論文ID: CJ-17-0786
Background: Aneurysm expansion, and consequent endoleaks, after endovascular aneurysm repair (EVAR) is a major problem. Accurate prediction of aneurysm expansion is demanding for surgeons and remains difficult.
Methods and Results: We retrospectively analyzed 157 cases of EVAR for abdominal aortic aneurysm (AAA) using a bifurcated main-body stent-graft. There were 62 cases of aneurysm shrinkage after EVAR, 63 cases of stable aneurysm, and 32 cases of aneurysm expansion. Type I endoleaks were significantly increased in the aneurysm expansion group (EXP) compared with the stable (STB) and shrinkage (SHR) groups (EXP: 15.6% vs. STB: 4.8% vs. SHR: 0%, P=0.005). Type II endoleaks were also significantly increased in EXP (EXP: 65.6% vs. STB: 36.5% vs. SHR: 6.5%, P<0.001). Aneurysm wall enhancement (AWE) on imaging, however, was significantly decreased in the EXP group (EXP: 18.8% vs. STB: 23.8% vs. SHR: 53.2%, P<0.001). In multivariate analysis, the occurrence of type II endoleaks significantly decreased (P<0.001) and that of AWE significantly increased the likelihood of aneurysm shrinkage (P=0.032).
Conclusions: AWE following EVAR may be associated with aneurysm shrinkage.
Endovascular aneurysm repair (EVAR) for abdominal aortic aneurysm (AAA) is less invasive and associated with a significant reduction in operative mortality and morbidity compared with conventional open surgery.1–3 However, aneurysm expansion after EVAR is a major problem because endoleaks are significantly associated with it, but accurate prediction of aneurysm expansion is demanding for surgeons and remains difficult.4–6
Aneurysm wall enhancement (AWE) in a cerebral aneurysm on magnetic resonance imaging (MRI) is a known risk factor for rupture of an untreated intracranial aneurysm.7 Moreover, AWE in a cerebral aneurysm after endovascular treatment is a common, asymptomatic, and self-limiting phenomenon.8,9 AWE in the delayed phase on computed tomography (CT) is reported to be associated with a larger diameter of AAA.10 Because there are no other reports of AWE in AAA, it is unknown whether AWE is associated with aneurysm prognosis after EVAR, so we retrospectively assessed the relationship between AWE and AAA prognosis.
We performed a retrospective review of patients who underwent EVAR with a Zenith stent-graft (Cook, Bloomington, IN, USA), Excluder stent-graft (W. L. Gore & Associates, Flagstaff, AZ, USA), Endurant stent-graft (Medtronic, Minneapolis, MN, USA), Aorfix stent-graft (Lombard Medical Technologies, Didcot, UK), or Powerlink and AFX stent-graft (Endologix, Irvine, CA, USA) at the Jikei University Kashiwa Hospital between August 2011 and July 2016. The Jikei University Kashiwa Hospital Institutional Review Board approved this study with informed consent waived because it was a retrospective observational study using a de-identified database (28-192 (8435)).
The patients included in this study underwent EVAR for AAA using a bifurcated main-body stent-graft. Patients who did not undergo contrast-enhanced CT and/or a short follow-up less than 6 months after EVAR were excluded. Primary endpoints were aneurysm expansion and shrinkage after EVAR, which were evaluated by CT scan.
PatientsPatients with an AAA that had a minimum external diameter of 5.5 cm (short axis) or saccular morphology were deemed eligible for EVAR. According to the hospital’s strategy for EVAR, any patient over 75 years of age, with severe frailty or a hostile abdomen undergoes EVAR first, unless the patient’s anatomical features indicate low Instructions for Use (IFU). Patients without these risks (i.e., advanced age, frailty, or hostile abdomen) undergo open repair instead. The exclusion criteria for elective EVAR included acute myocardial or cerebral infarction within 3 months before surgery and/or recent symptomatic blue toe syndrome. All patients were classified according to the American Society of Anesthesiologists system (grades 1–4). EVAR in patients with severe or very severe chronic obstructive pulmonary disease with a forced expiratory volume in 1 s (FEV1) of 50% [stage III or IV according to the global initiative for chronic obstructive lung disease (GOLD) classification] was performed under local or epidural anesthesia. Patients with end-stage renal disease were considered eligible for EVAR.
Data AnalysisAll patients included in the present study underwent preoperative and postoperative CT scanning using a 64-detector row CT scanner (Aquilion 64, Toshiba Co., Tokyo, Japan). A contrast-enhanced CT scan was performed 120 s after 30-s intravenous administration of 600 mg/kg contrast agent. The CT images with a slice thickness of 1 mm were transferred to a 3D image analysis workstation (Synapse Vincent, v4.4, Fujifilm Co., Tokyo, Japan) for evaluation. The mean CT value of the aneurysm wall was calculated from 5 random points in a slice with maximal aneurysm diameter. For CT measurements, values of 1 pixel (0.195×0.195 mm) of the AAA wall on unenhanced and enhanced images were measured. Sakuta et al have previously reported a technique for measuring AWE,10 which we replicated faithfully. The definition of AWE is an increase of >20 Hounsfield units (HU) in mean CT values. When there was calcification of the AAA wall, AWE was measured using a noncalcified portion of the wall (Figures 1,2). For AWE assessment, 2 of the authors, with 7 and 29 years of experience, respectively, in interpreting vascular CT images, independently compared images. The results of the 2 observers were compared to assess the interobserver variability for all cases. We diagnosed AWE on the contrast CT scan performed 1 month after EVAR, and compared the measured the aneurysm size on the preoperative non-contrast CT scan with the most recent non-contrast CT scan. We defined aneurysm expansion as >5 mm of expansion between the preoperative diameter and the most recent diameter, and aneurysm shrinkage as >5 mm of shrinkage. The author who judged aneurysm expansion and shrinkage was blinded to the clinical course of the patients. Preoperative CT scans were performed within 1 month before EVAR. Aneurysm wall thrombus was defined as a >5-mm-thick thrombus at the point of maximal thickness measured in a slice of the preoperative contrast CT scan with maximal aneurysm diameter by axial imaging.
(A) Preoperative enhanced CT scan showing AAA. (B) Postoperative enhanced CT scan showing AAA with AWE. (C) 12 months after EVAR, the aneurysm has shrunk. AAA, aortic abdominal aneurysm; AWE, aneurysm wall enhancement; EVAR, endovascular aneurysm repair.
(A) Preoperative enhanced CT scan showing AAA. (B) Postoperative enhanced CT scan showing AAA without AWE. (C) 12 months after EVAR, the aneurysm has expanded. Abbreviations as in Figure 1.
Interobserver variability was analyzed using the Cohen kappa coefficient test. Patients’ characteristics and outcomes were analyzed using the chi-square test, analysis of variance (ANOVA), and the Kruskal-Wallis test. Multivariate analyses were performed to identify patient- and procedure-related risk factors for aneurysm expansion and shrinkage after EVAR. ANOVA was used when continuous data exhibited a normal distribution. For comparison of nonparametric data, the Kruskal-Wallis test was used. Categorical data were evaluated using the chi-square or Fisher’s exact test. Subsequently, only factors that achieved a level of P<0.05 after univariate testing were entered in the logistic regression analysis. We compared AWE and these variables using the chi-square test. All statistical analyses were performed using Stata/IC (STATA Statistical Software, version 14.0; Stata Corp., College Station, TX, USA). Two-sided probability (P) values <0.05 were considered significant.
The median CT value in AWE cases was 34 HU (23–82 HU). At 6 months after EVAR, we evaluated 130 of the cases using enhanced CT scans. At that time, AWE had disappeared in 13.0% of cases in the AWE-positive group (6/46), but had newly appeared in 1.2% of cases in the AWE-negative group (1/84).
Interobserver Variability for Assessment of AWEThe kappa coefficiency for AWE was 0.887 (157 CT scans at 1 month after EVAR). Therefore, we considered this measurement reliable.
Baseline Characteristics and OutcomesIn the study period, a total of 219 AAAs were treated with EVAR using a bifurcated main-body stent-graft, and 62 patients were excluded: 42 patients in whom contrast-enhanced CT scan could not be performed within 1 month after EVAR and 20 patients with short follow-up period (within 6 months). Of the 157 identified patients who underwent EVAR for AAA, there were 127 men (80.9%) and 30 women (19.1%), aged 45–93 years (median: 75 years). The follow-up period after surgery was 6–59 months (median: 23.6 months). There were 62 cases of aneurysm shrinkage after EVAR (SHR), 63 cases of stable aneurysm (STB), and 32 cases of aneurysm expansion (EXP). There were no significant differences in the baseline characteristics of patients who underwent EVAR with respect to preoperative factors, anatomical factors, or operative details, except for preoperative stroke, aneurysm thrombus, proximal neck angulation and AWE before EVAR. There were significantly more preoperative strokes in the STB group than in the SHR or EXP group (EXP: 3.1% vs. STB: 15.9% vs. SHR: 4.8%, P=0.042). There were significantly fewer aneurysm wall thrombus in the EXP group than in the SHR or STB group (EXP: 59.4% vs. STB: 82.5% vs. SHR: 79.0%, P=0.035). The number of type I endoleaks was significantly increased in the EXP group compared with the STB and SHR groups (EXP: 15.6% vs. STB: 4.8% vs. SHR: 0%, P=0.005). That of type II endoleaks was significantly increased in the EXP group (EXP: 65.6% vs. STB: 36.5% vs. SHR: 6.5%, P<0.001). The occurrence of AWE, however, was significantly decreased in EXP (EXP: 18.8% vs. STB: 23.8% vs. SHR: 53.2%, P<0.001) (Table 1).
Variables | Total (n=157) | SHR (n=62) | STB (n=63) | EXP (n=32) | P value |
---|---|---|---|---|---|
n (%) or median (range) | |||||
Follow-up period (months) | 23.6 (6–59) | 23.6 (6–59) | 13.1 (6–51) | 36 (9–59) | <0.001** |
Preoperative factors | |||||
Age (years) | 75 (45–93) | 74 (45–87) | 75 (57–93) | 78 (57–89) | 0.155* |
Male | 127 (80.9) | 55 (88.7) | 48 (76.2) | 24 (75.0) | 0.131† |
Diabetes | 17 (10.8) | 7 (11.3) | 7 (11.1) | 3 (9.4) | 0.953† |
Hypertension | 113 (72.0) | 42 (67.7) | 48 (76.2) | 23 (71.9) | 0.575† |
CAD | 30 (19.1) | 15 (24.2) | 12 (19.5) | 3 (9.4) | 0.223† |
Dyslipidemia | 66 (42.0) | 23 (37.1) | 32 (20.8) | 11 (34.4) | 0.227† |
Stroke | 14 (8.9) | 3 (4.8) | 10 (15.9) | 1 (3.1) | 0.042† |
Anatomic factors | |||||
Aneurysm size (mm) | 50 (25–91) | 50 (25–91) | 48 (31–77) | 50 (34–85) | 0.321** |
Aneurysm wall thrombus | 120 (76.4) | 49 (79.0) | 52 (82.5) | 19 (59.4) | 0.035† |
Saccular aneurysm | 46 (29.3) | 17 (27.4) | 22 (34.9) | 7 (21.9) | 0.383† |
Aneurysm calcification | 0.133† | ||||
Non | 49 (31.2) | 25 (40.3) | 17 (27.0) | 7 (21.9) | |
Partial | 88 (56.1) | 28 (45.2) | 37 (58.7) | 23 (71.9) | |
Circumferential | 20 (12.7) | 9 (14.5) | 9 (14.3) | 2 (6.3) | |
Neck length (mm) | 26 (6–81) | 26 (8–77) | 28 (10–81) | 24 (6–51) | 0.155** |
Neck diameter (mm) | 22 (15–38) | 23 (16–30) | 22 (15–33) | 23 (16–38) | 0.891** |
Neck angulation (degrees) | 50 (3–138) | 49 (9–126) | 49 (14–121) | 71 (21–138) | 0.003** |
Neck calcification | 0.780† | ||||
Non | 49 (31.2) | 21 (33.9) | 21 (33.3) | 7 (21.9) | |
Partial | 96 (61.2) | 37 (59.7) | 37 (58.7) | 22 (68.8) | |
Circumferential | 12 (7.6) | 4 (6.5) | 5 (7.9) | 3 (9.4) | |
Neck thrombus | 51 (32.5) | 18 (29.0) | 20 (31.8) | 13 (40.6) | 0.517† |
Preoperative AWE | 27 (17.2) | 15 (24.2) | 11 (17.5) | 1 (3.1) | 0.037† |
Operative details | |||||
Device | 0.387† | ||||
Zenith Felx | 9 (5.7) | 3 (4.8) | 3 (4.8) | 3 (9.4) | |
Excluder | 64 (40.8) | 23 (37.1) | 23 (36.5) | 18 (56.3) | |
Endurant | 58 (36.9) | 23 (37.1) | 25 (39.7) | 10 (31.3) | |
Powerlink/AFX | 14 (8.9) | 7 (11.3) | 7 (11.1) | 0 | |
Aorfix | 12 (7.6) | 6 (9.7) | 5 (7.9) | 1 (3.1) | |
Surgical duration (min) | 145 (75–305) | 145.5 (87–305) | 135 (75–270) | 152.5 (95–235) | 0.722** |
Blood loss (mL) | 100 (10–950) | 100 (20–850) | 100 (10–950) | 100 (30–600) | 0.973** |
Fluoroscopy time (min) | 28 (4–97) | 28 (11–55) | 26 (11–97) | 30.5 (4–78) | 0.721** |
Contrast enema (mL) | 145 (56–420) | 140 (56–316) | 146 (70–237) | 152.5 (68–235) | 0.933** |
Postoperative details (1 month after EVAR) | |||||
Endoleak | |||||
Type I | 8 (5.1) | 0 | 3 (4.8) | 5 (15.6) | 0.005† |
Type II | 48 (30.6) | 4 (6.5) | 23 (36.5) | 21 (65.6) | <0.001† |
Type III | 1 (0.6) | 1 (1.5) | 0 | 0 | 0.463† |
AWE | 54 (34.4) | 33 (53.2) | 15 (23.8) | 6 (18.8) | <0.001† |
†Chi-square test; *ANOVA; **Kruskal-Wallis test. ANOVA, analysis of variance; AWE, aortic wall enhancement; CAD, coronary artery disease; EVAR, endovascular aneurysm repair; EXP, expansion of aneurysm size; SHR, shrinkage of aneurysm size; STB, stable aneurysm size.
Using the logistic regression model, the variables [stroke, neck angulation, aneurysm wall thrombus, types I and II endoleaks, and AWE (pre and post EVAR)] were selected. In the multivariate analysis, the presence of neck angulation and types I and II endoleaks significantly increased the risk of aneurysm expansion [neck angulation: odds ratio [OR]= 1.027 (1.008–1.047), P=0.006; type I endoleak: OR=22.91 (3.227–162.6), P=0.002; type II endoleak: OR=14.52 (4.622–45.63), P<0.001]. On the other hand, the presence of type II endoleak significantly decreased and that of AWE significantly increased the likelihood of aneurysm shrinkage [type II endoleak: OR=0.078 (0.025–0.244), P<0.001; AWE: OR=2.666 (1.090–6.524), P=0.032] (Table 2).
Variables | OR (95% CI) | P value |
---|---|---|
Aneurysm expansion | ||
Stroke | 0.160 (0.016–1.653) | 0.124 |
Neck angulation | 1.027 (1.008–1.047) | 0.006 |
Aneurysm wall thrombus | 0.818 (0.285–2.429) | 0.717 |
Type I endoleak | 22.91 (3.227–162.6) | 0.002 |
Type II endoleak | 14.52 (4.622–45.63) | <0.001 |
Preoperative AWE | 0.160 (0.016–1.580) | 0.117 |
Postoperative AWE | 1.053 (0.285–3.890) | 0.938 |
Aneurysm shrinkage | ||
Stroke | 0.459 (0.099–2.124) | 0.319 |
Neck angulation | 0.995 (0.980–1.010) | 0.500 |
Aneurysm wall thrombus | 0.617 (0.223–1.704) | 0.351 |
Type I endoleak | – | |
Type II endoleak | 0.078 (0.025–0.244) | <0.001 |
Preoperative AWE | 1.060 (0.337–3.335) | 0.920 |
Postoperative AWE | 2.666 (1.090–6.524) | 0.032 |
OR, odds ratio. Other abbreviations as in Table 1.
We compared AWE and the other variables of aneurysm wall thrombus and types I and II endoleaks. AWE significantly decreased the risk of type II endoleak (38.8% vs. 14.8%, P=0.002) (Table 3).
Variables | Total (n=157) | AWE | P value | |
---|---|---|---|---|
Negative (n=103) | Positive (n=54) | |||
Aneurysm wall thrombus | 120 (76.4%) | 75 (72.8%) | 45 (83.3%) | 0.140† |
Type I endoleak | 8 (5.1%) | 7 (6.8%) | 1 (1.9%) | 0.181† |
Type II endoleak | 48 (30.6%) | 40 (38.8%) | 8 (14.8%) | 0.002† |
†Chi-square test. AWE, aneurysm wall enhancement.
AAA is a common disease of increasing prevalence, particularly in elderly men.2 As the size of an aneurysm increases, so does the probability of rupture. The performance of EVAR has recently increased around the world because of its lower invasiveness and reduction in operative mortality and morbidity.1–3 However, aneurysm expansion after EVAR is a major problem with this procedure. Although endoleaks and other factors have been reported as risk factors associated with aneurysm expansion, the accurate prediction of aneurysm expansion and shrinkage remains difficult.4–6 Types I and III endoleaks are definitely associated with sac growth,11,12 but the clinical effect of type II endoleak is not well known and remains controversial.13–16 Type II endoleak occurs after EVAR in 6–45% of patients, with spontaneous resolution occurring after 6 months in 0–80% of patients.12–17 Some authors have concluded that type II endoleak is associated with aneurysm growth, secondary intervention, and rupture.11,13,14,18–20 Others, however, report a benign clinical course of type II endoleaks.16,21 In our study, type II endoleaks were significantly associated with sac growth and reduction.
Jalalzadeh and colleagues reported that findings of inflammation on fluorodeoxyglucose positron emission tomography (FDG-PET)-CT of the aortic aneurysm wall could be used as predictive biomarkers of growth or rupture.22 The mechanism of 18F-FDG uptake and aneurysm rupture is that chronic inflammation with proteolytic degradation of the aortic wall is the principal cause of aortic wall weakening and aneurysm growth.
Edjlali et al reported that AWE in cerebral aneurysm was an independent risk factor with unstable status.7 In their report, mural artery contrast uptake was indicated in vessel wall inflammation such as vasculitis, which is linked to vasa vasorum in intracranial vessels, a unique structural feature that does not usually exist in intracranial arteries. The surrounding cerebrospinal fluid would account for the usual lack of intracranial vasa vasorum, but in specific situations, such as aging, atherosclerosis, hypertension, or other cardiovascular risk factors, vasa vasorum develops around intracranial vessels.23 Moreover, it is known that there is more vasa vasorum in larger intracranial aneurysms. Intramural bleeding and occlusion of vasa vasorum is thought to induce inflammation and release growth factors that stimulate proliferation of the vessel wall. It has been reported that AWE after endovascular treatment is a common phenomenon, especially in large intracranial aneurysms.8,9 In untreated intracranial aneurysms, AWE is thought to be a risk factor for unstable aneurysm; after endovascular treatment AWE is reported to be an asymptomatic and self-limiting phenomenon.
There is only 1 report on the relationship between AWE and untreated AAA. Sakuta et al reported that AWE in patients was associated with larger AAA, higher C-reactive protein levels, thicker aneurysm wall, and more severe atheroma than in patients without AWE.10 They concluded that chronic vascular inflammation plays an important role in aneurysm growth.
Before this study, we predicted that AWE would be associated with sac expansion and the absence of sac reduction because AWE is thought to be caused by inflammation of the aneurysm wall. However, the results were opposite: the presence of AWE decreased the risk of type II endoleak and increased the occurrence of sac shrinkage. The mechanism of this is unknown. The stent-graft blocks blood flow from the intravascular space to the aneurysm wall, and blood flow of the aneurysm wall depends solely on vasa vasorum.24,25 Therefore, in type II endoleaks the aneurysm wall receives blood flow directly from the type II endoleak as well as from the vasa vasorum. In the absence of a type II endoleak, however, the aneurysm wall receives blood flow through the vasa vasorum only. We presume that blood flow from the vasa vasorum is increased in patients without type II endoleaks, and that this is the pathogenesis of AWE. Tanaka et al reported that hypoperfusion of the vasa vasorum develops the sac diameter of AAA.26,27 We hypothesized that AWE would represent hyperperfusion of the vasa vasorum and development of sac shrinkage.
Study LimitationsFirst, in this study, 32 of 157 patients (20.4%) had aneurysm expansion, and the incidence of type I endoleaks was 5.1%. The cause of these high incidences is thought to be our inclusion of patients treated outside the IFU as well as those with para-renal abdominal aneurysm. Second, this was a single-center study, so the results may include facility and selection bias. Third, because endoleaks change dynamically after EVAR, it is sometimes difficult to know when patients have them. Several patients were initially thought to have type II endoleaks, but were later found to have type I endoleaks. We diagnosed AWE and endoleak solely based on CT scan findings 1 month after EVAR. Fourth, this study lacked histopathological studies of our observed findings. All hypotheses were based on pathological reports from the literature. Fifth, in our analysis of the follow-up period, there was a significant difference between aneurysm prognoses. We excluded cases of patients with follow-up periods shorter than 6 months. In fact, the median follow-up period in the sac-stable group was longer than 1 year. It cannot be denied, therefore, that the difference in follow-up duration might be associated with differences in aneurysm prognosis between the AWE-positive and AWE-negative groups. Finally, our study was retrospective in nature and had a relatively small number of patients. A prospective, blinded clinical trial is required to establish the mechanism and physiology of AWE and aneurysm prognosis.
Aneurysm wall enhancement after EVAR may be associated with aneurysm shrinkage.
T.O. received advisory fees from W.L. Gore and Boston Scientific Corporation. The other authors have no conflicts of interest or financial ties to disclose.
None.