Article ID: CJ-15-0850
Background: Clinically distinct differences exist between East Asian and Caucasian subjects, but data for type 2 endoleak (T2EL) are limited in the East Asian population. The aim of this study was to analyze the characteristics of East Asian T2EL using a Japanese multicenter database.
Methods and Results: Retrospective analysis of 832 endovascular aneurysm repairs performed from 2008 to 2014 were conducted. T2EL was observed in 234 cases (28.1%), and in 32 cases (3.8%) it led to sac expansion >5 mm caused by isolated T2EL (median follow-up, 35.6 months). On univariate and multivariate analysis, non-smoker status (odds ratio [OR], 2.216; P<0.001), Excluder stent graft (OR, 2.027; P<0.001), and T2EL at final angiogram (OR, 2.080; P<0.001) were risk factors for T2EL. On multivariate analysis for isolated T2EL with sac expansion, only non-smoker status remained (OR, 2.671; P<0.001). Other than T1EL, isolated T2EL was the most significant risk factor for sac expansion (OR, 18.486; P<0.001). Furthermore, out of 11 transarterial embolization procedures initiated, 4 led to rupture during follow-up.
Conclusions: East Asian T2EL had a strong relationship with non-smoker status. Also, T2EL was a significant risk factor for sac expansion, which sometimes led to rupture even after intervention. Along with the high prevalence of T2EL observed, East Asian T2EL may not always be benign.
Endovascular aneurysm repair (EVAR) has become the gold standard for the treatment of abdominal aortic aneurysms (AAA) owing to its low invasiveness.1,2 Compared with conventional open surgical repair, however, EVAR is associated with high rates of late secondary intervention.1–5 One of the leading causes for secondary intervention after EVAR is the type 2 endoleak (T2EL). T2EL is defined as retrograde flow to the remaining aneurysm sac through aortic side branches.3 The reported incidence of T2EL varies between reports, usually between 10 and 20%.6–8 Although many reports have been published on late aneurysm sac growth caused by T2EL, interestingly, debate continues as to whether T2EL leads to worse outcome9–11 or whether it does not.11–15 In addition, no consensus has been reached on the surveillance protocol and timing of treatment for T2EL, and the effects of treatment for T2EL remain unclear.12,15–20
From previous reports, it has become clear that many clinically distinct differences exist between East Asian and Caucasian patients, including demographics,21 anatomy,22,23 and coagulability.24,25 Data for T2EL, however, which seems to be greatly influenced by these factors, are derived mainly from Caucasian countries, and data for East Asian populations are limited.6 We hypothesized that the lower coagulability observed in East Asian patients would lead to a higher incidence of T2EL and that the risk factors for T2EL might differ from those observed in previous Caucasian databases. The aim of this study was to analyze the characteristics and risk factors of East Asian T2EL using a Japanese multicenter database.
We retrospectively analyzed data from the prospectively maintained registry for patients who underwent EVAR at Keio University Hospital (Tokyo, Japan), Kawasaki Municipal Hospital (Kawasaki, Japan), Saisekai Yokohamashi Tobu Hospital (Kawasaki, Japan), Hiratsuka City Hospital (Hiratsuka, Japan), Tokyo Dental College Ichikawa General Hospital (Ichikawa, Japan), Saitama City Hospital (Saitama, Japan), Mito Red Cross Hospital (Mito, Japan), and Tokyo Medical Center (Tokyo, Japan) from 2007 to 2014. More than 50 categories of patient demographics, comorbidity, medication, anatomy, device used, perioperative outcome, and long-term outcome (including aneurysm size, types of endoleak, and secondary intervention) were recorded in the prospectively maintained registry. Exclusion criteria included isolated iliac artery aneurysm, EVAR that necessitated intraoperative open conversion, emergency EVAR for ruptured AAA, or when a category had missing data. The study protocol was reviewed and approved by the institutional review board at each participating hospital.
Each hospital had at least 1 dedicated vascular surgeon performing EVAR, usually under general anesthesia. Stent grafts used were the Excluder (W.L. Gore and Associates, Flagstaff, AZ, USA), Zenith (Cook, Bloomington, IN, USA), Powerlink (Endologix, Irvine, CA, USA), and Endurant (Medtronic, Santa Rosa, CA, USA) models, selected by physician preference.
After EVAR, patients underwent computed tomography (CT) with contrast medium at 1 month, 6 months, and annually thereafter as follow-up surveillance. Multiphasic CT angiography of 1.00–3.00-mm cuts (both early and late phase) was obtained for imaging analysis. If the patient had an allergy to the contrast medium or chronic renal insufficiency, duplex ultrasound with plain CT or magnetic resonance imaging were done. Endoleaks were diagnosed according to the Society for Vascular Surgery practice guidelines.3
Treatment for persisting T2EL was considered for patients with sac expansion >5 mm between 2 imaging studies, and all reintervention was carried out according to the discretion of the attending surgeon and with patient consent. Transarterial embolization using a triple coaxial system (1.9-Fr non-tapered microcatheter passed through a 2.7-Fr microcatheter advanced coaxially through a 4-Fr catheter)26 was the initial approach for treating all cases of T2EL. When the aneurysm sac was unapproachable through arterial access, direct sac puncture under ultrasound or CT guidance was performed.
Statistical AnalysisFor continuous measures, variables are given as mean±SD; for the categoric factors, frequencies and percentages are used. For the univariate analysis of risk factors for T2EL, chi-squared test was used for the categorical factors and Welch’s unpaired t-test for the continuous measures. Variables with P<0.10 were introduced into the logistic regression analysis for multivariate analysis. Results are expressed as odds ratio (OR) and corresponding 95% confidence interval (CI). P<0.05 was considered as significant. All analysis was conducted using SPSS 20.0 (IBM, Armonk, NY, USA).
During the study period, 1,003 EVAR procedures were performed at 8 participating institutes. After excluding patients with isolated iliac artery aneurysm, EVAR that necessitated intraoperative open conversion, emergency EVAR for ruptured AAA, or who had missing data, 832 EVAR procedures for AAA remained for retrospective analysis with a median follow-up period of 35.6 months. Average patient age was 75.4±8.3 years (range, 45–98 years), and 81.6% were male, with an average AAA transverse diameter of 50.4±10.9 mm (range, 25–96 mm). Stent grafts used for EVAR were the Zenith (33.4%), Excluder (35.3%), Powerlink (12.8%), and Endurant (17.3%) models, and average time for EVAR was 187.7±75.2 min (range, 47–653 min).
Incidence of T2EL and Clinical CourseA total of 234 cases of T2EL (28.1%) were observed after 832 EVAR procedures. Initial CT within 1 month following EVAR indicated 181 T2EL present from the beginning of the surveillance. During follow-up, spontaneous resolution of T2EL was also observed in 53 patients (6.4%), but late T2EL (defined as T2EL found >1 month after EVAR) appeared in another 53 patients (6.4%). In the entire cohort, 47 patients (5.6%) had sac expansion >5 mm during follow-up; isolated T2EL (defined as T2EL without any other endoleaks) was involved in 32 patients (3.8%).
Patient Demographics and CharacteristicsBaseline demographics and characteristics of patients with and without T2EL are summarized in Table 1. On univariate analysis there were no differences between patients with or without T2EL in relation to hypertension, diabetes, coronary artery disease, cerebrovascular disease, chronic renal insufficiency, hemodialysis, high American Society of Anesthesiologists score (>3), antiplatelet therapy, or anticoagulation. In contrast, age>80 years (P=0.003), female sex (P=0.001), and non-smoker status (P<0.001) remained statistically significant risk factors for T2EL; in addition, chronic obstructive pulmonary disease (COPD) and large aneurysm size (>55 mm) showed a trend toward increased risk for T2EL (P<0.10).
| Variable | All patients (n=832), n (%) |
Without T2EL (n=598), n (%) |
With T2EL (n=234), n (%) |
P-value |
|---|---|---|---|---|
| Age >80 years | 287 (34.5) | 188 (31.4) | 99 (42.3) | 0.003 |
| Male sex | 680 (81.6) | 505 (84.4) | 175 (74.8) | 0.001 |
| Aneurysm size >55 mm | 254 (30.5) | 172 (28.8) | 82 (35) | 0.077 |
| Hypertension | 634 (76.1) | 460 (76.9) | 174 (74.4) | 0.435 |
| Diabetes | 125 (15.0) | 95 (15.9) | 30 (12.8) | 0.266 |
| Coronary artery disease | 299 (35.9) | 213 (35.6) | 86 (36.8) | 0.759 |
| Cerebrovascular disease | 130 (15.6) | 96 (16.1) | 34 (14.5) | 0.586 |
| COPD | 114 (13.7) | 92 (15.4) | 9.4 (22) | 0.063 |
| CRI (eGFR <59) | 381 (45.7) | 280 (46.8) | 101 (43.2) | 0.341 |
| Hemodialysis | 24 (2.9) | 20 (3.3) | 4 (1.7) | 0.367 |
| History of smoking† | 544 (65.4) | 430 (71.9) | 114 (48.7) | <0.001 |
| ASA score >3 | 82 (9.9) | 60 (10.0) | 22 (9.4) | 0.783 |
| Antiplatelet therapy‡ | 374 (44.9) | 274 (45.8) | 100 (42.7) | 0.421 |
| Anticoagulation | 62 (7.4) | 44 (7.4) | 18 (7.7) | 0.325 |
†Includes current smoker and former smoker. ‡Includes aspirin, clopidogrel, ticlopidine, cilostazol, sarpogrelate, or beraprost. ASA, American Society of Anesthesiologists; COPD, chronic obstructive pulmonary disease; CRI, chronic renal insufficiency; eGFR, estimated glomerular filtration rate; T2EL, type 2 endoleak.
Operative risk factors that may influence the incidence of T2EL, including devices, are listed in Table 2. No difference was found between patients with or without T2EL in terms of bleeding, outside instruction for use, and T4EL at completion angiogram. A significantly higher incidence of T2EL, however, was observed in patients with a longer operative time (P=0.047), blood transfusion (P=0.005), and T2EL at completion angiogram (P<0.001). In terms of device differences, Zenith and Powerlink were associated with a significantly lower incidence of T2EL (P=0.020 and 0.001, respectively), and Endurant tended toward a lower incidence of T2EL (P<0.010). Only Excluder was associated with a significantly higher incidence of T2EL (P<0.001).
| Variable | All patients (n=832) | Without T2EL (n=598) | With T2EL (n=234) | P-value |
|---|---|---|---|---|
| Operative time (min) | 187.7±75.2 | 184.4±77.9 | 196.0±67.3 | 0.047 |
| Bleeding (ml)† | 190.9±324.5 | 184.9±339.9 | 206.2±281.3 | 0.396 |
| Blood transfusion | 59 (7.1) | 33 (5.5) | 26 (11.1) | 0.005 |
| Outside IFU | 210 (25.2) | 151 (25.3) | 59 (25.2) | 0.991 |
| Zenith stent graft | 278 (33.4) | 214 (35.8) | 64 (27.4) | 0.02 |
| Excluder stent graft | 294 (35.3) | 176 (29.4) | 118 (50.4) | <0.001 |
| Endurant stent graft | 144 (17.3) | 112 (18.7) | 32 (13.7) | 0.083 |
| Powerlink stent graft | 107 (12.8) | 91 (15.2) | 16 (6.8) | 0.001 |
| T2EL at final angiogram | 185 (22.2) | 107 (17.9) | 78 (33.3) | <0.001 |
| T4EL at final angiogram | 88 (10.6) | 67 (11.2) | 21 (9.0) | 0.347 |
Data given as mean±SD or n (%). EVAR, endovascular aneurysm repair; IFU, instruction for use; T2EL, type 2 endoleak; T4EL, type 4 endoleak.
Multivariate analysis was conducted for variables with P<0.10: age >80 years, male sex, large aneurysm size (>55 mm), COPD, non-smoker status, operative time, blood transfusion, T2EL at final angiogram, and Excluder stent graft (Table 3). Although other stent graft devices showed significant differences, only Excluder was included in the multivariate analysis because it had been associated with the increased risk of T2EL and had the strongest influence on T2EL over other devices. On logistic regression analysis, non-smoker status (OR 2.216; P<0.001), Excluder device (OR 2.027; P<0.001), and T2EL at final angiogram (OR 2.080; P<0.001) were all risk factors for T2EL.
| Factor | OR (95% CI) | P-value |
|---|---|---|
| Age >80 years | 1.344 (0.953–1.896) | 0.092 |
| Male sex | 0.943 (0.612–1.453) | 0.791 |
| Aneurysm >55 mm | 1.246 (0.881–1.762) | 0.213 |
| COPD | 0.713 (0.424–1.199) | 0.202 |
| Non-smoker† | 2.216 (1.552–3.165) | <0.001 |
| Excluder stent graft | 2.027 (1.457–2.818) | <0.001 |
| Operative time (min) | 1 (0.998–1.002) | 0.883 |
| Blood transfusion | 1.567 (0.844–2.910) | 0.155 |
| T2EL at final angiogram | 2.08 (1.443–2.996) | <0.001 |
†Includes only patients without any history of smoking. CI, confidence interval; OR, odds ratio. Other abbreviations as in Table 1.
Using the same method, multivariate analysis was also conducted for the risk factors of sac expansion after EVAR (Table 4) and for the risk factors of isolated T2EL with sac expansion (Table 5). On logistic analysis for the risk factors of sac expansion after EVAR, age >80 years, non-smoker status, T1EL and T2EL were significant risk factors; other than T1EL, however, isolated T2EL was the most significant risk factor for sac expansion after EVAR (OR, 18.486; P<0.001). For isolated T2EL with sac expansion, logistic regression analysis left only non-smoker status as a significant risk factor (OR, 2.671; P<0.031).
| Factor | OR (95% CI) | P-value |
|---|---|---|
| Age >80 years | 2.187 (1.026–4.66) | 0.043 |
| Male sex | 1.979 (0.817–4.794) | 0.131 |
| Aneurysm >55 mm | 1.544 (0.747–3.190) | 0.241 |
| Hypertension | 0.387 (0.181–0.827) | 0.014 |
| Diabetes | 0.343 (0.071–1.665) | 0.184 |
| Non-smoker† | 2.246 (1.025–4.924) | 0.043 |
| Excluder stent graft | 0.914 (0.443–1.886) | 0.808 |
| Operative time (min) | 0.996 (0.991–1.001) | 0.094 |
| T1EL | 22.081 (7.143–68.258) | <0.001 |
| T2EL | 18.486 (7.123–47.976) | <0.001 |
†Includes only patients without any history of smoking. T1EL, type 1 endoleak. Other abbreviations as in Tables 1–3.
| Factor | OR (95% CI) | P-value |
|---|---|---|
| Age >80 years | 1.987 (0.844–4.675) | 0.116 |
| Male sex | 2.79 (0.968–8.040) | 0.057 |
| Aneurysm >55 mm | 1.487 (0.656–3.374) | 0.342 |
| COPD | 0.286 (0.034–2.418) | 0.251 |
| Non-smoker† | 2.671 (1.096–6.508) | 0.031 |
| Excluder stent graft | 0.889 (0.392–2.019) | 0.779 |
| Operative time (min) | 0.995 (0.990–1.001) | 0.090 |
| Blood transfusion | 2.033 (0.697–5.931) | 0.194 |
| T2EL at final angiogram | 1.234 (0.530–2.870) | 0.626 |
†Includes only patients without any history of smoking. Abbreviations as in Table 1.
Out of 32 patients with isolated T2EL with sac expansion, 11 underwent transarterial embolization using a triple coaxial technique (Table 6). Reintervention for T2EL was performed on average 22.4 months after the initial EVAR. Embolization of the target vessel was successful in all cases, and only 1 patient required a transabdominal approach under CT guidance owing to the difficulty of approaching the lumbar artery from the hypogastric artery. During mean follow-up of 16 months, however, only 4 T2EL disappeared, leaving 7 persistent T2EL, 5 of which had enlarged aneurysm size and 4 of which ultimately led to rupture.
| Patient ID no. |
Time to reintervention (months) |
Size change† (mm) |
Clinical symptoms |
Target vessel |
Route | Latest aneurysm size‡ (mm) |
Result (follow-up after reintervention; months) |
|---|---|---|---|---|---|---|---|
| 1 | 31 | 57 ⇒ 79 | Impending rupture |
IMA LA | TAE | 94 | Rupture from persistent T2EL, conservatively managed (14) |
| 2 | 12 | 55 ⇒ 64 | None | IMA | TAE | 83 | Refused further treatment; death from rupture (29) |
| 3 | 26 | 61 ⇒ 68 | None | LA | TAE | 65 | Persistent T2EL with stable aneurysm sac (26) |
| 4 | 30 | 85 ⇒ 90 | None | LA | TAE | 90 | Open surgical repair for rupture from persistent T2EL (12) |
| 5 | 27 | 51 ⇒ 75 | None | IMA LA | Trans abd | 75 | No T2EL with stable aneurysm sac (36) |
| 6 | 26 | 53 ⇒ 60 | None | IMA LA | TAE | 47 | No T2EL with regression of aneurysm sac (26) |
| 7 | 33 | 50 ⇒ 62 | None | IMA LA | TAE | 62 | No T2EL with stable aneurysm sac (14) |
| 8 | 19 | 70 ⇒ 79 | None | IMA LA | TAE | 95 | Persistent T2EL with enlargement of aneurysm sac and death from another cause (13) |
| 9 | 8 | 70 ⇒ 100 | Impending rupture |
IMA | TAE | 110 | Refused further treatment; death from rupture (4) |
| 10 | 14 | 67 ⇒ 73 | None | IMA LA | TAE | 73 | No T2EL with stable aneurysm sac (4) |
| 11 | 20 | 60 ⇒ 65 | None | LA | TAE | 65 | Persistent T2EL (IMA) with stable aneurysm sac (2) |
†Change of aneurysm size from the time of initial EVAR to reintervention. ‡Aneurysm size at the latest (final) imaging study. IMA, inferior mesenteric artery; LA, lumbar artery; TAE, transarterial embolization; Trans abd, transabdominal embolization under computed tomography guidance. Other abbreviations as in Tables 1,2.
Even though T2EL is the most frequent complication observed after EVAR, debate continues on all aspects of T2EL, including terminology, predictability, clinical significance, surveillance modality, and treatment.12 Because the key to T2EL ceasing is thrombus formation within residual aneurysm sac, many reports have focused on the parameters that affect thrombus formation: from antiplatelet therapy and anticoagulation27–30 to metabolic syndrome,31 smoking,6,10 anatomy of the aortic branches,17,32 and preoperative embolization of the aortic branches.33,34 The current literature, however, lacks information about ethnic differences in T2EL.
It is clear that ethnic differences are significant with regard to demographics,21 anatomy22,23 and coagulability.24,25 In terms of coagulability, East Asian subjects are known to have a less prothrombotic state than Caucasian subjects,24,25 a fact that is well documented by the low incidence of venous thromboembolism reported in East Asian populations.35,36 Although the evidence for T2EL in the East Asian population has been limited, based on the lower prothrombotic state reported and higher incidence of endoleaks in patients with anticoagulation therapy,29,30 we postulated that the incidence of T2EL in the present database would be relatively high. Indeed, the incidence of T2EL in the database was 28.1%, higher than in previous reports.6 Given that other patient characteristics, such as age, sex, rates of hypertension, diabetes, and chronic renal insufficiency, were similar to those in the Caucasian database,1,6 this ethnic difference in coagulability may have played an important role in the incidence of T2EL.
As for the risk factors of T2EL in the East Asian population, multivariate analysis of the database indicated that non-smoker status, Excluder stent graft, and T2EL at final angiogram were significant risk factors. It is easily understood that the presence of T2EL at final angiogram would be a risk factor for T2EL during follow-up, given that we would be observing the same phenomenon. Non-smoker status as an independent risk factor for T2EL in the database seems contradictory given the relationship between smoking and AAA,37 but it supports the previous observation that atherosclerosis of the lumbar artery and inferior mesenteric artery (IMA) and the prothrombotic state caused by smoking lower the incidence of T2EL.6,10 Furthermore, multivariate analysis of risk factors for T2EL with sac expansion >5 mm left only non-smoker status as a significant risk factor, indicating the importance of anatomy and coagulability for the clinical course of T2EL.
Differences in the incidence of T2EL according to stent graft, with Excluder having the most T2EL, has already been reported by Sheehan et al,8 but they also reported that this difference was observed only during the early phase of follow-up, and the difference became less noticeable over the years. In contrast, according to the present database there was a remarkably high incidence of T2EL for Excluder stent grafts (40.1%) compared with the previous report.38 In this study, the incidence of T2EL for other stent grafts was 21.6%, making Excluder an independent risk factor for T2EL. The clinical significance of this observed difference is unclear, given that Excluder was not the significant risk factor for either sac expansion after EVAR or T2EL with sac expansion >5 mm. One of the device-specific configurations of Excluder that may have influenced incidence of T2EL is the short main body. Short main body may result in increased incidence of T2EL, because it has less chance to seal IMA or lumbar artery compared with other stent grafts when a long proximal neck exists; but, because this remarkably high incidence of T2EL after Excluder stenting was not reported previously in Caucasian databases, further investigation is warranted to clarify the interaction between the device-specific configuration of Excluder and ethnic differences.
For the treatment of T2EL, the present initial approach of transarterial embolization using a triple coaxial system (1.9-Fr non-tapered microcatheter passed through a 2.7-Fr microcatheter advanced coaxially through a 4-Fr catheter)26 was similar to the translumbar approach frequently used in previous reports of successful embolization of the target vessel, which in the present series was achieved in 10 out of 11 patients. The clinical course after reintervention, however, also replicated previous reports.16,17,19,20 There was a low rate of complete resolution of T2EL and a gradual expansion of the aneurysm sac even after reintervention, leading to 4 cases of rupture caused by isolated T2EL. These results clearly imply an urgent need to develop new treatment approaches for T2EL other than embolization through either a translumbar or transarterial approach.
Similar to previous reports,12–15 the clinical significance of T2EL in East Asian populations remained unclear, given that we observed a low rate of sac expansion >5 mm (3.8%) and only 4 cases of rupture (0.5%) in the present database. Also, all patients with rupture presented to hospital with only pain, enabling sufficient time for elective treatment to be performed, and 3 patients who did not receive any surgical treatment were able to be managed conservatively at initial hospitalization. The incidence of T2EL in the database, however, was high (28.1%) compared with previous reports that used Caucasian databases, and T2EL was clearly a significant risk factor for sac expansion after EVAR (OR, 18.486). Given that the present mean observation time was relatively short (35.6 months), we will probably observe more aneurysm sac expansions due to T2EL in the future. Together with the poor results for the current reintervention for T2EL, special attention to T2EL in East Asian populations may be required.
The present study had several limitations. First, this was a retrospective analysis of a prospectively maintained database with a relatively short mean follow-up time. Longer follow-up time (>5 years, if possible) may have resulted in more cases of sac expansion caused by T2EL, considering both the high incidence of T2EL and the significant influence of T2EL on sac expansion after EVAR in the present Japanese database. Second, we used the Japanese database to represent East Asian population, given that there seems to be distinct differences between non-East Asian and East Asian patients in terms of coagulability and bleeding risk factors.39–41 Also, similarity between East Asian countries with regard to incidence of protein S deficiency and Factor V Leiden has been observed,42,43 but whether the present results are applicable to all East Asian countries needs further investigation. Third, we did not have preoperative anatomical information, such as number of patent lumbar arteries, size of IMA, or amount of mural thrombus, which may have significantly influenced the results for T2EL. Furthermore, even though we discuss coagulability as a significant factor affecting incidence of T2EL, the present database showed no difference in incidence of T2EL between the anticoagulation and non-anticoagulation groups. We believe that given that other factors also affect coagulability, anticoagulation alone does not represent the state of coagulability in a patient. There is definitely need, however, for further investigation into the relationship between coagulability and incidence of T2EL. Last, we found no clear explanation for the high incidence of T2EL observed in the Excluder stent patients: the only significant risk factor that differed from previous studies of T2EL. To clarify whether ethnic differences had any influence on the incidence of T2EL in Excluder stent patients, well-planned international prospective studies or randomized controlled studies will be required in the future.
In the present Japanese multicenter database, a strong association was found between East Asian T2EL and non-smoker status, as previously described in the European Collaborators on Stent-Graft Techniques for Aortic Aneurysm Repair (EUROSTAR) database. Although the incidence of sac expansion caused by T2EL was relatively low, T2EL was a significant risk factor for sac expansion, which can lead to rupture even after intervention. Along with the high prevalence of T2EL observed in the present database, which may lead to increase of major adverse events in future, T2EL in the East Asian population may not always be benign.
The authors received no financial support for the research, authorship, and/or publication of this article.
