- Email: [email protected]
Select early type IA endoleaks after endovascular aneurysm repair will resolve without secondary intervention
From the Vascular and Endovascular Surgery Society
Select early type IA endoleaks after endovascular aneurysm repair will resolve without secondary intervention Thomas F. X. O’Donnell, MD, Michael R. Corey, MD, Sarah E. Deery, MD, Gregory Tsougranis, BS, Rohit Maruthi, BS, W. Darrin Clouse, MD, Richard P. Cambria, MD, and Mark F. Conrad, MD, MMSc, Boston, Mass
ABSTRACT Objective: Although it is traditionally considered ominous, the natural history of early proximal attachment site endoleaks (IA) after endovascular aneurysm repair (EVAR) is not well known. Our aim was to identify risk factors for persistent type IA endoleaks and to determine their effect on long-term outcomes after EVAR. Methods: All patients who underwent infrarenal EVAR at a single institution between 1998 and 2015 were identified. Preoperative axial imaging and intraoperative arteriograms were reviewed, and those patients with a type IA endoleak were further studied. Aneurysm features were characterized by two reviewers and were studied for predictors of persistent endoleaks at the conclusion of the case. Patient records and the Social Security Death Index were used to record 1-year and overall survival. Results: We identified 1484 EVARs, 122 (8%) of which were complicated by a type IA endoleak on arteriography after graft deployment, with a median follow-up of 4 years. The majority of patients underwent additional ballooning of the proximal site (52 [43%]) or placement of an aortic cuff (47 [39%]); 30 patients (25%) received a Palmaz stent, and four patients were treated with coils or anchors. At case end, only 43 (35%) of the type IA endoleaks remained; at 1 month, only 16 endoleaks persisted (13%), and only six persisted at 1 year (6%). In multivariable analysis, the only independent predictor of persistence of type IA endoleak at the conclusion of the case was the presence of extensive neck calcifications (odds ratio [OR], 9.9; 95% confidence interval [CI], 1.4-67.9; P ¼ .02). Thirteen patients (11%) underwent reintervention for type IA endoleaks, with a time frame ranging from 3 days postoperatively to 11 years. There were three patients (2.4%) who experienced aneurysm rupture. Postoperative type IA endoleak was associated with lower survival at 1 year (79% vs 91%; relative risk, 2.5; 95% CI, 1.1-5.4; P ¼ .02), but it did not affect long-term survival (log-rank, P ¼ .45). Both an increase in aneurysm sac size and failure of the endoleak to resolve by case end were independent predictors of a need for reintervention (growth: OR, 8.3; 95% CI, 2.2-31.6; P < .01; persistent endoleak: OR, 7.6; 95% CI, 1.8-31.5; P < .01). A persistent type IA endoleak was not independently associated with an increase in sac size on surveillance imaging (P ¼ .28). Conclusions: Aneurysm rupture secondary to persistent type IA endoleak is rare, and most will resolve within 1 year. Extensive neck calcification is the only independent predictor of persistent type IA endoleak, and an increase in sac size warrants reintervention. These data suggest that select early persistent type IA endoleaks can be safely observed. (J Vasc Surg 2018;67:119-25.)
Endovascular aneurysm repair (EVAR) was first introduced in 1991 and by 2008 accounted for 77% of the repairs performed in Medicare patients.1,2 Type IA endoleaks occur when there is persistent flow into the aneurysm sac due to inadequate proximal seal and are seen
From the Division of Vascular and Endovascular Surgery, Massachusetts General Hospital. Author conflict of interest: none. Presented at the Twenty-seventh Annual Meeting of the Vascular and Endovascular Surgery Society, Steamboat, Colo, February 2-5, 2017. Correspondence: Mark F. Conrad, MD, MMSc, Massachusetts General Hospital, 15 Parkman St, Boston, MA 02114 (e-mail: [email protected]). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest. 0741-5214 Copyright Ó 2017 by the Society for Vascular Surgery. Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jvs.2017.05.096
on completion angiography in up to 11% of EVAR cases.3 Despite their prevalence, the risk factors for, natural history of, and prognosis of these endoleaks are not well characterized. The Society for Vascular Surgery recommends that “every attempt should be made to resolve type IA endoleaks before the patient leaves the intervention suite.”3 This recommendation is based on the belief that unlike in type II endoleaks, where there is persistent flow from patent branch vessels of the aorta, continued flow through the proximal seal zone is more likely to produce high pressure that can lead to sac expansion and places the patient at continued risk of rupture. Indeed, data from the European Collaborators on Stent/graft Techniques for aortic Aneurysm Repair (EUROSTAR) trial showed type IA endoleak to be a risk factor for rupture,4-6 and Venermo et al7 found that type IA endoleak was associated with a 3-year aneurysm-related mortality of 11%. 119
120
Journal of Vascular Surgery
O’Donnell et al
January 2018
It has been our practice to exhaust all options for resolving type IA endoleaks before leaving the operating room. Despite this effort, some will persist, and a decision must be made as to how the patient should be observed and whether conversion to open explantation is warranted to prevent rupture. Recent data suggest that type IA endoleaks may not carry such a dire prognosis as previously believed. Several studies have demonstrated low rates of rupture and aneurysm-related mortality and high rates of spontaneous sealing in select patients,8-14 but these studies are limited by small numbers and limited follow-up. The goal of this study was to examine our institutional experience with type IA endoleaks in an effort to document long-term outcomes.
METHODS Subjects. This is a retrospective review of all patients who underwent infrarenal EVAR at the Massachusetts General Hospital from 1998 to 2015. Aneurysms that were a consequence of chronic dissection were excluded, but ruptured aneurysms were included in the cohort. Patients were identified through the use of Current Procedural Terminology codes. The charts were then reviewed to confirm the procedures performed and to identify patients who had a type IA endoleak on the initial angiogram after deployment. The Massachusetts General Hospital Institutional Review Board approved the study and waived the need for patient consent, given the retrospective nature of the study. Aneurysm characteristics. The preoperative computed tomography (CT) angiograms, intraoperative angiograms, and postoperative CT angiograms were reviewed by two authors (T.F.X.O. and M.R.C.). Preoperative CT angiograms were used to calculate the maximal anteroposterior aneurysm diameter, diameter of the aorta at the lowest renal artery, length of the neck, infrarenal and suprarenal neck angles, proximal and distal neck diameter, presence of a conical neck (>5-mm difference in diameter of proximal neck and the neck at its widest point), and presence of extensive neck calcifications (as assessed qualitatively by the two reviewers, defined as calcifications involving >50% of the circumference). Intraoperative events. Intraoperative angiograms and operative reports were reviewed by two authors (T.F.X.O. and M.R.C.) to assess for the persistence of type IA endoleak at the close of the procedure as well as to record any supplementary intraoperative interventions undertaken to ensure proximal seal. Outcomes. The primary outcome was overall survival, which was recorded through patient records and the Social Security Death Index. Secondary outcomes included persistent endoleak at the conclusion of the
ARTICLE HIGHLIGHTS d
d
d
Type of Research: Retrospective, single-center cohort study Take Home Message: Forty-three patients with persistent type IA endoleaks were observed after endovascular aneurysm repair; 16 endoleaks persisted at 1 month and six at 1 year. The only predictor of the presence of a postprocedural type IA endoleak was neck calcification. There were three instances of abdominal aortic aneurysm rupture and 13 cases of reintervention. Postoperative type IA endoleak was associated with lower 1-year survival but not longer term survival or increase in sac size. Recommendation: This study suggests that some early type IA endoleaks can be observed but that sac size increase should prompt reintervention.
case, at 1 month, and at 1 year, as assessed by follow-up CT angiography, and reintervention or aneurysm rupture. Secondary outcomes were ascertained through chart review. Statistical analysis. Categorical variables were compared with the Fisher exact test. Continuous variables were analyzed with the Wilcoxon rank sum test. Univariate survival analysis was carried out with the logrank test. Logistic regression was performed to determine aneurysm characteristics that were independent predictors of persistent endoleak. Variables that were included in the logistic regression were age, procedure year, female sex, ruptured aneurysms, maximal anteroposterior aneurysm diameter, diameter of the aorta at the lowest renal artery, length of the neck, infrarenal and suprarenal neck angles, maximal neck diameter, presence of a conical neck (>5-mm difference in diameter of proximal neck and the neck at its widest point), and presence of extensive neck calcifications. We tested whether being outside the manufacturer’s instructions for use (IFU) with respect to the proximal landing zone was independently associated with type IA endoleak by constructing a separate logistic regression model that included all of the aforementioned variables except for those factors that are part of the IFU (neck length, neck angles, and neck diameter) to avoid collinearity. Cox proportional hazards models were constructed to examine the association between persistent endoleak and death, independent of age and aneurysm size at presentation. Age and aneurysm size were chosen a priori to be controlled for in multivariable analyses as these have previously been shown to be independent predictors of survival. All statistical analyses were carried out using Stata 14.2 software (StataCorp, College Station, Tex). Sensitivity analysis. Given that ruptured aneurysms may represent a category of patients with less than ideal
Journal of Vascular Surgery
O’Donnell et al
121
Volume 67, Number 1
Table I. Baseline characteristics Endoleak resolved (n ¼ 79) Characteristic
Overall cohort (N ¼ 122)
Persistent endoleak (n ¼ 43)
Median (interquartile range) or No. (%)
P value
Age
80 (74-85)
79 (73-85)
82 (78-85)
.07
Female sex
38 (31)
24 (30)
14 (33)
.84
Rupture
15 (12)
8 (10)
7 (16)
.39
Aneurysm size
58 (53-64)
58 (53-67)
60 (52-63)
.97
Diameter at lowest renal
23 (21-26)
23 (21-27)
24 (22-26)
.34
Largest neck diameter
26 (23-29)
25 (23-29)
26 (23-29)
.48
Neck length
28 (21-38)
31 (21-38)
26 (20-37)
.31
Suprarenal neck angle
10 (5-20)
10 (5-24)
10 (5-15)
.27
Infrarenal neck angle
20 (10-40)
20 (10-38)
22 (10-45)
.33
Conical neck
31 (26)
16 (21)
15 (36)
.09
Outside IFU
29 (24)
16 (21)
13 (31)
.27
Neck calcifications
20 (17)
6 (8)
14 (33)
<.01
IFU, Instructions for use.
anatomy, a sensitivity analysis was performed excluding those patients with rupture from the analysis.
RESULTS Patients. Between 1998 and 2015, there were 1484 EVARs performed, 122 (8%) of which were complicated by a type IA endoleak on angiography after deployment, with a median follow-up of 48 months (interquartile range, 22-78 months). Baseline characteristics and anatomic data are presented in Table I. Of note, 24% of the patients had anatomies that were outside the manufacturer’s IFU. However, in univariate analysis, the only factor associated with persistent endoleak was presence of extensive neck calcifications. By the close of the procedure, 35% of those endoleaks persisted (43/122). At 1-month follow-up CT angiography, only 13% had evidence of persistent type IA endoleak (16/119) despite only three reinterventions, and only 6% were evident at 1 year (6/96) despite only an additional five reinterventions (Fig 1). None of the patients whose endoleaks resolved in the operating room had an endoleak at 1 year. These results were similar when only intact aneurysms were considered (12% at 1 month, 7% at 1 year). Of the 43 patients who left the operating room with a persistent endoleak, 13 (30%) had a type IA endoleak at 1 month and 6 (14%) had evidence of endoleak at 1 year. These rates were also similar when only intact aneurysms were considered (28% at 1 month, 24% at 1 year). Intraoperative procedures. After a type IA endoleak was identified following graft deployment, all patients underwent additional maneuvers in attempts to resolve the leak (Table II). The most common adjunct was additional ballooning alone, which was performed in 52 (43%) patients with a 62% rate of successful resolution of the
endoleak. An additional 47 (39%) patients underwent placement of an aortic cuff, and 30 (25%) patients received a Palmaz stent, with success rates of 70% and 53%, respectively. Three patients underwent anchor placement without success. None of the therapies were associated with endoleak resolution in univariate analysis, except anchors, which were associated with failure to resolve (relative risk, 2.98; 95% CI, 2.3-3.8; P ¼ .02). Predictors of persistent endoleak. In multivariable analysis, the only independent predictor of persistence of type IA endoleak at the conclusion of the case was the presence of extensive neck calcifications (odds ratio [OR], 11.5; 95% confidence interval [CI], 1.7-76.0; P ¼ .01; Table III). No other anatomic features were predictive of failure to resolve, including being outside the IFU (OR, 0.71; 95% CI, 0.17-3.0; P ¼ .64). Excluding ruptured aneurysms did not change the results. The proportion of endoleaks that persisted at the end of the case decreased over time (P < .01; Fig 2). Survival. At 1 year, 91% of those whose endoleak resolved by the end of the case were alive compared with 79% of those who left the operating room with a persistent endoleak (relative risk for death, 2.45; 95% CI, 1.12-5.36; P ¼ .02). After adjustment for age and aneurysm size, leaving the operating room with a type IA endoleak remained a significant risk factor for mortality at 1 year (OR, 3.79; 95% CI, 1.28-11.24; P ¼ .02). Despite the association with 1-year survival, there was no difference in overall survival in univariate analysis between patients who left the operating room with a persistent type IA endoleak and those who did not (log-rank, P ¼ .26; Fig 3). Five-year survival was 56% for those patients who left the operating room without an endoleak compared with 52% for those who left with
122
Journal of Vascular Surgery
O’Donnell et al
January 2018
Fig 1. Evolution of intraoperative endoleaks during 1-year follow-up. Interventions to the left of the arrows refer to any reintervention that occurred during the time interval. EVARs, Endovascular aneurysm repairs.
Table II. Intraoperative adjunctive procedures Procedure
No. (% of total)
Resolution achieved, %
Ballooning alone
52 (43)
62
Palmaz stent
30 (25)
53
Cuff
47 (39)
70
Anchors
3 (2.5)
0a
Coils
1 (1)
0
Significant at P < .05.
a
Table III. Logistic regression model for failure of type IA endoleak to resolve in the operating room Characteristic
OR
95% CI
P
Age
0.93
0.85-1.02
.12
Female sex
0.97
0.24-3.93
.97
1.04
0.98-1.10
.19
0.93
0.66-1.31
.68
1.00
0.74-1.37
.99
Aneurysm size Diameter at lowest renal Largest neck diameter Neck calcifications
10.36
1.54-69.86
.02
Neck length
1.02
0.98-1.07
.34
Suprarenal neck angle
1.05
0.99-1.13
.13
1.00
0.97-1.04
.83
Infrarenal neck angle Conical neck
0.18
0.02-1.96
.16
Ruptured aneurysm
0.30
0.05-1.95
.21
Outside IFU
0.38
0.04-3.41
.39
CI, Confidence interval; IFU, instructions for use; OR, odds ratio.
an endoleak. This lack of association remained after adjustment for age and aneurysm size (hazard ratio, 0.64; 95% CI, 0.38-1.07; P ¼ .09). Excluding those patients with ruptures did not change the results, as persistent endoleak remained
associated with 1-year survival (P ¼ .47) but not overall survival (P ¼ .5). Reintervention and late rupture. There were 13 (11%) patients who underwent at least one reintervention, with a median time to intervention of 17 months (range, 3 days-11 years) from the date of the initial operation. There were three ruptures (2.4% of the overall cohort, 7.0% of the patients with persistent endoleaks), two from type IA endoleaks (1.6% overall, 4.7% of persistent endoleaks). One patient was observed for 3 years with a type IA endoleak. Her sac continued to enlarge but she refused intervention until she presented with a rupture 3 years after EVAR. She was successfully treated with an open repair and survived an additional 3 years. Another patient was taken back to the operating room on postoperative day 3 for attempted endovascular repair of a large type IA endoleak after the patient and family refused open intervention. This was unsuccessful, and she was discharged to hospice care where she subsequently died, presumably from rupture, but no autopsy was performed. The last rupture was secondary to a type III endoleak (component separation) at 9 months and was successfully treated with repeated EVAR. There were two open operations, one of which was the rupture at 3 years; the second was an open juxtarenal aneurysm repair for aneurysmal degeneration at the proximal extent. The other 11 patients were all treated with endovascular interventions, mostly cuffs. Both an increase in size of the aneurysm sac and failure of the endoleak to resolve by the case end were independent predictors of a need for reintervention (growth: OR, 8.3; 95% CI, 2.2-31.6; P < .01; persistent endoleak: OR, 7.6; 95% CI, 1.8-31.5; P < .01). However, a persistent endoleak was not associated with an increase in sac size on surveillance imaging (P ¼ .28).
Journal of Vascular Surgery
O’Donnell et al
123
Volume 67, Number 1
Fig 2. Rate of persistent endoleak at case end over time.
Fig 3. Overall survival, by presence of persistent endoleak. SE, Standard error.
DISCUSSION Type IA endoleaks are present on the initial angiogram after deployment in up to 10% of patients, and despite additional adjunctive procedures, 35% of patients in this study had a persistent endoleak at the end of the procedure. A persistent type IA endoleak has traditionally been considered an absolute indication for reintervention, including conversion to open aneurysm repair to prevent rupture. However, this study demonstrates that many type IA endoleaks that were persistent at the end of the procedure will spontaneously resolve with reversal of systemic anticoagulation and can often be observed with close surveillance with a low rate (4.7%) of subsequent rupture. Similar results have been seen in several small series in which >50% of persistent type IA endoleaks have resolved by the first follow-up CT scan.7,8,10,13 With longer follow-up, only 14% of the endoleaks present at the case end were still present at 1 year. More important, there was no difference in overall survival in those patients who left the operating room with a persistent endoleak compared with those whose endoleaks resolved. This is consistent with the findings of Bastos Gonçalves et al,8 who
followed up 15 patients with persistent type IA endoleaks and found that all had resolved by 1 year after the initial procedure, and Millen et al,13 who studied 33 persistent endoleaks and found that 31 (94%) resolved by the first postoperative scan. The ultimate goal of elective abdominal aortic aneurysm repair regardless of approach (open vs EVAR) is to prevent aneurysm rupture and subsequent death. Consequently, any aneurysm-related event after EVAR should be considered a failure of the procedure. It is reassuring that in this series, there were only two ruptures attributable to type IA endoleaks, both of which were potentially avoidable as both patients had refused further intervention before their ruptures. One patient presented with an early rupture. She was considered a poor open candidate, and her neck anatomy was not favorable for endovascular repair. Despite aggressive attempts to obtain adequate seal in the proximal zone, there was a persistent, large endoleak in a large aneurysm. She refused open conversion and was discharged to hospice, where she died a few days later, presumably of rupture. The second rupture occurred after 3 years, during which time the patient had a known endoleak with expanding aortic sac but refused any further intervention. Similar to our experience, a fatal early rupture at 2 days postoperatively was reported by Bastos Gonçalves et al,8 for which an open repair was attempted but was unsuccessful. In addition to the outcomes after type IA endoleaks, the nature and frequency of reinterventions have not been well defined. Previous studies suggested that most reoperations are performed open, but they suffered from small sample sizes and limited follow-up, with very few reoperations captured. Bastos Gonçalves et al8 reported six reinterventions of 15 endoleaks, three of which were performed open and three endovascularly. Kim et al10 performed two reoperations on 10 endoleaks; again, one was open and one endovascular. Tan et al14 studied 80 endoleaks and found three perioperative conversions to open, and Millen et al13 studied 33 persistent endoleaks and found that only one required Onyx (ev3, Irvine, Calif). In our study, most of the reinterventions were performed endovascularly, with only three open operations of 11. The open operations were all traditional open aneurysm resections, with graft explantation. Endovascular approaches ranged from aortic cuffs to Palmaz stents, redo EVARs, Onyx glue, Aptus anchors (Aptus Endosystems, Sunnyvale, Calif), and one hybrid operation with an aortouni-iliac limb with a femorofemoral bypass. Tzortzis et al15 described 22 endoleaks, 15 of which were treated with periaortic banding around the proximal zone, but this technique was not used in our series. Of note is the substantial improvement in the rate of persistent endoleak at case end over time. This is likely due to improvements in grafts over time as well as to the surgeon’s technique and experience with the procedure.
124
Journal of Vascular Surgery
O’Donnell et al
January 2018
Although the rates of rupture and reintervention in this study were low, patients with persistent type IA endoleaks were observed with a surveillance protocol that was more aggressive than our standard imaging followup. Most patients who left the operating room with a type IA endoleak underwent CT angiography before discharge. This was repeated in 4 to 6 weeks, and if the endoleak did not resolve, patients were imaged at 6 months and 1 year after the initial procedure. Patients with persistent endoleaks as well as those who continue to show aneurysm sac growth should be considered for a more aggressive imaging schedule as these factors independently predicted the need for reintervention in this study. As these endoleaks frequently resolve spontaneously, growth of the aneurysm sac captured on surveillance imaging, rather than the mere presence of an endoleak, should be the impetus for reintervention. Indeed, Venermo et al7 found that type IA endoleak was associated with lower rates of rupture than in patients with untreated abdominal aortic aneurysms. This study highlights the spectrum of type IA endoleaks seen in clinical practice and their varying implications on patient outcomes. The study focused on immediate type IA endoleaks that were present on the initial angiogram after deployment and excluded endoleaks that appeared in a delayed fashion. Whereas early endoleaks are usually secondary to hostile neck anatomy or a poorly placed graft, those that appear during follow-up, the so-called late endoleaks, are the result of distal migration of the graft or aneurysmal degeneration at the proximal seal zone and are different clinical entities that tend to behave differently. Those leaks expose a sac that has been depressurized and likely weakened over time to a new level of aortic pressure. In contrast, with early endoleaks, the aneurysm sac has not yet been excluded from systemic aortic pressure and is less likely to continue to expand in the presence of a small gutter leak. The patients who left the operating room with persistent endoleaks were thought to have these lower pressure gutter leaks, or more aggressive interventions, including conversion to open, would have been considered. The fact that persistent type IA endoleaks in this study were not associated with aneurysm sac increase and that there was a high rate of spontaneous seal support this theory. Indeed, the current finding that certain persistent type IA endoleaks are benign should not be applied to those that occur in a delayed fashion or to those early endoleaks with high flow. Only the presence of a heavily calcified neck was associated with a persistent endoleak at case end. Surprisingly, having anatomy that is outside of the manufacturer’s IFU was not associated with failure of the endoleak to resolve, and neither was neck angle or neck length. This is consistent with previous reports that have failed to show any reproducible predictors of either type IA endoleaks or their failure to resolve
intraoperatively.14-21 Bastos Gonçalves et al8 reported that neck length and neck thrombus or calcifications predicted adverse neck events, whereas Tan et al14 found that age, female gender, large main body graft diameter, and unplanned graft extension were the only predictive factors. These data must be interpreted in the context of the study design. Although the current report is the largest to date, it is still a relatively small, single-center study. It is possible that there was not full capture of all ruptures or aneurysm-related mortality because the cause of death for some of the patients was unknown. With a single-institutional study, one can capture only the rupture or reinterventions that occurred within our center or were reported on follow-up, so the true reintervention and rupture rates are likely underestimated. However, most patients who have been treated at our institution will be sent back if there is an aorta-related issue. Finally, the goal of this study was to focus on anatomic factors that predict persistent endoleak, and we did not take patient factors such as anticoagulation into account. This is often difficult to control in a retrospective fashion as it is impossible to determine the compliance of patients with medications or the degree to which they were subtherapeutic or supratherapeutic on their medications during follow-up, and these data are more accurately followed in a prospective fashion.
CONCLUSIONS Type IA endoleaks are present in 10% of patients on the initial angiogram after deployment. Despite aggressive intervention, many of these will persist beyond the initial procedure. However, aneurysm rupture secondary to persistent type IA endoleak is rare, and most of these endoleaks will resolve within 1 year. These patients require close follow-up, and intervention should be reserved for patients who show continued sac growth on serial imaging. Extensive neck calcification is the only independent predictor of persistent type IA endoleak in this study, and patients with this anatomy should be approached cautiously. These data suggest that early persistent type IA endoleaks can be safely observed in most patients.
AUTHOR CONTRIBUTIONS Conception and design: TO, MRC, DC, RC, MFC Analysis and interpretation: TO, MRC, SD, DC, RC, MFC Data collection: TO, MRC, SD, GT, RM Writing the article: TO, MRC, SD, MFC Critical revision of the article: TO, MRC, SD, GT, RM, DC, RC, MFC Final approval of the article: TO, MRC, SD, GT, RM, DC, RC, MFC Statistical analysis: TO, SD, MFC Obtained funding: Not applicable Overall responsibility: MFC
Journal of Vascular Surgery
O’Donnell et al
125
Volume 67, Number 1
REFERENCES 1. Schermerhorn ML, Buck DB, O’Malley AJ, Curran T, McCallum JC, Darling J, et al. Long-term outcomes of abdominal aortic aneurysm in the Medicare population. N Engl J Med 2015;373:328-38. 2. Parodi JC, Palmaz JC, Barone HD. Transfemoral intraluminal graft implantation for abdominal aortic aneurysms. Ann Vasc Surg 1991;5:491-9. 3. Chaikof EL, Brewster DC, Dalman RL, Makaroun MS, Illig KA, Sicard GA, et al. The care of patients with an abdominal aortic aneurysm: the Society for Vascular Surgery practice guidelines. J Vasc Surg 2009;50:S2-49. 4. Harris PL, Vallabhaneni SR, Desgranges P, Becquemin JP, van Marrewijk C, Laheij RJ. Incidence and risk factors of late rupture, conversion, and death after endovascular repair of infrarenal aortic aneurysms: the EUROSTAR experience. European Collaborators on Stent/graft techniques for aortic aneurysm repair. J Vasc Surg 2000;32:739-49. 5. van Marrewijk C, Buth J, Harris PL, Norgren L, Nevelsteen A, Wyatt MG. Significance of endoleaks after endovascular repair of abdominal aortic aneurysms: the EUROSTAR experience. J Vasc Surg 2002;35:461-73. 6. Schurink GW, Aarts NJ, Wilde J, van Baalen JM, Chuter TA, Schultze Kool LJ, et al. Endoleakage after stent-graft treatment of abdominal aneurysm: implications on pressure and imagingdan in vitro study. J Vasc Surg 1998;28: 234-41. 7. Venermo MA, Arko FR, Salenius JP, Saarinen JP, Zvaigzne A, Zarins CK. EVAR may reduce the risk of aneurysm rupture despite persisting type IA endoleaks. J Endovasc Ther 2011;18:676-82. 8. Bastos Gonçalves F, Verhagen HJ, Vasanthananthan K, Zandvoort HJ, Moll FL, van Herwaarden JA. Spontaneous delayed sealing in selected patients with a primary type-Ia endoleak after endovascular aneurysm repair. Eur J Vasc Endovasc Surg 2014;48:53-9. 9. Candell L, Tucker LY, Goodney P, Walker J, Okuhn S, Hill B, et al. Early and delayed rupture after endovascular abdominal aortic aneurysm repair in a 10-year multicenter registry. J Vasc Surg 2014;60:1146-53. 10. Kim SM, Ra HD, Min SI, Jae HJ, Ha J, Min SK, et al. Clinical significance of type I endoleak on completion angiography. Ann Surg Treat Res 2014;86:95. 11. Bastos Gonçalves F, Hoeks SE, Teijink JA, Moll FL, Castro JA, Stolker RJ, et al. Risk factors for proximal neck complications
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
after endovascular aneurysm repair using the Endurant stentgraft. Eur J Vasc Endovasc Surg 2015;49:156-62. Oliveira NF, Bastos Gonçalves FM, de Vries JP, Ultee KH, Werson DA, Hoeks SE, et al. Mid-term results of EVAR in severe proximal aneurysm neck angulation. Eur J Vasc Endovasc Surg 2015;49:19-27. Millen AM, Osman K, Antoniou GA, McWilliams RG, Brennan JA, Fisher RK. Outcomes of persistent intraoperative type IA endoleak after standard endovascular aneurysm repair. J Vasc Surg 2015;61:1185-91. Tan TW, Eslami M, Rybin D, Doros G, Zhang WW, Farber A. Outcomes of patients with type I endoleak at completion of endovascular abdominal aneurysm repair. J Vasc Surg 2016;63:1420-7. Tzortzis E, Hinchliffe RJ, Hopkinson BR. Adjunctive procedures for the treatment of proximal type I endoleak: the role of peri-aortic ligatures and Palmaz stenting. J Endovasc Ther 2003;10:233-9. Schanzer A, Greenberg RK, Hevelone N, Robinson WP, Eslami MH, Goldberg RJ, et al. Predictors of abdominal aortic aneurysm sac enlargement after endovascular repair. Circulation 2011;123:2848-55. Boniakowski AE, De Martino RR, Coleman DM, Eliason JL, Goodney PP, Rectenwald JE. The natural history of type II endoleaks after endovascular aneurysm repair for ruptured abdominal aortic aneurysm. J Vasc Surg 2016;64:1645-51. Lal BK, Zhou W, Li Z, Kyriakides T, Matsumura J, Lederle FA, et al. Predictors and outcomes of endoleaks in the Veterans Affairs Open Versus Endovascular Repair (OVER) trial of abdominal aortic aneurysms. J Vasc Surg 2015;62: 1394-404. Lowe C, Abbas A, Rogers S, Smith L, Ghosh J, McCollum C. Three-dimensional contrast-enhanced ultrasound improves endoleak detection and classification after endovascular aneurysm repair. J Vasc Surg 2017;65:1453-9. Massimi TM, Kostun ZW, Woo EY. Transcaval embolization of a type I gutter endoleak after three-vessel chimney endovascular aneurysm repair. J Vasc Surg 2017;65:1515-7. Ullery BW, Tran K, Itoga NK, Dalman RL, Lee JT. Natural history of gutter-related type IA endoleaks after snorkel/ chimney endovascular aneurysm repair. J Vasc Surg 2017;65: 981-90.
Submitted Feb 27, 2017; accepted May 5, 2017.
Select early type IA endoleaks after endovascular aneurysm repair will resolve without secondary intervention Thomas F. X. O’Donnell, MD, Michael R. Corey, MD, Sarah E. Deery, MD, Gregory Tsougranis, BS, Rohit Maruthi, BS, W. Darrin Clouse, MD, Richard P. Cambria, MD, and Mark F. Conrad, MD, MMSc, Boston, Mass
ABSTRACT Objective: Although it is traditionally considered ominous, the natural history of early proximal attachment site endoleaks (IA) after endovascular aneurysm repair (EVAR) is not well known. Our aim was to identify risk factors for persistent type IA endoleaks and to determine their effect on long-term outcomes after EVAR. Methods: All patients who underwent infrarenal EVAR at a single institution between 1998 and 2015 were identified. Preoperative axial imaging and intraoperative arteriograms were reviewed, and those patients with a type IA endoleak were further studied. Aneurysm features were characterized by two reviewers and were studied for predictors of persistent endoleaks at the conclusion of the case. Patient records and the Social Security Death Index were used to record 1-year and overall survival. Results: We identified 1484 EVARs, 122 (8%) of which were complicated by a type IA endoleak on arteriography after graft deployment, with a median follow-up of 4 years. The majority of patients underwent additional ballooning of the proximal site (52 [43%]) or placement of an aortic cuff (47 [39%]); 30 patients (25%) received a Palmaz stent, and four patients were treated with coils or anchors. At case end, only 43 (35%) of the type IA endoleaks remained; at 1 month, only 16 endoleaks persisted (13%), and only six persisted at 1 year (6%). In multivariable analysis, the only independent predictor of persistence of type IA endoleak at the conclusion of the case was the presence of extensive neck calcifications (odds ratio [OR], 9.9; 95% confidence interval [CI], 1.4-67.9; P ¼ .02). Thirteen patients (11%) underwent reintervention for type IA endoleaks, with a time frame ranging from 3 days postoperatively to 11 years. There were three patients (2.4%) who experienced aneurysm rupture. Postoperative type IA endoleak was associated with lower survival at 1 year (79% vs 91%; relative risk, 2.5; 95% CI, 1.1-5.4; P ¼ .02), but it did not affect long-term survival (log-rank, P ¼ .45). Both an increase in aneurysm sac size and failure of the endoleak to resolve by case end were independent predictors of a need for reintervention (growth: OR, 8.3; 95% CI, 2.2-31.6; P < .01; persistent endoleak: OR, 7.6; 95% CI, 1.8-31.5; P < .01). A persistent type IA endoleak was not independently associated with an increase in sac size on surveillance imaging (P ¼ .28). Conclusions: Aneurysm rupture secondary to persistent type IA endoleak is rare, and most will resolve within 1 year. Extensive neck calcification is the only independent predictor of persistent type IA endoleak, and an increase in sac size warrants reintervention. These data suggest that select early persistent type IA endoleaks can be safely observed. (J Vasc Surg 2018;67:119-25.)
Endovascular aneurysm repair (EVAR) was first introduced in 1991 and by 2008 accounted for 77% of the repairs performed in Medicare patients.1,2 Type IA endoleaks occur when there is persistent flow into the aneurysm sac due to inadequate proximal seal and are seen
From the Division of Vascular and Endovascular Surgery, Massachusetts General Hospital. Author conflict of interest: none. Presented at the Twenty-seventh Annual Meeting of the Vascular and Endovascular Surgery Society, Steamboat, Colo, February 2-5, 2017. Correspondence: Mark F. Conrad, MD, MMSc, Massachusetts General Hospital, 15 Parkman St, Boston, MA 02114 (e-mail: [email protected]). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest. 0741-5214 Copyright Ó 2017 by the Society for Vascular Surgery. Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jvs.2017.05.096
on completion angiography in up to 11% of EVAR cases.3 Despite their prevalence, the risk factors for, natural history of, and prognosis of these endoleaks are not well characterized. The Society for Vascular Surgery recommends that “every attempt should be made to resolve type IA endoleaks before the patient leaves the intervention suite.”3 This recommendation is based on the belief that unlike in type II endoleaks, where there is persistent flow from patent branch vessels of the aorta, continued flow through the proximal seal zone is more likely to produce high pressure that can lead to sac expansion and places the patient at continued risk of rupture. Indeed, data from the European Collaborators on Stent/graft Techniques for aortic Aneurysm Repair (EUROSTAR) trial showed type IA endoleak to be a risk factor for rupture,4-6 and Venermo et al7 found that type IA endoleak was associated with a 3-year aneurysm-related mortality of 11%. 119
120
Journal of Vascular Surgery
O’Donnell et al
January 2018
It has been our practice to exhaust all options for resolving type IA endoleaks before leaving the operating room. Despite this effort, some will persist, and a decision must be made as to how the patient should be observed and whether conversion to open explantation is warranted to prevent rupture. Recent data suggest that type IA endoleaks may not carry such a dire prognosis as previously believed. Several studies have demonstrated low rates of rupture and aneurysm-related mortality and high rates of spontaneous sealing in select patients,8-14 but these studies are limited by small numbers and limited follow-up. The goal of this study was to examine our institutional experience with type IA endoleaks in an effort to document long-term outcomes.
METHODS Subjects. This is a retrospective review of all patients who underwent infrarenal EVAR at the Massachusetts General Hospital from 1998 to 2015. Aneurysms that were a consequence of chronic dissection were excluded, but ruptured aneurysms were included in the cohort. Patients were identified through the use of Current Procedural Terminology codes. The charts were then reviewed to confirm the procedures performed and to identify patients who had a type IA endoleak on the initial angiogram after deployment. The Massachusetts General Hospital Institutional Review Board approved the study and waived the need for patient consent, given the retrospective nature of the study. Aneurysm characteristics. The preoperative computed tomography (CT) angiograms, intraoperative angiograms, and postoperative CT angiograms were reviewed by two authors (T.F.X.O. and M.R.C.). Preoperative CT angiograms were used to calculate the maximal anteroposterior aneurysm diameter, diameter of the aorta at the lowest renal artery, length of the neck, infrarenal and suprarenal neck angles, proximal and distal neck diameter, presence of a conical neck (>5-mm difference in diameter of proximal neck and the neck at its widest point), and presence of extensive neck calcifications (as assessed qualitatively by the two reviewers, defined as calcifications involving >50% of the circumference). Intraoperative events. Intraoperative angiograms and operative reports were reviewed by two authors (T.F.X.O. and M.R.C.) to assess for the persistence of type IA endoleak at the close of the procedure as well as to record any supplementary intraoperative interventions undertaken to ensure proximal seal. Outcomes. The primary outcome was overall survival, which was recorded through patient records and the Social Security Death Index. Secondary outcomes included persistent endoleak at the conclusion of the
ARTICLE HIGHLIGHTS d
d
d
Type of Research: Retrospective, single-center cohort study Take Home Message: Forty-three patients with persistent type IA endoleaks were observed after endovascular aneurysm repair; 16 endoleaks persisted at 1 month and six at 1 year. The only predictor of the presence of a postprocedural type IA endoleak was neck calcification. There were three instances of abdominal aortic aneurysm rupture and 13 cases of reintervention. Postoperative type IA endoleak was associated with lower 1-year survival but not longer term survival or increase in sac size. Recommendation: This study suggests that some early type IA endoleaks can be observed but that sac size increase should prompt reintervention.
case, at 1 month, and at 1 year, as assessed by follow-up CT angiography, and reintervention or aneurysm rupture. Secondary outcomes were ascertained through chart review. Statistical analysis. Categorical variables were compared with the Fisher exact test. Continuous variables were analyzed with the Wilcoxon rank sum test. Univariate survival analysis was carried out with the logrank test. Logistic regression was performed to determine aneurysm characteristics that were independent predictors of persistent endoleak. Variables that were included in the logistic regression were age, procedure year, female sex, ruptured aneurysms, maximal anteroposterior aneurysm diameter, diameter of the aorta at the lowest renal artery, length of the neck, infrarenal and suprarenal neck angles, maximal neck diameter, presence of a conical neck (>5-mm difference in diameter of proximal neck and the neck at its widest point), and presence of extensive neck calcifications. We tested whether being outside the manufacturer’s instructions for use (IFU) with respect to the proximal landing zone was independently associated with type IA endoleak by constructing a separate logistic regression model that included all of the aforementioned variables except for those factors that are part of the IFU (neck length, neck angles, and neck diameter) to avoid collinearity. Cox proportional hazards models were constructed to examine the association between persistent endoleak and death, independent of age and aneurysm size at presentation. Age and aneurysm size were chosen a priori to be controlled for in multivariable analyses as these have previously been shown to be independent predictors of survival. All statistical analyses were carried out using Stata 14.2 software (StataCorp, College Station, Tex). Sensitivity analysis. Given that ruptured aneurysms may represent a category of patients with less than ideal
Journal of Vascular Surgery
O’Donnell et al
121
Volume 67, Number 1
Table I. Baseline characteristics Endoleak resolved (n ¼ 79) Characteristic
Overall cohort (N ¼ 122)
Persistent endoleak (n ¼ 43)
Median (interquartile range) or No. (%)
P value
Age
80 (74-85)
79 (73-85)
82 (78-85)
.07
Female sex
38 (31)
24 (30)
14 (33)
.84
Rupture
15 (12)
8 (10)
7 (16)
.39
Aneurysm size
58 (53-64)
58 (53-67)
60 (52-63)
.97
Diameter at lowest renal
23 (21-26)
23 (21-27)
24 (22-26)
.34
Largest neck diameter
26 (23-29)
25 (23-29)
26 (23-29)
.48
Neck length
28 (21-38)
31 (21-38)
26 (20-37)
.31
Suprarenal neck angle
10 (5-20)
10 (5-24)
10 (5-15)
.27
Infrarenal neck angle
20 (10-40)
20 (10-38)
22 (10-45)
.33
Conical neck
31 (26)
16 (21)
15 (36)
.09
Outside IFU
29 (24)
16 (21)
13 (31)
.27
Neck calcifications
20 (17)
6 (8)
14 (33)
<.01
IFU, Instructions for use.
anatomy, a sensitivity analysis was performed excluding those patients with rupture from the analysis.
RESULTS Patients. Between 1998 and 2015, there were 1484 EVARs performed, 122 (8%) of which were complicated by a type IA endoleak on angiography after deployment, with a median follow-up of 48 months (interquartile range, 22-78 months). Baseline characteristics and anatomic data are presented in Table I. Of note, 24% of the patients had anatomies that were outside the manufacturer’s IFU. However, in univariate analysis, the only factor associated with persistent endoleak was presence of extensive neck calcifications. By the close of the procedure, 35% of those endoleaks persisted (43/122). At 1-month follow-up CT angiography, only 13% had evidence of persistent type IA endoleak (16/119) despite only three reinterventions, and only 6% were evident at 1 year (6/96) despite only an additional five reinterventions (Fig 1). None of the patients whose endoleaks resolved in the operating room had an endoleak at 1 year. These results were similar when only intact aneurysms were considered (12% at 1 month, 7% at 1 year). Of the 43 patients who left the operating room with a persistent endoleak, 13 (30%) had a type IA endoleak at 1 month and 6 (14%) had evidence of endoleak at 1 year. These rates were also similar when only intact aneurysms were considered (28% at 1 month, 24% at 1 year). Intraoperative procedures. After a type IA endoleak was identified following graft deployment, all patients underwent additional maneuvers in attempts to resolve the leak (Table II). The most common adjunct was additional ballooning alone, which was performed in 52 (43%) patients with a 62% rate of successful resolution of the
endoleak. An additional 47 (39%) patients underwent placement of an aortic cuff, and 30 (25%) patients received a Palmaz stent, with success rates of 70% and 53%, respectively. Three patients underwent anchor placement without success. None of the therapies were associated with endoleak resolution in univariate analysis, except anchors, which were associated with failure to resolve (relative risk, 2.98; 95% CI, 2.3-3.8; P ¼ .02). Predictors of persistent endoleak. In multivariable analysis, the only independent predictor of persistence of type IA endoleak at the conclusion of the case was the presence of extensive neck calcifications (odds ratio [OR], 11.5; 95% confidence interval [CI], 1.7-76.0; P ¼ .01; Table III). No other anatomic features were predictive of failure to resolve, including being outside the IFU (OR, 0.71; 95% CI, 0.17-3.0; P ¼ .64). Excluding ruptured aneurysms did not change the results. The proportion of endoleaks that persisted at the end of the case decreased over time (P < .01; Fig 2). Survival. At 1 year, 91% of those whose endoleak resolved by the end of the case were alive compared with 79% of those who left the operating room with a persistent endoleak (relative risk for death, 2.45; 95% CI, 1.12-5.36; P ¼ .02). After adjustment for age and aneurysm size, leaving the operating room with a type IA endoleak remained a significant risk factor for mortality at 1 year (OR, 3.79; 95% CI, 1.28-11.24; P ¼ .02). Despite the association with 1-year survival, there was no difference in overall survival in univariate analysis between patients who left the operating room with a persistent type IA endoleak and those who did not (log-rank, P ¼ .26; Fig 3). Five-year survival was 56% for those patients who left the operating room without an endoleak compared with 52% for those who left with
122
Journal of Vascular Surgery
O’Donnell et al
January 2018
Fig 1. Evolution of intraoperative endoleaks during 1-year follow-up. Interventions to the left of the arrows refer to any reintervention that occurred during the time interval. EVARs, Endovascular aneurysm repairs.
Table II. Intraoperative adjunctive procedures Procedure
No. (% of total)
Resolution achieved, %
Ballooning alone
52 (43)
62
Palmaz stent
30 (25)
53
Cuff
47 (39)
70
Anchors
3 (2.5)
0a
Coils
1 (1)
0
Significant at P < .05.
a
Table III. Logistic regression model for failure of type IA endoleak to resolve in the operating room Characteristic
OR
95% CI
P
Age
0.93
0.85-1.02
.12
Female sex
0.97
0.24-3.93
.97
1.04
0.98-1.10
.19
0.93
0.66-1.31
.68
1.00
0.74-1.37
.99
Aneurysm size Diameter at lowest renal Largest neck diameter Neck calcifications
10.36
1.54-69.86
.02
Neck length
1.02
0.98-1.07
.34
Suprarenal neck angle
1.05
0.99-1.13
.13
1.00
0.97-1.04
.83
Infrarenal neck angle Conical neck
0.18
0.02-1.96
.16
Ruptured aneurysm
0.30
0.05-1.95
.21
Outside IFU
0.38
0.04-3.41
.39
CI, Confidence interval; IFU, instructions for use; OR, odds ratio.
an endoleak. This lack of association remained after adjustment for age and aneurysm size (hazard ratio, 0.64; 95% CI, 0.38-1.07; P ¼ .09). Excluding those patients with ruptures did not change the results, as persistent endoleak remained
associated with 1-year survival (P ¼ .47) but not overall survival (P ¼ .5). Reintervention and late rupture. There were 13 (11%) patients who underwent at least one reintervention, with a median time to intervention of 17 months (range, 3 days-11 years) from the date of the initial operation. There were three ruptures (2.4% of the overall cohort, 7.0% of the patients with persistent endoleaks), two from type IA endoleaks (1.6% overall, 4.7% of persistent endoleaks). One patient was observed for 3 years with a type IA endoleak. Her sac continued to enlarge but she refused intervention until she presented with a rupture 3 years after EVAR. She was successfully treated with an open repair and survived an additional 3 years. Another patient was taken back to the operating room on postoperative day 3 for attempted endovascular repair of a large type IA endoleak after the patient and family refused open intervention. This was unsuccessful, and she was discharged to hospice care where she subsequently died, presumably from rupture, but no autopsy was performed. The last rupture was secondary to a type III endoleak (component separation) at 9 months and was successfully treated with repeated EVAR. There were two open operations, one of which was the rupture at 3 years; the second was an open juxtarenal aneurysm repair for aneurysmal degeneration at the proximal extent. The other 11 patients were all treated with endovascular interventions, mostly cuffs. Both an increase in size of the aneurysm sac and failure of the endoleak to resolve by the case end were independent predictors of a need for reintervention (growth: OR, 8.3; 95% CI, 2.2-31.6; P < .01; persistent endoleak: OR, 7.6; 95% CI, 1.8-31.5; P < .01). However, a persistent endoleak was not associated with an increase in sac size on surveillance imaging (P ¼ .28).
Journal of Vascular Surgery
O’Donnell et al
123
Volume 67, Number 1
Fig 2. Rate of persistent endoleak at case end over time.
Fig 3. Overall survival, by presence of persistent endoleak. SE, Standard error.
DISCUSSION Type IA endoleaks are present on the initial angiogram after deployment in up to 10% of patients, and despite additional adjunctive procedures, 35% of patients in this study had a persistent endoleak at the end of the procedure. A persistent type IA endoleak has traditionally been considered an absolute indication for reintervention, including conversion to open aneurysm repair to prevent rupture. However, this study demonstrates that many type IA endoleaks that were persistent at the end of the procedure will spontaneously resolve with reversal of systemic anticoagulation and can often be observed with close surveillance with a low rate (4.7%) of subsequent rupture. Similar results have been seen in several small series in which >50% of persistent type IA endoleaks have resolved by the first follow-up CT scan.7,8,10,13 With longer follow-up, only 14% of the endoleaks present at the case end were still present at 1 year. More important, there was no difference in overall survival in those patients who left the operating room with a persistent endoleak compared with those whose endoleaks resolved. This is consistent with the findings of Bastos Gonçalves et al,8 who
followed up 15 patients with persistent type IA endoleaks and found that all had resolved by 1 year after the initial procedure, and Millen et al,13 who studied 33 persistent endoleaks and found that 31 (94%) resolved by the first postoperative scan. The ultimate goal of elective abdominal aortic aneurysm repair regardless of approach (open vs EVAR) is to prevent aneurysm rupture and subsequent death. Consequently, any aneurysm-related event after EVAR should be considered a failure of the procedure. It is reassuring that in this series, there were only two ruptures attributable to type IA endoleaks, both of which were potentially avoidable as both patients had refused further intervention before their ruptures. One patient presented with an early rupture. She was considered a poor open candidate, and her neck anatomy was not favorable for endovascular repair. Despite aggressive attempts to obtain adequate seal in the proximal zone, there was a persistent, large endoleak in a large aneurysm. She refused open conversion and was discharged to hospice, where she died a few days later, presumably of rupture. The second rupture occurred after 3 years, during which time the patient had a known endoleak with expanding aortic sac but refused any further intervention. Similar to our experience, a fatal early rupture at 2 days postoperatively was reported by Bastos Gonçalves et al,8 for which an open repair was attempted but was unsuccessful. In addition to the outcomes after type IA endoleaks, the nature and frequency of reinterventions have not been well defined. Previous studies suggested that most reoperations are performed open, but they suffered from small sample sizes and limited follow-up, with very few reoperations captured. Bastos Gonçalves et al8 reported six reinterventions of 15 endoleaks, three of which were performed open and three endovascularly. Kim et al10 performed two reoperations on 10 endoleaks; again, one was open and one endovascular. Tan et al14 studied 80 endoleaks and found three perioperative conversions to open, and Millen et al13 studied 33 persistent endoleaks and found that only one required Onyx (ev3, Irvine, Calif). In our study, most of the reinterventions were performed endovascularly, with only three open operations of 11. The open operations were all traditional open aneurysm resections, with graft explantation. Endovascular approaches ranged from aortic cuffs to Palmaz stents, redo EVARs, Onyx glue, Aptus anchors (Aptus Endosystems, Sunnyvale, Calif), and one hybrid operation with an aortouni-iliac limb with a femorofemoral bypass. Tzortzis et al15 described 22 endoleaks, 15 of which were treated with periaortic banding around the proximal zone, but this technique was not used in our series. Of note is the substantial improvement in the rate of persistent endoleak at case end over time. This is likely due to improvements in grafts over time as well as to the surgeon’s technique and experience with the procedure.
124
Journal of Vascular Surgery
O’Donnell et al
January 2018
Although the rates of rupture and reintervention in this study were low, patients with persistent type IA endoleaks were observed with a surveillance protocol that was more aggressive than our standard imaging followup. Most patients who left the operating room with a type IA endoleak underwent CT angiography before discharge. This was repeated in 4 to 6 weeks, and if the endoleak did not resolve, patients were imaged at 6 months and 1 year after the initial procedure. Patients with persistent endoleaks as well as those who continue to show aneurysm sac growth should be considered for a more aggressive imaging schedule as these factors independently predicted the need for reintervention in this study. As these endoleaks frequently resolve spontaneously, growth of the aneurysm sac captured on surveillance imaging, rather than the mere presence of an endoleak, should be the impetus for reintervention. Indeed, Venermo et al7 found that type IA endoleak was associated with lower rates of rupture than in patients with untreated abdominal aortic aneurysms. This study highlights the spectrum of type IA endoleaks seen in clinical practice and their varying implications on patient outcomes. The study focused on immediate type IA endoleaks that were present on the initial angiogram after deployment and excluded endoleaks that appeared in a delayed fashion. Whereas early endoleaks are usually secondary to hostile neck anatomy or a poorly placed graft, those that appear during follow-up, the so-called late endoleaks, are the result of distal migration of the graft or aneurysmal degeneration at the proximal seal zone and are different clinical entities that tend to behave differently. Those leaks expose a sac that has been depressurized and likely weakened over time to a new level of aortic pressure. In contrast, with early endoleaks, the aneurysm sac has not yet been excluded from systemic aortic pressure and is less likely to continue to expand in the presence of a small gutter leak. The patients who left the operating room with persistent endoleaks were thought to have these lower pressure gutter leaks, or more aggressive interventions, including conversion to open, would have been considered. The fact that persistent type IA endoleaks in this study were not associated with aneurysm sac increase and that there was a high rate of spontaneous seal support this theory. Indeed, the current finding that certain persistent type IA endoleaks are benign should not be applied to those that occur in a delayed fashion or to those early endoleaks with high flow. Only the presence of a heavily calcified neck was associated with a persistent endoleak at case end. Surprisingly, having anatomy that is outside of the manufacturer’s IFU was not associated with failure of the endoleak to resolve, and neither was neck angle or neck length. This is consistent with previous reports that have failed to show any reproducible predictors of either type IA endoleaks or their failure to resolve
intraoperatively.14-21 Bastos Gonçalves et al8 reported that neck length and neck thrombus or calcifications predicted adverse neck events, whereas Tan et al14 found that age, female gender, large main body graft diameter, and unplanned graft extension were the only predictive factors. These data must be interpreted in the context of the study design. Although the current report is the largest to date, it is still a relatively small, single-center study. It is possible that there was not full capture of all ruptures or aneurysm-related mortality because the cause of death for some of the patients was unknown. With a single-institutional study, one can capture only the rupture or reinterventions that occurred within our center or were reported on follow-up, so the true reintervention and rupture rates are likely underestimated. However, most patients who have been treated at our institution will be sent back if there is an aorta-related issue. Finally, the goal of this study was to focus on anatomic factors that predict persistent endoleak, and we did not take patient factors such as anticoagulation into account. This is often difficult to control in a retrospective fashion as it is impossible to determine the compliance of patients with medications or the degree to which they were subtherapeutic or supratherapeutic on their medications during follow-up, and these data are more accurately followed in a prospective fashion.
CONCLUSIONS Type IA endoleaks are present in 10% of patients on the initial angiogram after deployment. Despite aggressive intervention, many of these will persist beyond the initial procedure. However, aneurysm rupture secondary to persistent type IA endoleak is rare, and most of these endoleaks will resolve within 1 year. These patients require close follow-up, and intervention should be reserved for patients who show continued sac growth on serial imaging. Extensive neck calcification is the only independent predictor of persistent type IA endoleak in this study, and patients with this anatomy should be approached cautiously. These data suggest that early persistent type IA endoleaks can be safely observed in most patients.
AUTHOR CONTRIBUTIONS Conception and design: TO, MRC, DC, RC, MFC Analysis and interpretation: TO, MRC, SD, DC, RC, MFC Data collection: TO, MRC, SD, GT, RM Writing the article: TO, MRC, SD, MFC Critical revision of the article: TO, MRC, SD, GT, RM, DC, RC, MFC Final approval of the article: TO, MRC, SD, GT, RM, DC, RC, MFC Statistical analysis: TO, SD, MFC Obtained funding: Not applicable Overall responsibility: MFC
Journal of Vascular Surgery
O’Donnell et al
125
Volume 67, Number 1
REFERENCES 1. Schermerhorn ML, Buck DB, O’Malley AJ, Curran T, McCallum JC, Darling J, et al. Long-term outcomes of abdominal aortic aneurysm in the Medicare population. N Engl J Med 2015;373:328-38. 2. Parodi JC, Palmaz JC, Barone HD. Transfemoral intraluminal graft implantation for abdominal aortic aneurysms. Ann Vasc Surg 1991;5:491-9. 3. Chaikof EL, Brewster DC, Dalman RL, Makaroun MS, Illig KA, Sicard GA, et al. The care of patients with an abdominal aortic aneurysm: the Society for Vascular Surgery practice guidelines. J Vasc Surg 2009;50:S2-49. 4. Harris PL, Vallabhaneni SR, Desgranges P, Becquemin JP, van Marrewijk C, Laheij RJ. Incidence and risk factors of late rupture, conversion, and death after endovascular repair of infrarenal aortic aneurysms: the EUROSTAR experience. European Collaborators on Stent/graft techniques for aortic aneurysm repair. J Vasc Surg 2000;32:739-49. 5. van Marrewijk C, Buth J, Harris PL, Norgren L, Nevelsteen A, Wyatt MG. Significance of endoleaks after endovascular repair of abdominal aortic aneurysms: the EUROSTAR experience. J Vasc Surg 2002;35:461-73. 6. Schurink GW, Aarts NJ, Wilde J, van Baalen JM, Chuter TA, Schultze Kool LJ, et al. Endoleakage after stent-graft treatment of abdominal aneurysm: implications on pressure and imagingdan in vitro study. J Vasc Surg 1998;28: 234-41. 7. Venermo MA, Arko FR, Salenius JP, Saarinen JP, Zvaigzne A, Zarins CK. EVAR may reduce the risk of aneurysm rupture despite persisting type IA endoleaks. J Endovasc Ther 2011;18:676-82. 8. Bastos Gonçalves F, Verhagen HJ, Vasanthananthan K, Zandvoort HJ, Moll FL, van Herwaarden JA. Spontaneous delayed sealing in selected patients with a primary type-Ia endoleak after endovascular aneurysm repair. Eur J Vasc Endovasc Surg 2014;48:53-9. 9. Candell L, Tucker LY, Goodney P, Walker J, Okuhn S, Hill B, et al. Early and delayed rupture after endovascular abdominal aortic aneurysm repair in a 10-year multicenter registry. J Vasc Surg 2014;60:1146-53. 10. Kim SM, Ra HD, Min SI, Jae HJ, Ha J, Min SK, et al. Clinical significance of type I endoleak on completion angiography. Ann Surg Treat Res 2014;86:95. 11. Bastos Gonçalves F, Hoeks SE, Teijink JA, Moll FL, Castro JA, Stolker RJ, et al. Risk factors for proximal neck complications
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
after endovascular aneurysm repair using the Endurant stentgraft. Eur J Vasc Endovasc Surg 2015;49:156-62. Oliveira NF, Bastos Gonçalves FM, de Vries JP, Ultee KH, Werson DA, Hoeks SE, et al. Mid-term results of EVAR in severe proximal aneurysm neck angulation. Eur J Vasc Endovasc Surg 2015;49:19-27. Millen AM, Osman K, Antoniou GA, McWilliams RG, Brennan JA, Fisher RK. Outcomes of persistent intraoperative type IA endoleak after standard endovascular aneurysm repair. J Vasc Surg 2015;61:1185-91. Tan TW, Eslami M, Rybin D, Doros G, Zhang WW, Farber A. Outcomes of patients with type I endoleak at completion of endovascular abdominal aneurysm repair. J Vasc Surg 2016;63:1420-7. Tzortzis E, Hinchliffe RJ, Hopkinson BR. Adjunctive procedures for the treatment of proximal type I endoleak: the role of peri-aortic ligatures and Palmaz stenting. J Endovasc Ther 2003;10:233-9. Schanzer A, Greenberg RK, Hevelone N, Robinson WP, Eslami MH, Goldberg RJ, et al. Predictors of abdominal aortic aneurysm sac enlargement after endovascular repair. Circulation 2011;123:2848-55. Boniakowski AE, De Martino RR, Coleman DM, Eliason JL, Goodney PP, Rectenwald JE. The natural history of type II endoleaks after endovascular aneurysm repair for ruptured abdominal aortic aneurysm. J Vasc Surg 2016;64:1645-51. Lal BK, Zhou W, Li Z, Kyriakides T, Matsumura J, Lederle FA, et al. Predictors and outcomes of endoleaks in the Veterans Affairs Open Versus Endovascular Repair (OVER) trial of abdominal aortic aneurysms. J Vasc Surg 2015;62: 1394-404. Lowe C, Abbas A, Rogers S, Smith L, Ghosh J, McCollum C. Three-dimensional contrast-enhanced ultrasound improves endoleak detection and classification after endovascular aneurysm repair. J Vasc Surg 2017;65:1453-9. Massimi TM, Kostun ZW, Woo EY. Transcaval embolization of a type I gutter endoleak after three-vessel chimney endovascular aneurysm repair. J Vasc Surg 2017;65:1515-7. Ullery BW, Tran K, Itoga NK, Dalman RL, Lee JT. Natural history of gutter-related type IA endoleaks after snorkel/ chimney endovascular aneurysm repair. J Vasc Surg 2017;65: 981-90.
Submitted Feb 27, 2017; accepted May 5, 2017.