRE-OPERATIVE PATIENT CHARACTERISTICS AND PROCEDURAL DETAILS BY TYPE OF ENDOGRAFT. Group A (Zenith® Fenestrated) Group B (Fenestrated Anaconda™) Difference
Variable Values Values P value
Sex Male 94 (85.5%) 30 (85.7%) .97
Female 16 (14.5%) 5 (14.3%)
Age, mean in years (SD) 72.4 (7.1) 72.3 (7.3) .94
Weight, mean in kg (SD) 84.5 (14.2) 86.4 (13.8) .75
Height, mean in cm (SD) 176.1 (8.9) 173.6 (7.8) .17
BMI, mean kg/m2 (SD) 27.6 (3.9) 28.7 (4.5) .20
Plasma creatinine level, median in μmol/L (IQR)† 89 (77 - 109) 94 (76 - 108) .88
Pre-operative smoker No 40 (36.4%) 5 (14.3%) .14 Yes 39 (35.5%) 10 (28.6%) Previous 25 (22.3%) 10 (28.6%) Unknown 6 (5.5%) 10 (28.6%) Hypertensiona 0 21 (19.1%) 7 (20.0%) .27 1 39 (35.5%) 10 (28.6%) 2 34 (30.9%) 8 (22.9%) 3 16 (14.5%) 10 (28.6%) Hypercholesterolemia, number (%) 77 (70.0%) 28 (80.0%) .25
Diabetes mellitus, number (%) 13 (11.8%) 5 (14.3%) .77
Cerebrovascular disease, number (%)b 14 (12.7%) 3 (8.6%) .36
Peripheral artery disease, number (%) 9 (10.0%) 5 (14.3%) .53
Cardiac statusa 0 42 (38.2%) 19 (54.3%) .015 1 21 (19.1%) 11 (31.4%) 2 32 (29.1%) 5 (14.3%) 3 15 (13.6%) 0 (0%) Pulmonary statusa 0 72 (65.5%) 31 (88.6%) .028 1 17 (15.5%) 4 (11.4%) 2 15 (13.6%) 0 (0.0%) 3 6 (5.5%) 0 (0.0%) ASA-classification I 0 (0.0%) 0 (0%) .19 II 29 (26.4%) 15 (42.9%) III 73 (66.4%) 20 (57.1%) IV 2 (1.8%) 0 (0.0%) Unknown 6 (5.5% 0 (0.0%)
Aneurysm location Infrarenal 43 (39.1%) 20 (57.1%) .27
Juxtarenal 56 (50.9%) 13 (37.1%)
Suprarenal 11 (1.8%) 2 (5.7%)
Type IV thoraco-abdominal 2 (1.8%) 0 (0.0%)
Previous operation OSR 10 (9.9%) 1 (2.9%) .59
EVAR 8 (7.2%) 3 (8.6%)
† Not normally distributed, presented with median and IQR. a as described by Chaikof et al. 200221. b A history of
cerebrovascular accident or transient ischemic attack. IQR: Interquartile range, SD: Standard deviation, COPD: Chronic obstructive pulmonary disease, OSR: Previous open surgical repair with proximal paraanastomotic aneurysm, EVAR: Previous endovascular aneurysm repair with complicated course.
5
TABLE II: PROCEDURAL RESULTS
Group A (Zenith® Fenestrated)
Group B (Fenestrated
Anaconda™) valueP
Procedure time (median in min, IQR) 215 (IQR 180 – 291)† 238 (IQR 170 – 315)† .81
Contrast volume (median in ml, IQR, Iodine 300ml/L) 180 (IQR 150 – 220)† 130 (IQR 116 – 194)† .001
Estimated blood loss (median in ml, IQR) 250 (IQR 150 – 497)† 100 (IQR 63 – 175)† <.001
Iliac extension Bifurcated 96 (87.3%) 31 (88.6%) .40
Uni-iliac 2 (1.8%) 2 (5.7%)
Cuff 12 (10.9%) 2 (5.7%)
Mean number of fenestrations (SD) 2.4 (0.8) 2.6 (0.8)
Cases with number of
fenestrations 12 16 (14.5%)49 (44.5%) 15 (42.9%)2 (5.7%) .49 3 36 (32.7%) 14 (40.0%) 4 9 (8.2%) 4 (11.4%) Adjunctive procedure 33 (30%) 10 (28.6%) .87 Endovascular 30 (27.3%) 8 (22.9%) .59 Open 6 (5.5%) 3 (8.6%) .69
† Not normally distributed, presented with median and IQR. IQR: Interquartile range, SD: Standard deviation.
Technical success rate was 88.2% (n = 97) in group A and 82.9% (n = 29) in group B (P = .402). Assisted technical success rate was 91.8% (n = 101) in group A and 92.9% (n = 29) in group B (P = .198).
Time between baseline CTA and treatment was 4.0 months (IQR 3.1 to 5.1) in group A and 3.3 months (IQR 2.3 to 4.3) in group B. Time between treatment and first post-operative CTA was 1.4 months (IQR 1.1 to 1.7) in group A and 1.0 month (IQR 0.1 to 1.8) in group B. Time between baseline CTA and treatment was shorter in group B (P = .012), and no difference was seen between groups in time between operation and first post-operative CTA (P = .055). Table III shows the combined variables with the ICC and the corresponding reliability. All tested variables had a moderate, good or excellent ICC, and therefore could be used for analysis in this study.
Ch
ap
ter 5
88
TABLE III: INTRA- AND INTER-OBSERVER VARIABILITY
Variable Intra-observer variability (observer 1)a Intra-observer variability (observer 2)b Inter-observer variability Aquarius iNtuition c Inter-observer variability Philips Intellispace d Aortic diameter (n) 80 80 80 80 Mean (diff) 25.9 (0.1) 27.4 (0.3) 26.4 (1.2) 26.2 (0.9) ICC 0.946 0.930 0.747 0.761 95% CI 0.917 – 0.965 0.872 – 0.963 0.632 – 0.830 0.651 – 0.840
Target vessel tortuosity index (n) 160 77 160 154
Mean (diff) 1.11 (0.00) 1.11 (0.00) 1.13 (0.03) 1.11 (0.01)
ICC 0.725 0.931 0.818 0.822
95% CI 0.642 – 0.791 0.893 – 0.955 0.759 – 0.863 0.764 – 0.868
Target vessel clock position (n) 160 77 160 154
Mean (diff) 14.3 (1.3) 10.1 (0.3) 11.4 (4.4) 10.0 (0.8)
ICC 0.963 0.999 0.942 0.954
95% CI 0.950 – 0.973 0.998 – 0.999 0.921 – 0.957 0.937 – 0.966
Angle of target vessel (n) 160 77 160 151
Mean (diff) 124.2 (0.2) 112.7 (0.5) 124.0 (0.7) 116.0 (0.1)
ICC 0.845 0.946 0.696 0.590
95% CI 0.794 – 0.884 0.916 – 0.965 0.607 – 0.768 0.475 – 0.685
Target vessel angle distal of stent (n) 50 52 50 52
Mean (diff) 153.8 (1.5) 144.9 (0.8) 154.1 (0.6) 151.0 (12.9)
ICC 0.750 0.924 0.717 0.561
95% CI 0.598 – 0.850 0.871 – 0.956 0.549 – 0.829 0.343 – 0.722
Mean is shown with mean difference between measurements. Pre- and post-CTA combined, including post-ope-rative measurements of unstented target vessels. Variables with the same measuring technique were combined to enlarge groups. In a few cases one of the target vessels was occluded pre-operatively or got occluded post-operati-vely, therefore could not be measured explaining the lower number of cases in target vessels. Only in stented target vessels the angle distal of stent could be measured.
An excellent observer reliability was observed for target vessel clock position. A moderate to excellent observer reliability was seen for aortic diameter, target vessel tortuosity index, angle of target vessel and target vessel angle
distal of stent. a In 20 randomly assigned cases repeated measurements by observer 1 were done (combined pre-
and post-operative CTA) in Aquarius iNtuition™, b In 20 randomly assigned cases measurements by observer 2
were done with Philips IntelliSpace and compared to repeated measurements in the post-operative CTA, c
Measu-rements by Observer 2 (combined pre- and post-operative CTAs) of 20 randomly assigned cases in Aquarius
iNtuition™ were compared to initial measurements in the same cases by observer 1. d Measurements by Observer 2
(combined pre- and post-operative CTAs) of 20 randomly assigned cases in Philips Intellispace were compared to initial measurements by observer 1 in the same cases. CTA: Computed tomography angiography, ICC: intra-class correlation coefficient, CI: 95% confidence interval, CI: Confidence interval
5
Baseline differences in anatomy are shown in table IV. Fifteen fenestrations were used for the CA (10 in group A and 5 in group B), and 65 fenestrations for the SMA (47 in group A and 18 in group B). The baseline aortic diameter at the level of the SMA was 2mm larger in group A compared to group B (P = .007). Table V shows changes within groups and difference in change between groups for all variables, comparing pre-operative and first post-operative CTA. A statistically significant change within group A was seen for the aortic diameter at the level of the CA and SMA, the angles of the CA, SMA, LRA and RRA, and clock position of the LRA and RRA. A statistically significant change within group B was seen for SMA TI and the angles of the LRA and RRA. A statistically significant difference between groups was seen in the change of the aortic diameter at SMA and RA, the change of the LRA and RRA clock position and in the SMA angle distal of the target vessel stent.
No difference between BE covered, SE bare metal or combined stents was seen in anatomic change of the LRA and RRA in TI (P = .295 and P = .734, respectively), clock position (P = .457 and P = .060, respectively) or target vessel angle relative to aortic CLL (P = .774 and P = .882, respectively). In the CA only covered stents were used. In the SMA only in one case a combination of stents was used, therefore could not be statistically analyzed. No difference was observed between used stents in target vessel angle distal of the stent for the LRA (P = .396) or RRA (P = .863).
TABLE IV: PRE-OPERATIVE ANATOMY FOR STENTED TARGET VESSELS ONLY Group A (Zenith®
Fenestrated) Group B (Fenestrated Anaconda™) Difference between
groups (P value)
Anatomic variable n Mean/
Median SD/IQR n MedianMean/ SD/IQR
Aortic diameter at SMA (mm)† 110 27 25 - 30 35 25 23 - 28 .01
Aortic diameter at RA (mm)† 110 27 25 - 33 35 28 23 - 31 .16
CA tortuosity index† 10 1.08 1.05 - 1.12 5 1.05 1.04 - 1.11 .44
CA clock position (o)† 10 14 Mar-25 5 21 14 - 25 .59
Angle of CA (o)† 10 137 124 - 147 5 123 121 - 129 .17
SMA tortuosity index† 47 1.04 1.02 - 1.07 18 1.04 1.02 - 1.11 .52
SMA clock position (o)† 47 7 -24 18 7 Feb-14 .54
Angle of SMA (o)† 47 127 115 - 136 18 120 118 - 127 .56
LRA tortuosity index† 105 1.10 1.05 - 1.18 32 1.15 1.05 - 1.24 .21
LRA clock position (o) 105 84 18 32 86 20 .72
Angle of LRA (o) 105 114 17 32 117 15 .46
RRA tortuosity index† 98 1.12 1.05 - 1.19 34 1.15 1.11 - 1.22 .02
RRA clock position (o) 98 -68 17 34 -64 15 .15
Angle of RRA (o) 98 117 16 34 117 19 .63
†Not normally distributed, presented with median and IQR. A p-value of <0.05 is considered statistically significant.
SD: Standard deviation, IQR: Interquartile range, SMA: Superior Mesenteric Artery, RA: Renal arteries, CA: Celiac artery, LRA: Left renal artery, RRA: Right renal artery.
Ch ap ter 5 90 TAB LE V : AN AT O MI C CH AN GE B EFO RE AND AFTER IMP LANT ATI O N WITHIN GR O UPS AND B ET WEEN GR O UPS Diff er en ce b et w een gr ou ps f or a na to mic al ch an ge (P va lue) <.001 <.001 .36 .84 .40 .08 .56 .61 .84 .019 .28 .70 .06 .48 .14 .003 .09 .58 † N ot n or m al ly di str ib ut ed , p res en te d w ith m edi an a nd I Q R. * S ten t a ng les a re o nl y a pp lic ab le p os t-o pera tiv ely . ‡ In o ne c as e t her e wa s a n in ab ili ty t o s ten t t he ce liac a rt er y a nd in ten tio na lly o ver sten te d w ith a c uff . Diff er en ce (Δ) i s s ho w n in m edi an an d in ter qu ar tile ra ng e (I Q R). A p-va lue of <0.05 is co nsider ed s ta tis tic al ly sig nific an t. S D: S ta nd ar d de vi at io n, I Q R: I nt er qu ar tile ra ng e, S MA: S up er io r M es en ter ic A rt er y, R A: R en al a rt er ies, CA: C eli ac a rt er y, LR A: L eft r en al a rt er y, RR A: R ig ht r en al a rt er y. G ro up B (F en es tra te d A naco nd a™) P va lue .68 .87 .72 .47 .29 .04 .70 .46 .58 .67 .001 .33 .774 .000 - - - -SD /IQ R -1 t o 2 -2 t o 2 -0.03 t o 0.13 -14 t o 4 -19 t o 1 0.00 t o 0.05 -10 t o 5 -11 t o 9 -0.07 t o 0.04 15 19 -0.05 t o 0.03 14 14 133 t o 150 167 t o 173 14 15 M ea n/M edi an Δ 1.0 † 0.0 † 0.01 † -4 † -6 † 0.02 † -1 † -4 † 0.01 † 2 -13 -0.02 † 1 -12 †148 †169 156 150 n 35 35 ‡4 ‡4 ‡4 18 18 18 32 32 32 34 34 34 4 18 32 34 G ro up A (Z eni th® F en es tra te d) P va lue <.001 <.001 .29 .29 .012 .19 .92 .02 .59 <.001 <.001 .13 <.001 <.001 - - - -SD /IQ R -4 t o -1 -4 t o -1 -0.07 t o 0.28 -5 t o 3 -17 t o -7 -0.01 t o 0.02 -6 t o 6 -11 t o 4 -0.04 t o 0.04 11 16 -0.04 t o 0.02 12 14 141 t o 167 158 t o 168 17 12 M ea n/M edi an Δ -3 † -2 † -0.02 † -4 † -14 † 0.00 † -1 † -4 † 0.00 † -4 -9 -0.01 5 -10 †157 †163 150 149 n 110 110 10 10 10 47 47 47 105 105 105 98 98 98 10 47 104 98 A na to mic va ria ble Ao rt ic di am et er a t S MA (mm) Ao rt ic di am et er a t R A (mm) CA t or tuosi ty in dex CA c lo ck p osi tio n ( o) A ng le o f CA ( o) SMA t or tuosi ty in dex SMA c lo ck p osi tio n ( o) A ng le o f S MA ( o) LR A t or tuosi ty in dex LR A c lo ck p osi tio n ( o) A ng le o f LR A ( o) RR A t or tuosi ty in dex RR A c lo ck p osi tio n ( o) A ng le o f RR A ( o) CA a ng le s ten t* SMA a ng le s ten t* LR A a ng le s ten t* RR A a ng le s ten t*
5
Discussion
The implantation of a fenestrated endograft altered the anatomy of the proximal abdominal aorta and its visceral branches, without large differences between the two different endografts. However, results suggest that the Fenestrated AnacondaTM is more conformable compared to
the Zenith® Fenestrated, as there were fewer changes in the native anatomy after placement. In both groups, anatomical changes were observed for angles in the renal arteries, but not in the mesenteric arteries. Endograft planning was done relative to the mesenteric arteries, potentially leading to a mismatch in measurements of the renal arteries. Alternatively, the position of the mesenteric arteries is more rigidly fixed to the surrounding tissues and as a consequence they may be less influenced by stenting. The sample size of stented renal arteries was higher than mesenteric arteries and results of stented mesenteric arteries could therefore be false negative.
A difference between groups was seen in change of clock position of the renal arteries. In group A, RA clock position moved anteriorly, while this was not seen in group B. During placement of the Zenith® Fenestrated the endograft was partly deployed, the target vessels were cannulated, the diameter reducing ties were released and finally the target vessels were stented. As a consequence, after release of the diameter reducing ties, the stented arteries can be pushed anteriorly. Furthermore, the main part of the Fenestrated AnacondaTM is
unres-tricted by circular stents and the aortic blood pressure can push the graft to the aortic wall. Alternatively, the circular Z-stents of the Zenith® Fenestrated prevents expansion resulting in aortic size alteration. Consequently to the expansion of the main part in the Fenestrated AnacondaTM, the distal part of the endograft could have moved upward, and thereafter might
lead to folding of the graft material.
The angle distal to the stent in the SMA, and a perpendicular movement of the SMA to the aortic CLL were more pronounced in group A. The absence of SMA angle change in group B was explained by the part of the Fenestrated AnacondaTM unrestricted by circular stents
and allowed the fenestration to move after stenting. One would expect to see change in other target vessels of group A too, but this was only seen to a limited extent in our study for the LRA.
The observation that the tortuosity index (TI) of target vessels did not differ between groups may seem to be logical, as in most cases in both groups similar BE covered stents were used to
Ch
ap
ter 5
92
accommodate the fenestrations. Nevertheless, the docking of these stents in the fenestrations is more rigid in group A, due to their position relative to the struts of the Z-stent. This may well be the reason that the clock position of the renal arteries and the angles of the SMA and CA changed significantly in group A and not in group B. A main body that is unrestricted by struts has greater adaptation to native anatomy, and may therefore have less risk of strain on the stents and consequently less chance of stent fractures. Nevertheless, both devices caused an equal straightening of the renal arteries, and change in the SMA TI was only observed in group B, indicating that the difference between the two endografts in conformational change to the target vessels is probably limited.
The conformability of an endograft is an important factor for outcome prediction after EVAR, especially in those with more severe aortic and iliac angulations and iliac tortuosity indices. 14-17 As shown in this study, there are anatomic changes of the stented visceral arteries after
im-plantation of a fenestrated endograft. After FEVAR an increased risk of renal and neurologic injury has been described, but the exact relation to endograft conformability is yet unknown.25
One of the considerations in choosing an endograft should include the angles and tortuosity indices of the aorta, the iliac arteries and visceral arteries, but also the ability of the endograft to conform to the patient’s anatomy. After implantation a comparative analysis should be done between the pre-operative and post-operative CTA, to predict complications related to the endograft conformability. In this study the number of adverse events was too small to do a reliable regression analysis. Therefore we were not able to relate the conformability to com-plications and reinterventions.
Treatment with the fenestrated endograft for complex AAA was primarily done in patients unfit for open surgery.26 Over time fenestrated EVAR was also chosen for patients with less
co- morbidities. The time gap in our cohort between introduction of the Zenith® Fenestrated and the Fenestrated AnacondaTM nearly reached 10 years and may explain the difference in
the co-morbidities.
Customizing the endograft is time-consuming, and led to a time-delay of a few months between baseline CTA and treatment. The exact time between industry contact and final approval was not known, but possibly this took more time for the Zenith® Fenestrated, especially in the early period, potentially explaining the difference in time between CTA and treatment between both groups. With regard to the differences between both groups in timing of the first follow-up: in one center, standard clinical practice means that the first
postopera-5
tive CTA is at around four weeks post- implantation; in the other (having implanted solely Fenestrated Anaconda), the first postoperative CTA is performed before discharge of the patient after treatment.
The ICC was high for nearly all variables, both between two repeated measurements by the same observer and between the two independent observers. The measurements for different variables were performed in the same way. All pre- and post-operative aortic diameters, target vessel TI, target vessel clock position and target vessel angles relative to the aortic CLL were combined (Table III). It would be ideal to analyze variables separately, but the sample sizes were too small to have a reliable agreement for individual variables. Consensus in anatomical measurement techniques, making them reproducible, might eventually help prevent errors in measurements and sequentially clinical complications.27
Pressure changes during the cardiac cycle can influence anatomic configurations. In this study the CTA does not account for the cardiac cycle. Dynamic CTA allows scanning at specific moments during the cardiac cycle.28 Measuring these changes is too complex and too labor
intensive. Computer algorithms might help in future research to show continuing movement of the endograft.29
The influence of small anatomic changes on outcome is still unclear and the clinical consequence and cut-off value for clinical relevance need to be examined. A large prospective randomized trial with both endografts would help to understand which patient benefits the most from which fenestrated EVAR, but is unrealistic due to the large series needed to show very small changes in anatomy and relations between different variables. Until the clinical relevance of these changes has been shown, our study shows good conformability of both designs.
Conclusion
The implantation of a custom-made fenestrated endograft for complex abdominal aortic AAAs evidently altered patients’ anatomy, but there is no difference between current endografts for target vessel tortuosity index. This study suggests the unconstrained proximal body of the Fenestrated AnacondaTM endograft may have better conformability compared to the circular
Ch
ap
ter 5
94
1.Prinssen M, Verhoeven EL, Buth J C, P.W., van Sambeek M.R., Balm R, Buskens E, et al. A randomized trial comparing conventional and endovascular repair of abdominal aortic aneurysms. N Engl J Med. 2004;351:1607-18.
2.Bruin JLD, Baas AF, Buth J, Prinssen M, Verhoeven EL, Cuypers PW, et al. Long-term outcome of open or endovascular repair of abdominal aortic aneurysm. N Engl J Med. 2010;362:1881-9.
3.Ketelsen D, Thomas C, Schmehl J, Konig CW, Syha R, Rittig K, et al. Endovascular aneurysm repair of abdominal aortic aneurysms: standards, technical options and advanced indications. Rofo. 2014;186:337-47.
4.Jonker LT, de Niet A, Reijnen MMPJ, Tielliu IFJ, Zeebregts CJ. Mid- and long-term outcome of currently available endografts for the treatment of infrarenal abdominal aortic aneurysm. Surg Technol Int. 2018;33:239-50.
5.de Niet A, Reijnen MM, Tielliu IF, Lardenoije JW, Zeebregts CJ. Fenestrated endografts for complex abdominal aortic aneurysm repair. Surg Technol Int. 2016;29:220-30.
6.Wyers MC, Fillinger MF, Schermerhorn ML, Powell RJ, Rzucidlo EM, Walsh DB, et al. Endovascular repair of abdominal aortic aneurysm without preoperative arteriography. J Vasc Surg. 2003;38:730-8.
7.Sobocinski J, Chenorhokian H, Maurel B, Midulla M, Hertault A, Le Roux M, et al. The benefits of EVAR planning using a 3D workstation. Eur J Vasc Endovasc Surg. 2013;46:418-23.
8.Sprouse LR, Meier GH, Parent FN, DeMasi RJ, Stokes GK, LeSar CJ, et al. Is three-dimensional computed tomography reconstruction justified before endovascular aortic aneurysm repair? J Vasc Surg. 2004;40:443-7. 9.Macia I, de Blas M, Legarreta JH, Kabongo L, Hernandez O, Egana JM, et al. Standard and fenestrated endograft sizing in EVAR planning: Description and validation of a semi-automated 3D software. Comput Med Imaging Graph. 2016;50:9-23.
10.Verhoeven EL, Katsargyris A, Oikonomou K, Kouvelos G, Renner H, Ritter W. Fenestrated endovascular aortic aneurysm repair as a first line treatment option to treat Short necked, juxtarenal, and suprarenal aneurysms. Eur J Vasc Endovasc Surg. 2016;51:775-81.
11.Verhoeven EL, Katsargyris A, Bekkema F, Oikonomou K, Zeebregts CJ, Ritter W, et al. Editor's choice - Ten-year experience with endovascular repair of thoracoabdominal aortic aneurysms: Results from
166 consecutive patients. Eur J Vasc Endovasc Surg. 2015;49:524-31.
12.Bungay PM, Burfitt N, Sritharan K, Muir L, Khan SL, De Nunzio MC, et al. Initial experience with a new fenestrated stent graft. J Vasc Surg. 2011;54:1832-8. 13.Blankensteijn LL, Dijkstra ML, Tielliu IFJ, Reijnen MM, Zeebregts CJ. Midterm results of the fenestrated Anaconda endograft for short-neck infrarenal and juxtarenal abdominal aortic aneurysm repair. J Vasc Surg. 2017;65:303-10.
14.Lee K, Leci E, Forbes T, Dubois L, DeRose G, Power A. Endograft conformability and aortoiliac tortuosity in endovascular abdominal aortic aneurysm repair. J Endovasc Ther. 2014;21:728-34.
15.Parra JR, Ayerdi J, McLafferty R, Gruneiro L, Ramsey D, Solis M, et al. Conformational changes associated with proximal seal zone failure in abdominal aortic endografts. J Vasc Surg. 2003;37:106-11.
16.Iwakoshi S, Nakai T, Ichihashi S, Inoue T, Sakaguchi S, Hirose T, et al. Conformability and efficacy of the Zenith Spiral Z leg compared with the Zenith Flex leg in endovascular aortic aneurysm repair. Ann Vasc Surg. 2019[Epub ahead of print].
17.Della Schiava N, Arsicot M, Boudjelit T, Feugier P, Lermusiaux P, Millon A. Conformability of GORE Excluder iliac branch endoprosthesis and COOK Zenith bifurcated iliac side branched iliac stent grafts. Ann Vasc Surg. 2016;36:139-44.
18.Ullery BW, Suh GY, Lee JT, Liu B, Stineman R, Dalman RL, et al. Comparative geometric analysis of renal artery anatomy before and after fenestrated or snorkel/chimney endovascular aneurysm repair. J Vasc Surg. 2016;63:922-9.
19.Oshin OA, How TV, Brennan JA, Fisher RK, McWilliams RG, Vallabhaneni SR. Magnitude of the forces acting on target vessel stents as a result of a mismatch between native aortic anatomy and fenestrated stent-grafts. J of Endovasc Ther. 2011;18:569-75. 20.Riley R, Holman C, Fletcher D. Inter-rater reliability of the ASA physical status classification in a sample of anaesthetists in Western Australia. Anaesth Intensive Care. 2014;42:614-8.
21.Chaikof EL, Fillinger MF, Matsumura JS, Rutherford RB, White GH, Blankensteijn JD, et al. Identifying and grading factors that modify the outcome of endovascular aortic aneurysm repair. J Vasc Surg. 2002;35:1061-6. 22.Chaikof EL, Blankensteijn JD, Harris PL, White GH, Zarins CK, Bernhard VM, et al. Reporting standards References
5
for endovascular aortic aneurysm repair. J Vasc Surg. 2002;35:1048-60.
23.Wolf YG, Tillich M, Lee WA, Rubin GD, Fogarty TJ, Zarins CK. Impact of aortoiliac tortuosity on endovascular repair of abdominal aortic aneurysms: evaluation of 3D computer-based assessment. J Vasc Surg. 2001;34:594-9.
24.Koo TK, Li MY. A guideline of selecting and reporting intraclass correlation coefficients for reliability research. J Chiropr Med. 2016;15:155-63.
25.Locham S, Faateh M, Dhaliwal J, Nejim B, Dakour-Aridi H, Malas MB. Outcomes and cost of fenestrated versus standard endovascular repair of intact abdominal aortic aneurysm in the United States. J Vasc Surg. 2018;69:1036-44.
26.Rao R, Lane TR, Franklin IJ, Davies AH. Open repair versus fenestrated endovascular aneurysm repair of juxtarenal aneurysms. J Vasc Surg. 2015;61:242-55. 27.Oshin OA, England A, McWilliams RG, Brennan JA, Fisher RK, Vallabhaneni SR. Intra- and interobserver variability of target vessel measurement for fenestrated endovascular aneurysm repair. J Endovasc Ther. 2010;17:402-7.
28.Klein A, Oostveen LJ, Greuter MJ, Hoogeveen Y, Kool LJ, Slump CH, et al. Detectability of motions in AAA with ECG-gated CTA: a quantitative study. Med Phys. 2009;36:4616-24.
29.Klein A, van der Vliet JA, Oostveen LJ, Hoogeveen Y, Kool LJ, Renema WK, et al. Automatic segmentation of the wire frame of stent grafts from CT data. Med Image Anal. 2012;16:127-39.
