- Email: [email protected]
Preoperative variables predict persistent type 2 endoleak after endovascular aneurysm repair
From the New England Society for Vascular Surgery
Preoperative variables predict persistent type 2 endoleak after endovascular aneurysm repair Christopher J. Abularrage, MD,a Robert S. Crawford, MD,a Mark F. Conrad, MD, MMSc,a Hang Lee, PhD,b Christopher J. Kwolek, MD,a David C. Brewster, MD,a Richard P. Cambria, MD,a and Glenn M. LaMuraglia, MD,a Boston, Mass Objective: Persistent type 2 endoleaks (PT2, present >6 months) after endovascular aneurysm repair (EVAR) are associated with adverse outcomes. This study evaluated the preoperative risk factors and natural history of PT2 in order to define a population at high risk. Methods: From January 1999 to December 2007, 595 of 832 EVAR patients had long-term computed tomography follow-up and comprised the study cohort. Preoperative anatomic and clinical variables were correlated with PT2 using Cox regression. Composite hazard ratios (HRs) were constructed with clusters of high-risk preoperative variables. Primary end points, including spontaneous resolution, sac enlargement >5 mm, and freedom from reintervention, were evaluated using Kaplan-Meier analysis. Results: There were 136 PT2 patients (23%) with a median follow-up of 34.8 months (range, 6.4-121.2 months). Positive predictive factors included patent inferior mesenteric artery (IMA; HR, 4.00; 95% confidence interval [CI], 1.62-9.90; P ⴝ .003), increasing number of patent lumbar arteries (HR, 1.24; 95% CI, 1.10-1.41; P ⴝ .0006), increasing age (HR, 1.04; 95% CI, 1.01-1.06; P ⴝ .005), and increasing luminal diameter on CT-contrast opacified lumen (HR, 1.03; 95% CI, 1.02-1.05; P ⴝ .0001). During follow-up, spontaneous PT2 resolution occurred in 34 patients (25%), sac diameter remained stable in 63 (46%), and rupture occurred in 2 (1.5%). Kaplan-Meier analysis estimated that 35.2% ⴞ 5.6% (95% CI, 23.8%-46.2%) of PT2 resolve spontaneously at 5 years after the index procedure. Freedom from sac enlargement >5 mm was 54.6% ⴞ 7.2% (95% CI, 40.6%-69.4%) at 5 years. Fifty-nine reinterventions were performed in 39 patients with PT2. Freedom from reintervention was 67.3% ⴞ 5.0% (95% CI, 57.0%-77.0%) at 5 years. The combination of a patent IMA and one risk factor of more than six patent lumbar arteries, maximum luminal diameter >30 mm, or age >70 years increased the odds of PT2 approximately ninefold. The combination of a patent IMA and any two risk factors increased the odds of PT2 approximately 18-fold. Conclusions: Several readily identifiable preoperative variables are associated with PT2 whose natural history was benign in but 35% of patients. On the basis of the composite high-risk HRs, there is accordingly a cohort of patients in whom perioperative interventions to preclude PT2 should be considered. ( J Vasc Surg 2010;52:19-24.)
Endovascular abdominal aortic aneurysm repair (EVAR) has become mainstream therapy for abdominal aortic aneurysms (AAAs) with suitable anatomy. EVAR has been associated with a 65% to 70% reduction in the 30-day mortality compared with open AAA repair.1,2 Clinically relevant late outcomes also have now been documented as favorable with EVAR.3 EVAR requires continued surveillance because up to 11% of patients need reintervention for graft-related comFrom the Division of Vascular and Endovascular Surgery,a and Biostatistics Center,b Massachusetts General Hospital and Harvard Medical School. This study was supported in part by grants from the Monte and Rita Goldman Foundation, Mr John F. Murphy, and the Bay State Federal Savings Foundation. Competition of interest: Dr. Richard Cambria has a consulting agreement with W.L. Gore. Presented at the Thirty sixth Annual Meeting of the New England Society for Vascular Surgery, Boston, Mass, Oct 2-4, 2009. Reprint requests: Glenn M. LaMuraglia, MD, Division of Vascular and Endovascular Surgery, Massachusetts General Hospital, WAC440, 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 competition of interest. 0741-5214/$36.00 Copyright © 2010 by the Society for Vascular Surgery. doi:10.1016/j.jvs.2010.02.023
plications, in particular endoleak.4 The presence of type 1 and 3 endoleaks is a clear indication for reintervention, whereas the clinical significance of type 2 endoleaks is not well established. Type 2 endoleaks occur in 10% to 30% of patients after EVAR and may be associated with aneurysm growth and rupture.5,6 Subdivision of type 2 endoleaks into transient (resolving ⱕ6 months) and persistent (present ⬎6 months) has been found to predict EVARrelated complications. Persistent type 2 endoleaks (PT2) have been associated with an increased incidence of adverse outcomes, including aneurysm sac growth, reintervention rate, the need for conversion to open repair, and rupture.7 Accordingly, there has been enthusiasm in some reports for preemptive adjunctive procedures, such as inferior mesenteric artery (IMA) coil embolization or intrasac thrombin injection, to preclude type 2 endoleaks.8,9 This study evaluated the preoperative risk factors and natural history of PT2 during a 10-year period at a single academic medical center to better define a population at high-risk for such lesions postoperatively. METHODS This study was approved by the Institutional Review Board of the Massachusetts General Hospital. 19
JOURNAL OF VASCULAR SURGERY July 2010
20 Abularrage et al
Data collection. All EVARs performed between January 1999 and December 2007 were retrospectively identified from a prospectively maintained database. Exclusion criteria included patients with thoracic aortic aneurysms, anastomotic aneurysms, isolated iliac artery aneurysms, and ⬍6 months of follow-up. Preoperative data collection was obtained from the patient hospital and office records and included demographics, medical history, and laboratory results. Estimated glomerular filtration rate (GFR) was calculated according to the Modification of Diet in Renal Disease Study.10 Chronic renal insufficiency (CRI) was defined as a GFR ⬍60 mL/min/1.73 m2, or chronic kidney disease stages 3 to 5.11 All preoperative and postoperative computed tomography (CT) scans were reviewed by an attending radiologist, an attending vascular surgeon, and a vascular fellow. Patients underwent preoperative CT imaging with 2.0- to 2.5-mm cuts. Preoperative anatomic variables included maximum aneurysm diameter, maximum aneurysm luminal diameter, patency of the IMA, number of patent lumbar arteries, occlusion of internal iliac vessels (chronic or by preoperative embolization), and percentage of aneurysm sac thrombosis. The maximum aneurysm luminal diameter was defined as the shortest diagonal of the contrastopacified lumen measured on the same cross-sectional CT image with the maximum aneurysm diameter. Thrombus load was defined as the area of the aneurysm sac measured on the same cross-sectional CT image with the maximum aneurysm diameter minus the maximum aneurysm luminal area. The percentage of sac thrombosis was calculated by the method of Sampaio et al12 and defined as the ratio of the area of the thrombus load to that of the aneurysm sac. All patients in the study received a follow-up CT scan with intravenous contrast and thin collimation ⱖ6 months postoperatively. After a nonenhanced CT scan was performed, a bolus injection of contrast was administered at 4 mL/s with a 25-second preparation delay. Delayed-phase images were obtained and reconstructed with 2.0- to 2.5-mm cuts. The presence of a type 2 endoleak was determined by the radiologist or surgeon, or both, and was corroborated in a blind fashion by a study vascular surgeon, or fellow, or both. Arteriography was performed when it was not possible to differentiate between different types of endoleak. PT2 was defined as any type 2 endoleak present ⱖ6 months postoperatively and included late-appearing type 2 endoleaks that were not present before 6 months. Further follow-up imaging with CT angiography was typically obtained at 6-month to 1-year intervals. The decision to intervene and the method of treatment were not uniform due to the different methods of treatment available during the 10-year period. Data analysis. Univariate comparisons of patient demographic and risk factors were performed using the CoxMantel test of the Kaplan-Meier survival estimates. All variables with P ⬍ .10 and with five or more patients were included in multivariate Cox regression analysis. Composite hazard ratio (HR) high-risk analysis was performed first
by converting continuous independent variables to binary variables. This was accomplished by choosing multiple points above and below the mean and selecting the point with the lowest P value when comparing patients with and without PT2. Variables with P ⬍ .05 on Cox regression were combined in different permutations, and adjusted HRs and 95% confidence intervals (CIs) were calculated. Only those composite high-risk variables with an HR greater than all single variables were reported. All values of P ⬍ .05 (two-tailed) were considered significant. Analyses were performed using StatView software (SAS Inc, Cary, NC). RESULTS During the study period, EVAR was performed in 832 patients. Four stent graft designs were included in this study: AneuRx (Medtronic, Minneapolis, Minn), Excluder (W. L. Gore & Associates, Flagstaff, Ariz), Powerlink (Endologix, Irvine, Calif), and Zenith (Cook, Bloomington, Ind). Overall results from our EVAR experience have been published previously.3 The study group consisted of 595 patients who underwent a follow-up CT scan ⱖ6 months postoperatively and 47 had an initial transient type 2 endoleak that was not detected ⬎6 months. PT2 was identified in 136 patients (23%) with a median follow-up of 34.8 months (range, 6.4-121.2 months), and a PT2 was discovered in 62 of these patients that was not detected on the initial postoperative CT scan. Preoperative patient characteristics and anatomic variables are summarized in Table I. On univariate Cox-Mantel test of the Kaplan-Meier estimates (Table II), a PT2 was more likely to develop in older patients (P ⫽ .007) and those taking warfarin (P ⫽ .03) but was less likely to develop in those who smoked tobacco (P ⫽ .03). A CoxMantel test of the preoperative CT imaging Kaplan-Meier estimates showed a PT2 was more likely to develop in patients with a patent IMA (P ⬍ .0001), an increased number of patent lumbar arteries (P ⬍ .0001), increased maximum aneurysm luminal diameter (P ⬍ .0001), and a decreased percentage of aneurysm sac thrombosis (P ⬍ .0001). No other significant anatomic or preoperative variables predicted the development of a PT2 on univariate analysis. Factors predictive of PT2 on Cox regression analysis are summarized in Table III. These included patent IMA (HR, 4.00; 95% CI, 1.62-9.90; P ⫽ .003), increasing number of patent lumbar arteries (HR, 1.24; 95% CI, 1.10-1.41; P ⫽ .0006), increasing age (HR, 1.04; 95% CI, 1.01-1.06; P ⫽ .005), and increasing maximum aneurysm luminal diameter (HR, 1.03; 95% CI, 1.02-1.05; P ⫽ .0001). Decreasing percentage of sac thrombosis was also predictive of PT2 but was not included in the final model due to a high degree of correlation with increasing maximum aneurysm luminal diameter (correlation coefficient, – 0.81). Overall, spontaneous PT2 resolution occurred in 34 patients (25%) at a mean of 32 months, and rupture occurred in 2 (1.5%). On Cox regression analysis, only a
JOURNAL OF VASCULAR SURGERY Volume 52, Number 1
Abularrage et al 21
Table I. Demographic and anatomic variables of patients with and without persistent type 2 endoleaks Variablea Patients Preoperative comorbidities Age Sex Male Female Any tobacco use Peripheral vascular disease History of deep venous thrombosis Hypercholesterolemia Coronary artery disease Cerebrovascular accident Chronic renal insufficiencyb Chronic obstructive pulmonary disease Diabetes mellitus Hypertension Congestive heart failure Medications Warfarin Statin Clopidogrel Aspirin ACE inhibitor Steroids -blocker Preoperative anatomic characteristics Patent inferior mesenteric artery Patent lumbar arteries, No. Max aneurysm luminal diameter, mm Sac thrombosis percentage Hypogastric coil embolization or occlusion Maximum aneurysm diameter, mm
PT2 endoleak
No PT2 endoleak
136
459
77.21 ⫾ 0.58 75.16 ⫾ 0.37 117 (86) 19 (14) 89 (65) 11 (8)
369 (80) 90 (20) 346 (75) 75 (16)
7 (5) 73 (54) 63 (46) 11 (8) 55 (40)
15 (3) 268 (58) 227 (49) 47 (10) 67 (15)
26 (19) 15 (11) 100 (74) 15 (11)
106 (23) 62 (14) 342 (75) 59 (13)
21 (15) 64 (47) 9 (7) 57 (42) 34 (25) 11 (8) 62 (46)
49 (11) 250 (54) 21 (5) 221 (48) 112 (24) 45 (10) 218 (48)
127 (93) 6.39 ⫾ 0.15
347 (76) 5.72 ⫾ 0.08
37.87 ⫾ 0.82 32.50 ⫾ 0.47 45.29 ⫾ 2.14 55.96 ⫾ 1.14 21 (15)
87 (19)
Table II. Univariate comparisons of Kaplan-Meier estimates of event time distributions for predictors of persistent type 2 endoleak Predictora
P value
Preoperative comorbidities Age, y Any tobacco use Peripheral vascular disease History of deep venous thrombosis Sex Hypercholesterolemia Coronary artery disease Cerebrovascular accident Chronic renal insufficiencyb Chronic obstructive pulmonary disease Diabetes mellitus Hypertension Congestive heart failure Medications Warfarin Statin Clopidogrel Aspirin ACE inhibitor Steroids -blocker Preoperative anatomic characteristics Patent inferior mesenteric artery Patent lumbar arteries, No. Maximum aneurysm luminal diameter, mm Sac thrombosis percentage Maximum aneurysm diameter, mm Hypogastric coil embolization or occlusion
history of chronic obstructive pulmonary disease predicted spontaneous resolution (HR, 2.28; 95% CI, 1.07-4.85; P ⫽ .03). Reinterventions were not done in 63 patients (46%) who had continued PT2 with stable sac size and were monitored with serial CT scans. During follow-up, KaplanMeier analysis estimated that 35.2% ⫾ 5.6% (95% CI, 23.8-46.2) of PT2 resolve spontaneously (Fig 1), and freedom from sac enlargement ⬎5 mm was 54.6% ⫾ 7.2% (95% CI, 40.6-69.4) at 5 years (Fig 2). On Cox regression analysis, only increasing age predicted sac enlargement ⬎5 mm (HR, 1.06; 95% CI, 1.00-1.12; P ⫽ .048). The average change in aneurysm sac size was –2.7 ⫾ 1.5 mm in patients with spontaneous resolution of the PT2, –1.7 ⫾ 0.7 mm in patients with continued PT2 not undergoing reintervention, 3.9 ⫾ 1.6 mm in patients requiring a reintervention, and 9.0 ⫾ 4.0 mm in patients with aneurysm rupture.
.03 .16 .20 .27 .51 .75 .99 ⬍.0001 ⬍.0001 ⬍.0001 ⬍.0001 .23 .29
ACE, Angiotensin-converting enzyme. a Continuous variables were measured as mean ⫾ standard error of the mean. b Defined as a glomerular filtration rate of ⬍60 mL/min/1.73 m2.
Table III. Multivariate Cox regression analysis for predictors of persistent type 2 endoleak
52.90 ⫾ 0.66 51.83 ⫾ 0.45
ACE, Angiotensin-converting enzyme inhibitor; PT2, persistent type 2. a Continuous data are presented as mean ⫾ standard error of the mean; categoric data as number (%). b Defined as a glomerular filtration rate of ⬍60 mL/min/1.73 m2.
.007 .03 .07 .12 .21 .36 .41 .42 .42 .44 .48 .86 .96
Predictor Patent inferior mesenteric artery Increasing number of patent lumbar arteries Increasing age, y Increasing MALD, mm
HR (95% CI)
P value
4.00 (1.62-9.90)
.003
1.24 (1.10-1.41) 1.04 (1.01-1.06) 1.03 (1.02-1.05)
.0006 .005 .0001
CI, Confidence interval; HR, hazard ratio; MALD, maximum aneurysm luminal diameter.
Fifty-nine reinterventions were performed in 39 patients with PT2. Freedom from reintervention was 67.3% ⫾ 5.0% (95% CI, 57.0-77.0) at 5 years (Fig 3). Reinterventions included 14 IMA coil embolizations, 14 sac coil embolizations, 10 glue injections with or without coil embolizations, 6 lumbar coil embolizations, 2 open ligation of bleeding vessels, and 12 failed attempts (inability to access the aneurysm sac). One patient refused an endovascular attempt at PT2 treatment and requested a graft explant, which was performed.
JOURNAL OF VASCULAR SURGERY July 2010
22 Abularrage et al
Fig 1. Kaplan-Meier survival plot shows time to spontaneous resolution in patients with persistent type 2 (PT2) endoleak after endovascular aneurysm repair.
Fig 3. Kaplan-Meier survival plot shows freedom from reintervention in patients with persistent type 2 (PT2) endoleak after endovascular aneurysm repair.
Table IV. Composite hazard ratio high-risk analysis of variables predictive of persistent type 2 endoleak Variables Patent IMA plus ⬎6 patent lumbar arteries Age ⬎70 years MALD ⬎30 mm ⬎6 patent lumbar arteries and age ⬎70 y Patent IMA and MALD ⬎30 mm plus ⬎6 patent lumbar arteries Age ⬎70 y ⬎6 patent lumbar arteries and age ⬎70 y
No.
HR (95% CI)
57 109 98
8.76 (3.33-23.06) 8.94 (3.16-25.17) 9.14 (3.36-24.86)
46 17.70 (5.91-52.87) 46 18.10 (6.26-52.32) 81 18.47 (5.98-56.78) 35 36.57 (11.23-118.76)
CI, Confidence interval; HR, hazard ratio; IMA, inferior mesenteric artery; MALD, maximum aneurysm luminal diameter.
Fig 2. Kaplan-Meier survival plot shows freedom from sac enlargement ⬎5 mm in patients with persistent type 2 (PT2) endoleak after endovascular aneurysm repair.
Composite HR high-risk analysis identified multiple combinations of preoperative characteristics and anatomic variables that increased the risk that a PT2 would develop (Table IV). An increase in the risk of PT2 was found in patients with a patent IMA and one of the following risk factors: (1) more than six patent lumbar arteries (8.76-fold increase), (2) age ⬎70 years (8.94-fold increase), or (3) a maximum luminal diameter ⬎30 mm (9.14-fold increase). The risk increased approximately 18-fold in patients with a patent IMA and any two of these risk factors. DISCUSSION Continued perfusion and pressurization of an aneurysm sac due to endoleak remains a significant complication
of EVAR. Although type 2 endoleaks are less likely to require secondary reintervention than type 1 or 3 endoleaks, they have been identified as a risk factor for aneurysm sac growth and rupture.7 Previous studies have reported a 9% to 25% incidence of type 2 endoleak after EVAR.5,7,13,14 The incidence in this study was slightly higher, with type 2 endoleak developing in 31% of patients; however, this may be because many type 2 endoleaks were not seen on the initial postoperative CT scans but were present later in the follow-up period. Type 2 endoleaks that persist ⬎6 months are less likely to resolve and are associated with an increased risk of adverse events compared with transient type 2 endoleaks.15 Data from our own institution found a decreased freedom from rupture at 1, 3, and 5 years in patients with PT2 compared with those with transient type 2 endoleaks.7 The previous study from
JOURNAL OF VASCULAR SURGERY Volume 52, Number 1
our institution defined PT2 as a type 2 endoleak present ⱕ1 month of EVAR placement and persisting ⱖ6 months after EVAR. The current study included all type 2 endoleaks, including those not seen on early (⬍6 month) CT scans, in an effort to capture all PT2 because they behave similarly and are treated in a similar manner. Given the increased risk of adverse events, some have attempted to identify risk factors for PT2. Previous studies have shown that both patent IMA5,16,17 and patent lumbar arteries18 increase the risk of type 2 endoleak after EVAR. Furthermore, data from our institution and others have shown that the rate of type 2 endoleaks increases with an increasing combined number of patent aortic branch vessels.19,20 Although we believe this is the first study to find increasing age as a risk factor, other reports have implicated not only patent aortic branch vessels but also aneurysm sac thrombus load in the development of type 2 endoleaks. Sampaio et al12 found an inverse relationship between the probability of type 2 endoleak and the percentage of thrombosed cross-sectional area of the aneurysm sac. Maximum aneurysm luminal diameter, a correlate of the percentage of sac thrombosis, has also been shown to increase the risk of PT2.21 When the composite HRs of patent branch vessels and maximum aneurysm luminal diameter in the current study were examined, there was an approximately 18-fold increased risk that a PT2 would develop. It is likely that the increased flow channel within the aneurysm sac coupled with a larger number of patent aortic branch vessels leads to increased flow velocities within the sac, thus decreasing the likelihood of type 2 endoleak resolution.19 The natural history of PT2 has been difficult to define due to the conflicting reports that have been published. Some authors have advocated an aggressive approach at reintervention due to the association with adverse events and limited sensitivity of CT scanning,15 whereas others have argued for a more conservative course due to the low rate of these events.5 Larger series of patients have found, similar to this study, that it is safe to continue monitoring patients with PT2 as long as their aneurysm sacs are not enlarging.6,22 Approximately three-quarters of the patients with PT2 in this study had stable aneurysm sac size or spontaneous resolution, whereas only one-quarter required reintervention. With such a low rate of rupture (2 of 136 patients with PT2), it seems prudent to adopt a conservative approach with continued radiologic surveillance as long as there is close follow-up and no evidence of aneurysm growth. Of all the factors examined, only chronic obstructive pulmonary disease (COPD) was associated with spontaneous resolution of a PT2. This association has been reported previously, but the exact cause is not known.5,6 Some have hypothesized that COPD may be a marker for generalized atherosclerosis of side branch vessels.5 Alternatively, increased blood viscosity in patients with COPD may contribute to the thrombosis of the PT2.6 Aneurysm sac enlargement ⬎5 mm due to type 2 endoleaks ranges from 5% to 10%6,22 but may be seen in as
Abularrage et al 23
many as 40% to 55% of patients with PT2 endoleaks.7,21 Previous studies have identified maximum endoleak cavity as predictive of aneurysm enlargement.21 In this study, increasing preoperative age was associated with sac enlargement ⬎5 mm in patients with PT2. This association is unclear, because no anatomic variables predicted sac enlargement. Expansion of the aneurysm sac in the presence of a PT2 should be considered an indication for reintervention. Multiple postoperative techniques have been used, including sac or branch vessel coil embolization, or both, or intrasac injection of prothrombotic material, with a reported success of 11% to 100%.4,15 Long-term outcomes of reintervention were not assessed in this study because of a limited follow-up, but it is apparent that some of those patients with PT2 will require more than one reintervention. Owing to this fact and the high-risk composite HR, it is likely that PT2 is a dynamic process with changing branch vessel involvement and flow patterns around a common central sac channel.23 Patency of aortic branch vessels combined with a large flow lumen provides multiple possible inflow and outflow channels for the PT2, thus emphasizing the need for continued surveillance in this patient population. Identification of patients at increased risk for PT2, and thus adverse events, has prompted some to adopt a preoperative strategy of branch vessel management.24 Axelrod et al25 reported a decrease in the rate of PT2 from 48% to 17% after preoperative IMA coil embolization. Muthu et al26 found a decrease in the incidence of type 2 endoleaks from 26% to 14% after preoperative IMA coil embolization and intrasac thrombin injection, although this did not reach statistical significance. Finally, Zanchetta et al9 were able to decrease the incidence of PT2 to 2.4% with intraoperative sac fibrin glue injection. Although these results are promising, the patient populations were small, nonrandomized, and had limited follow-up. Furthermore, the results of the current study suggest that 93% of patients would have undergone an unnecessary procedure, because only 39 of 595 patients required reintervention for PT2. The data presented here would argue for selective perioperative endovascular intervention in those patients with a substantial risk based on the high-risk composite HRs. Our study is limited by its retrospective nature and by the differing practice patterns of the surgical staff. Toward the end of the study period, research from our institution recognized the persistence of an early type 2 endoleak ⬎6 months as an independent predictor of adverse outcomes. For this reason, some surgeons chose early reintervention at 6 months after EVAR rather than surveillance of nonexpanding aneurysms with PT2. Another limitation is that preoperative and postoperative anatomic variables were based on CT scan imaging, and the accurate identification of patency of small aortic branch vessels or type 2 endoleaks, or both, can sometimes be difficult. Others have shown that color duplex ultrasound imaging provides hemodynamic information not available by CT and may be superior in detecting type 2 endoleaks.23
24 Abularrage et al
Finally, the composite HR risks were obtained by mathematical calculation based on the available data set. Further studies based on independent data sets are needed to validate these results. CONCLUSIONS The current study represents the largest series to date, to our knowledge, of patients presenting with PT2 after EVAR with the longest follow-up. Preoperative demographic and anatomic factors increasing the risk that a PT2 would develop include patent IMA, increasing maximum luminal diameter, number of patent lumbar arteries, and age. The combination of these readily identifiable preoperative variables can identify a cohort of patients that should be considered for perioperative interventions to preclude PT2. AUTHOR CONTRIBUTIONS Conception and design: CA, RS, MC, CK, DB, RC, GL Analysis and interpretation: CA, MC, HL, GL Data collection: CA Writing the article: CA, GL Critical revision of the article: CA, RS, MC, HL, CK, DB, RC, GL Final approval of the article: CA, RS, MC, HL, CK, DB, RC, GL Statistical analysis: CA, RC, MC, HL Obtained funding: Not applicable Overall responsibility: CA REFERENCES 1. Lee WA, Carter JW, Upchurch G, Seeger JM, Huber TS. Perioperative outcomes after open and endovascular repair of intact abdominal aortic aneurysms in the United States during 2001. J Vasc Surg 2004;39: 491-6. 2. Greenhalgh RM, Brown LC, Kwong GP, Powell JT, Thompson SG. Comparison of endovascular aneurysm repair with open repair in patients with abdominal aortic aneurysm (EVAR trial 1), 30-day operative mortality results: randomised controlled trial. Lancet 2004;364:843-8. 3. Brewster DC, Jones JE, Chung TK, Lamuraglia GM, Kwolek CJ, Watkins MT, et al. Long-term outcomes after endovascular abdominal aortic aneurysm repair: the first decade. Ann Surg 2006;244:426-38. 4. Conrad MF, Adams AB, Guest JM, Paruchuri V, Brewster DC, LaMuraglia GM, et al. Secondary intervention after endovascular abdominal aortic aneurysm repair. Ann Surg 2009;250:383-9. 5. van Marrewijk CJ, Fransen G, Laheij RJ, Harris PL, Buth J; EUROSTAR Collaborators. Is a type II endoleak after EVAR a harbinger of risk? Causes and outcome of open conversion and aneurysm rupture during follow-up. Eur J Vasc Endovasc Surg 2004;27:128-37. 6. Silverberg D, Baril DT, Ellozy SH, Carroccio A, Greyrose SE, Lookstein RA, et al. An 8-year experience with type II endoleaks: natural history suggests selective intervention is a safe approach. J Vasc Surg 2006;44:453-9. 7. Jones JE, Atkins MD, Brewster DC, Chung TK, Kwolek CJ, LaMuraglia GM, et al. Persistent type 2 endoleak after endovascular repair of abdominal aortic aneurysm is associated with adverse late outcomes. J Vasc Surg 2007;46:1-8. 8. Parry DJ, Kessel DO, Robertson I, Denton L, Patel JV, Berridge DC, et al. Type II endoleaks: predictable, preventable, and sometimes treatable? J Vasc Surg 2002;36:105-10.
JOURNAL OF VASCULAR SURGERY July 2010
9. Zanchetta M, Faresin F, Pedon L, Ronsivalle S. Intraoperative intrasac thrombin injection to prevent type II endoleak after endovascular abdominal aortic aneurysm repair. J Endovasc Ther 2007;14:176-83. 10. Klahr S. The modification of diet in renal disease study. N Engl J Med 1989;320:864-6. 11. National Kidney Foundation. K/DOQI clinical practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Am J Kidney Dis 2002;39(2 suppl 1):S1-266. 12. Sampaio SM, Panneton JM, Mozes GI, Andrews JC, Bower TC, Kalra M, et al. Aneurysm sac thrombus load predicts type II endoleaks after endovascular aneurysm repair. Ann Vasc Surg 2005;19:302-9. 13. Sheehan MK, Ouriel K, Greenberg R, McCann R, Murphy M, Fillinger M, et al. Are type II endoleaks after endovascular aneurysm repair endograft dependent? J Vasc Surg 2006;43:657-61. 14. Veith FJ, Baum RA, Ohki T, Amor M, Adiseshiah M, Blankensteijn JD, et al. Nature and significance of endoleaks and endotension: summary of opinions expressed at an international conference. J Vasc Surg 2002;35:1029-35. 15. Gelfand DV, White GH, Wilson SE. Clinical significance of type II endoleak after endovascular repair of abdominal aortic aneurysm. Ann Vasc Surg 2006;20:69-74. 16. Velazquez OC, Baum RA, Carpenter JP, Golden MA, Cohn M, Pyeron A, et al. Relationship between preoperative patency of the inferior mesenteric artery and subsequent occurrence of type II endoleak in patients undergoing endovascular repair of abdominal aortic aneurysms. J Vasc Surg 2000;32:777-88. 17. Warrier R, Miller R, Bond R, Robertson IK, Hewitt P, Scott A. Risk factors for type II endoleaks after endovascular repair of abdominal aortic aneurysms. ANZ J Surg 2008;78:61-3. 18. Fritz GA, Deutschmann HA, Schoellnast H, Stessel U, Sorantin E, Portugaller HR, et al. Frequency and significance of lumbar and inferior mesenteric artery perfusion after endovascular repair of abdominal aortic aneurysms. J Endovasc Ther 2004;11:649-58. 19. Arko FR, Filis KA, Siedel SA, Johnson BL, Drake AR, Fogarty TJ, et al. Intrasac flow velocities predict sealing of type II endoleaks after endovascular abdominal aortic aneurysm repair. J Vasc Surg 2003;37:8-15. 20. Fan CM, Rafferty EA, Geller SC, Kaufman JA, Brewster DC, Cambria RP, et al. Endovascular stent-graft in abdominal aortic aneurysms: the relationship between patent vessels that arise from the aneurysmal sac and early endoleak. Radiology 2001;218:176-82. 21. Timaran CH, Ohki T, Rhee SJ, Veith FJ, Gargiulo NJ 3rd, Toriumi H, et al. Predicting aneurysm enlargement in patients with persistent type II endoleaks. J Vasc Surg 2004;39:1157-62. 22. Steinmetz E, Rubin BG, Sanchez LA, Choi ET, Geraghty PJ, Baty J, et al. Type II endoleak after endovascular abdominal aortic aneurysm repair: a conservative approach with selective intervention is safe and cost-effective. J Vasc Surg 2004;39:306-13. 23. Parent FN, Meier GH, Godziachvili V, LeSar CJ, Parker FM, Carter KA, et al. The incidence and natural history of type I and II endoleak: a 5-year follow-up assessment with color duplex ultrasound scan. J Vasc Surg 2002;35:474-81. 24. Bonvini R, Alerci M, Antonucci F, Tutta P, Wyttenbach R, Bogen M, et al. Preoperative embolization of collateral side branches: a valid means to reduce type II endoleaks after endovascular AAA repair. J Endovasc Ther 2003;10:227-32. 25. Axelrod DJ, Lookstein RA, Guller J, Nowakowski FS, Ellozy S, Carroccio A, et al. Inferior mesenteric artery embolization before endovascular aneurysm repair: technique and initial results. J Vasc Interv Radiol 2004;15:1263-7. 26. Muthu C, Maani J, Plank LD, Holden A, Hill A. Strategies to reduce the rate of type II endoleaks: routine intraoperative embolization of the inferior mesenteric artery and thrombin injection into the aneurysm sac. J Endovasc Ther 2007;14:661-8.
Submitted Oct 15, 2009; accepted Feb 7, 2010.
Preoperative variables predict persistent type 2 endoleak after endovascular aneurysm repair Christopher J. Abularrage, MD,a Robert S. Crawford, MD,a Mark F. Conrad, MD, MMSc,a Hang Lee, PhD,b Christopher J. Kwolek, MD,a David C. Brewster, MD,a Richard P. Cambria, MD,a and Glenn M. LaMuraglia, MD,a Boston, Mass Objective: Persistent type 2 endoleaks (PT2, present >6 months) after endovascular aneurysm repair (EVAR) are associated with adverse outcomes. This study evaluated the preoperative risk factors and natural history of PT2 in order to define a population at high risk. Methods: From January 1999 to December 2007, 595 of 832 EVAR patients had long-term computed tomography follow-up and comprised the study cohort. Preoperative anatomic and clinical variables were correlated with PT2 using Cox regression. Composite hazard ratios (HRs) were constructed with clusters of high-risk preoperative variables. Primary end points, including spontaneous resolution, sac enlargement >5 mm, and freedom from reintervention, were evaluated using Kaplan-Meier analysis. Results: There were 136 PT2 patients (23%) with a median follow-up of 34.8 months (range, 6.4-121.2 months). Positive predictive factors included patent inferior mesenteric artery (IMA; HR, 4.00; 95% confidence interval [CI], 1.62-9.90; P ⴝ .003), increasing number of patent lumbar arteries (HR, 1.24; 95% CI, 1.10-1.41; P ⴝ .0006), increasing age (HR, 1.04; 95% CI, 1.01-1.06; P ⴝ .005), and increasing luminal diameter on CT-contrast opacified lumen (HR, 1.03; 95% CI, 1.02-1.05; P ⴝ .0001). During follow-up, spontaneous PT2 resolution occurred in 34 patients (25%), sac diameter remained stable in 63 (46%), and rupture occurred in 2 (1.5%). Kaplan-Meier analysis estimated that 35.2% ⴞ 5.6% (95% CI, 23.8%-46.2%) of PT2 resolve spontaneously at 5 years after the index procedure. Freedom from sac enlargement >5 mm was 54.6% ⴞ 7.2% (95% CI, 40.6%-69.4%) at 5 years. Fifty-nine reinterventions were performed in 39 patients with PT2. Freedom from reintervention was 67.3% ⴞ 5.0% (95% CI, 57.0%-77.0%) at 5 years. The combination of a patent IMA and one risk factor of more than six patent lumbar arteries, maximum luminal diameter >30 mm, or age >70 years increased the odds of PT2 approximately ninefold. The combination of a patent IMA and any two risk factors increased the odds of PT2 approximately 18-fold. Conclusions: Several readily identifiable preoperative variables are associated with PT2 whose natural history was benign in but 35% of patients. On the basis of the composite high-risk HRs, there is accordingly a cohort of patients in whom perioperative interventions to preclude PT2 should be considered. ( J Vasc Surg 2010;52:19-24.)
Endovascular abdominal aortic aneurysm repair (EVAR) has become mainstream therapy for abdominal aortic aneurysms (AAAs) with suitable anatomy. EVAR has been associated with a 65% to 70% reduction in the 30-day mortality compared with open AAA repair.1,2 Clinically relevant late outcomes also have now been documented as favorable with EVAR.3 EVAR requires continued surveillance because up to 11% of patients need reintervention for graft-related comFrom the Division of Vascular and Endovascular Surgery,a and Biostatistics Center,b Massachusetts General Hospital and Harvard Medical School. This study was supported in part by grants from the Monte and Rita Goldman Foundation, Mr John F. Murphy, and the Bay State Federal Savings Foundation. Competition of interest: Dr. Richard Cambria has a consulting agreement with W.L. Gore. Presented at the Thirty sixth Annual Meeting of the New England Society for Vascular Surgery, Boston, Mass, Oct 2-4, 2009. Reprint requests: Glenn M. LaMuraglia, MD, Division of Vascular and Endovascular Surgery, Massachusetts General Hospital, WAC440, 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 competition of interest. 0741-5214/$36.00 Copyright © 2010 by the Society for Vascular Surgery. doi:10.1016/j.jvs.2010.02.023
plications, in particular endoleak.4 The presence of type 1 and 3 endoleaks is a clear indication for reintervention, whereas the clinical significance of type 2 endoleaks is not well established. Type 2 endoleaks occur in 10% to 30% of patients after EVAR and may be associated with aneurysm growth and rupture.5,6 Subdivision of type 2 endoleaks into transient (resolving ⱕ6 months) and persistent (present ⬎6 months) has been found to predict EVARrelated complications. Persistent type 2 endoleaks (PT2) have been associated with an increased incidence of adverse outcomes, including aneurysm sac growth, reintervention rate, the need for conversion to open repair, and rupture.7 Accordingly, there has been enthusiasm in some reports for preemptive adjunctive procedures, such as inferior mesenteric artery (IMA) coil embolization or intrasac thrombin injection, to preclude type 2 endoleaks.8,9 This study evaluated the preoperative risk factors and natural history of PT2 during a 10-year period at a single academic medical center to better define a population at high-risk for such lesions postoperatively. METHODS This study was approved by the Institutional Review Board of the Massachusetts General Hospital. 19
JOURNAL OF VASCULAR SURGERY July 2010
20 Abularrage et al
Data collection. All EVARs performed between January 1999 and December 2007 were retrospectively identified from a prospectively maintained database. Exclusion criteria included patients with thoracic aortic aneurysms, anastomotic aneurysms, isolated iliac artery aneurysms, and ⬍6 months of follow-up. Preoperative data collection was obtained from the patient hospital and office records and included demographics, medical history, and laboratory results. Estimated glomerular filtration rate (GFR) was calculated according to the Modification of Diet in Renal Disease Study.10 Chronic renal insufficiency (CRI) was defined as a GFR ⬍60 mL/min/1.73 m2, or chronic kidney disease stages 3 to 5.11 All preoperative and postoperative computed tomography (CT) scans were reviewed by an attending radiologist, an attending vascular surgeon, and a vascular fellow. Patients underwent preoperative CT imaging with 2.0- to 2.5-mm cuts. Preoperative anatomic variables included maximum aneurysm diameter, maximum aneurysm luminal diameter, patency of the IMA, number of patent lumbar arteries, occlusion of internal iliac vessels (chronic or by preoperative embolization), and percentage of aneurysm sac thrombosis. The maximum aneurysm luminal diameter was defined as the shortest diagonal of the contrastopacified lumen measured on the same cross-sectional CT image with the maximum aneurysm diameter. Thrombus load was defined as the area of the aneurysm sac measured on the same cross-sectional CT image with the maximum aneurysm diameter minus the maximum aneurysm luminal area. The percentage of sac thrombosis was calculated by the method of Sampaio et al12 and defined as the ratio of the area of the thrombus load to that of the aneurysm sac. All patients in the study received a follow-up CT scan with intravenous contrast and thin collimation ⱖ6 months postoperatively. After a nonenhanced CT scan was performed, a bolus injection of contrast was administered at 4 mL/s with a 25-second preparation delay. Delayed-phase images were obtained and reconstructed with 2.0- to 2.5-mm cuts. The presence of a type 2 endoleak was determined by the radiologist or surgeon, or both, and was corroborated in a blind fashion by a study vascular surgeon, or fellow, or both. Arteriography was performed when it was not possible to differentiate between different types of endoleak. PT2 was defined as any type 2 endoleak present ⱖ6 months postoperatively and included late-appearing type 2 endoleaks that were not present before 6 months. Further follow-up imaging with CT angiography was typically obtained at 6-month to 1-year intervals. The decision to intervene and the method of treatment were not uniform due to the different methods of treatment available during the 10-year period. Data analysis. Univariate comparisons of patient demographic and risk factors were performed using the CoxMantel test of the Kaplan-Meier survival estimates. All variables with P ⬍ .10 and with five or more patients were included in multivariate Cox regression analysis. Composite hazard ratio (HR) high-risk analysis was performed first
by converting continuous independent variables to binary variables. This was accomplished by choosing multiple points above and below the mean and selecting the point with the lowest P value when comparing patients with and without PT2. Variables with P ⬍ .05 on Cox regression were combined in different permutations, and adjusted HRs and 95% confidence intervals (CIs) were calculated. Only those composite high-risk variables with an HR greater than all single variables were reported. All values of P ⬍ .05 (two-tailed) were considered significant. Analyses were performed using StatView software (SAS Inc, Cary, NC). RESULTS During the study period, EVAR was performed in 832 patients. Four stent graft designs were included in this study: AneuRx (Medtronic, Minneapolis, Minn), Excluder (W. L. Gore & Associates, Flagstaff, Ariz), Powerlink (Endologix, Irvine, Calif), and Zenith (Cook, Bloomington, Ind). Overall results from our EVAR experience have been published previously.3 The study group consisted of 595 patients who underwent a follow-up CT scan ⱖ6 months postoperatively and 47 had an initial transient type 2 endoleak that was not detected ⬎6 months. PT2 was identified in 136 patients (23%) with a median follow-up of 34.8 months (range, 6.4-121.2 months), and a PT2 was discovered in 62 of these patients that was not detected on the initial postoperative CT scan. Preoperative patient characteristics and anatomic variables are summarized in Table I. On univariate Cox-Mantel test of the Kaplan-Meier estimates (Table II), a PT2 was more likely to develop in older patients (P ⫽ .007) and those taking warfarin (P ⫽ .03) but was less likely to develop in those who smoked tobacco (P ⫽ .03). A CoxMantel test of the preoperative CT imaging Kaplan-Meier estimates showed a PT2 was more likely to develop in patients with a patent IMA (P ⬍ .0001), an increased number of patent lumbar arteries (P ⬍ .0001), increased maximum aneurysm luminal diameter (P ⬍ .0001), and a decreased percentage of aneurysm sac thrombosis (P ⬍ .0001). No other significant anatomic or preoperative variables predicted the development of a PT2 on univariate analysis. Factors predictive of PT2 on Cox regression analysis are summarized in Table III. These included patent IMA (HR, 4.00; 95% CI, 1.62-9.90; P ⫽ .003), increasing number of patent lumbar arteries (HR, 1.24; 95% CI, 1.10-1.41; P ⫽ .0006), increasing age (HR, 1.04; 95% CI, 1.01-1.06; P ⫽ .005), and increasing maximum aneurysm luminal diameter (HR, 1.03; 95% CI, 1.02-1.05; P ⫽ .0001). Decreasing percentage of sac thrombosis was also predictive of PT2 but was not included in the final model due to a high degree of correlation with increasing maximum aneurysm luminal diameter (correlation coefficient, – 0.81). Overall, spontaneous PT2 resolution occurred in 34 patients (25%) at a mean of 32 months, and rupture occurred in 2 (1.5%). On Cox regression analysis, only a
JOURNAL OF VASCULAR SURGERY Volume 52, Number 1
Abularrage et al 21
Table I. Demographic and anatomic variables of patients with and without persistent type 2 endoleaks Variablea Patients Preoperative comorbidities Age Sex Male Female Any tobacco use Peripheral vascular disease History of deep venous thrombosis Hypercholesterolemia Coronary artery disease Cerebrovascular accident Chronic renal insufficiencyb Chronic obstructive pulmonary disease Diabetes mellitus Hypertension Congestive heart failure Medications Warfarin Statin Clopidogrel Aspirin ACE inhibitor Steroids -blocker Preoperative anatomic characteristics Patent inferior mesenteric artery Patent lumbar arteries, No. Max aneurysm luminal diameter, mm Sac thrombosis percentage Hypogastric coil embolization or occlusion Maximum aneurysm diameter, mm
PT2 endoleak
No PT2 endoleak
136
459
77.21 ⫾ 0.58 75.16 ⫾ 0.37 117 (86) 19 (14) 89 (65) 11 (8)
369 (80) 90 (20) 346 (75) 75 (16)
7 (5) 73 (54) 63 (46) 11 (8) 55 (40)
15 (3) 268 (58) 227 (49) 47 (10) 67 (15)
26 (19) 15 (11) 100 (74) 15 (11)
106 (23) 62 (14) 342 (75) 59 (13)
21 (15) 64 (47) 9 (7) 57 (42) 34 (25) 11 (8) 62 (46)
49 (11) 250 (54) 21 (5) 221 (48) 112 (24) 45 (10) 218 (48)
127 (93) 6.39 ⫾ 0.15
347 (76) 5.72 ⫾ 0.08
37.87 ⫾ 0.82 32.50 ⫾ 0.47 45.29 ⫾ 2.14 55.96 ⫾ 1.14 21 (15)
87 (19)
Table II. Univariate comparisons of Kaplan-Meier estimates of event time distributions for predictors of persistent type 2 endoleak Predictora
P value
Preoperative comorbidities Age, y Any tobacco use Peripheral vascular disease History of deep venous thrombosis Sex Hypercholesterolemia Coronary artery disease Cerebrovascular accident Chronic renal insufficiencyb Chronic obstructive pulmonary disease Diabetes mellitus Hypertension Congestive heart failure Medications Warfarin Statin Clopidogrel Aspirin ACE inhibitor Steroids -blocker Preoperative anatomic characteristics Patent inferior mesenteric artery Patent lumbar arteries, No. Maximum aneurysm luminal diameter, mm Sac thrombosis percentage Maximum aneurysm diameter, mm Hypogastric coil embolization or occlusion
history of chronic obstructive pulmonary disease predicted spontaneous resolution (HR, 2.28; 95% CI, 1.07-4.85; P ⫽ .03). Reinterventions were not done in 63 patients (46%) who had continued PT2 with stable sac size and were monitored with serial CT scans. During follow-up, KaplanMeier analysis estimated that 35.2% ⫾ 5.6% (95% CI, 23.8-46.2) of PT2 resolve spontaneously (Fig 1), and freedom from sac enlargement ⬎5 mm was 54.6% ⫾ 7.2% (95% CI, 40.6-69.4) at 5 years (Fig 2). On Cox regression analysis, only increasing age predicted sac enlargement ⬎5 mm (HR, 1.06; 95% CI, 1.00-1.12; P ⫽ .048). The average change in aneurysm sac size was –2.7 ⫾ 1.5 mm in patients with spontaneous resolution of the PT2, –1.7 ⫾ 0.7 mm in patients with continued PT2 not undergoing reintervention, 3.9 ⫾ 1.6 mm in patients requiring a reintervention, and 9.0 ⫾ 4.0 mm in patients with aneurysm rupture.
.03 .16 .20 .27 .51 .75 .99 ⬍.0001 ⬍.0001 ⬍.0001 ⬍.0001 .23 .29
ACE, Angiotensin-converting enzyme. a Continuous variables were measured as mean ⫾ standard error of the mean. b Defined as a glomerular filtration rate of ⬍60 mL/min/1.73 m2.
Table III. Multivariate Cox regression analysis for predictors of persistent type 2 endoleak
52.90 ⫾ 0.66 51.83 ⫾ 0.45
ACE, Angiotensin-converting enzyme inhibitor; PT2, persistent type 2. a Continuous data are presented as mean ⫾ standard error of the mean; categoric data as number (%). b Defined as a glomerular filtration rate of ⬍60 mL/min/1.73 m2.
.007 .03 .07 .12 .21 .36 .41 .42 .42 .44 .48 .86 .96
Predictor Patent inferior mesenteric artery Increasing number of patent lumbar arteries Increasing age, y Increasing MALD, mm
HR (95% CI)
P value
4.00 (1.62-9.90)
.003
1.24 (1.10-1.41) 1.04 (1.01-1.06) 1.03 (1.02-1.05)
.0006 .005 .0001
CI, Confidence interval; HR, hazard ratio; MALD, maximum aneurysm luminal diameter.
Fifty-nine reinterventions were performed in 39 patients with PT2. Freedom from reintervention was 67.3% ⫾ 5.0% (95% CI, 57.0-77.0) at 5 years (Fig 3). Reinterventions included 14 IMA coil embolizations, 14 sac coil embolizations, 10 glue injections with or without coil embolizations, 6 lumbar coil embolizations, 2 open ligation of bleeding vessels, and 12 failed attempts (inability to access the aneurysm sac). One patient refused an endovascular attempt at PT2 treatment and requested a graft explant, which was performed.
JOURNAL OF VASCULAR SURGERY July 2010
22 Abularrage et al
Fig 1. Kaplan-Meier survival plot shows time to spontaneous resolution in patients with persistent type 2 (PT2) endoleak after endovascular aneurysm repair.
Fig 3. Kaplan-Meier survival plot shows freedom from reintervention in patients with persistent type 2 (PT2) endoleak after endovascular aneurysm repair.
Table IV. Composite hazard ratio high-risk analysis of variables predictive of persistent type 2 endoleak Variables Patent IMA plus ⬎6 patent lumbar arteries Age ⬎70 years MALD ⬎30 mm ⬎6 patent lumbar arteries and age ⬎70 y Patent IMA and MALD ⬎30 mm plus ⬎6 patent lumbar arteries Age ⬎70 y ⬎6 patent lumbar arteries and age ⬎70 y
No.
HR (95% CI)
57 109 98
8.76 (3.33-23.06) 8.94 (3.16-25.17) 9.14 (3.36-24.86)
46 17.70 (5.91-52.87) 46 18.10 (6.26-52.32) 81 18.47 (5.98-56.78) 35 36.57 (11.23-118.76)
CI, Confidence interval; HR, hazard ratio; IMA, inferior mesenteric artery; MALD, maximum aneurysm luminal diameter.
Fig 2. Kaplan-Meier survival plot shows freedom from sac enlargement ⬎5 mm in patients with persistent type 2 (PT2) endoleak after endovascular aneurysm repair.
Composite HR high-risk analysis identified multiple combinations of preoperative characteristics and anatomic variables that increased the risk that a PT2 would develop (Table IV). An increase in the risk of PT2 was found in patients with a patent IMA and one of the following risk factors: (1) more than six patent lumbar arteries (8.76-fold increase), (2) age ⬎70 years (8.94-fold increase), or (3) a maximum luminal diameter ⬎30 mm (9.14-fold increase). The risk increased approximately 18-fold in patients with a patent IMA and any two of these risk factors. DISCUSSION Continued perfusion and pressurization of an aneurysm sac due to endoleak remains a significant complication
of EVAR. Although type 2 endoleaks are less likely to require secondary reintervention than type 1 or 3 endoleaks, they have been identified as a risk factor for aneurysm sac growth and rupture.7 Previous studies have reported a 9% to 25% incidence of type 2 endoleak after EVAR.5,7,13,14 The incidence in this study was slightly higher, with type 2 endoleak developing in 31% of patients; however, this may be because many type 2 endoleaks were not seen on the initial postoperative CT scans but were present later in the follow-up period. Type 2 endoleaks that persist ⬎6 months are less likely to resolve and are associated with an increased risk of adverse events compared with transient type 2 endoleaks.15 Data from our own institution found a decreased freedom from rupture at 1, 3, and 5 years in patients with PT2 compared with those with transient type 2 endoleaks.7 The previous study from
JOURNAL OF VASCULAR SURGERY Volume 52, Number 1
our institution defined PT2 as a type 2 endoleak present ⱕ1 month of EVAR placement and persisting ⱖ6 months after EVAR. The current study included all type 2 endoleaks, including those not seen on early (⬍6 month) CT scans, in an effort to capture all PT2 because they behave similarly and are treated in a similar manner. Given the increased risk of adverse events, some have attempted to identify risk factors for PT2. Previous studies have shown that both patent IMA5,16,17 and patent lumbar arteries18 increase the risk of type 2 endoleak after EVAR. Furthermore, data from our institution and others have shown that the rate of type 2 endoleaks increases with an increasing combined number of patent aortic branch vessels.19,20 Although we believe this is the first study to find increasing age as a risk factor, other reports have implicated not only patent aortic branch vessels but also aneurysm sac thrombus load in the development of type 2 endoleaks. Sampaio et al12 found an inverse relationship between the probability of type 2 endoleak and the percentage of thrombosed cross-sectional area of the aneurysm sac. Maximum aneurysm luminal diameter, a correlate of the percentage of sac thrombosis, has also been shown to increase the risk of PT2.21 When the composite HRs of patent branch vessels and maximum aneurysm luminal diameter in the current study were examined, there was an approximately 18-fold increased risk that a PT2 would develop. It is likely that the increased flow channel within the aneurysm sac coupled with a larger number of patent aortic branch vessels leads to increased flow velocities within the sac, thus decreasing the likelihood of type 2 endoleak resolution.19 The natural history of PT2 has been difficult to define due to the conflicting reports that have been published. Some authors have advocated an aggressive approach at reintervention due to the association with adverse events and limited sensitivity of CT scanning,15 whereas others have argued for a more conservative course due to the low rate of these events.5 Larger series of patients have found, similar to this study, that it is safe to continue monitoring patients with PT2 as long as their aneurysm sacs are not enlarging.6,22 Approximately three-quarters of the patients with PT2 in this study had stable aneurysm sac size or spontaneous resolution, whereas only one-quarter required reintervention. With such a low rate of rupture (2 of 136 patients with PT2), it seems prudent to adopt a conservative approach with continued radiologic surveillance as long as there is close follow-up and no evidence of aneurysm growth. Of all the factors examined, only chronic obstructive pulmonary disease (COPD) was associated with spontaneous resolution of a PT2. This association has been reported previously, but the exact cause is not known.5,6 Some have hypothesized that COPD may be a marker for generalized atherosclerosis of side branch vessels.5 Alternatively, increased blood viscosity in patients with COPD may contribute to the thrombosis of the PT2.6 Aneurysm sac enlargement ⬎5 mm due to type 2 endoleaks ranges from 5% to 10%6,22 but may be seen in as
Abularrage et al 23
many as 40% to 55% of patients with PT2 endoleaks.7,21 Previous studies have identified maximum endoleak cavity as predictive of aneurysm enlargement.21 In this study, increasing preoperative age was associated with sac enlargement ⬎5 mm in patients with PT2. This association is unclear, because no anatomic variables predicted sac enlargement. Expansion of the aneurysm sac in the presence of a PT2 should be considered an indication for reintervention. Multiple postoperative techniques have been used, including sac or branch vessel coil embolization, or both, or intrasac injection of prothrombotic material, with a reported success of 11% to 100%.4,15 Long-term outcomes of reintervention were not assessed in this study because of a limited follow-up, but it is apparent that some of those patients with PT2 will require more than one reintervention. Owing to this fact and the high-risk composite HR, it is likely that PT2 is a dynamic process with changing branch vessel involvement and flow patterns around a common central sac channel.23 Patency of aortic branch vessels combined with a large flow lumen provides multiple possible inflow and outflow channels for the PT2, thus emphasizing the need for continued surveillance in this patient population. Identification of patients at increased risk for PT2, and thus adverse events, has prompted some to adopt a preoperative strategy of branch vessel management.24 Axelrod et al25 reported a decrease in the rate of PT2 from 48% to 17% after preoperative IMA coil embolization. Muthu et al26 found a decrease in the incidence of type 2 endoleaks from 26% to 14% after preoperative IMA coil embolization and intrasac thrombin injection, although this did not reach statistical significance. Finally, Zanchetta et al9 were able to decrease the incidence of PT2 to 2.4% with intraoperative sac fibrin glue injection. Although these results are promising, the patient populations were small, nonrandomized, and had limited follow-up. Furthermore, the results of the current study suggest that 93% of patients would have undergone an unnecessary procedure, because only 39 of 595 patients required reintervention for PT2. The data presented here would argue for selective perioperative endovascular intervention in those patients with a substantial risk based on the high-risk composite HRs. Our study is limited by its retrospective nature and by the differing practice patterns of the surgical staff. Toward the end of the study period, research from our institution recognized the persistence of an early type 2 endoleak ⬎6 months as an independent predictor of adverse outcomes. For this reason, some surgeons chose early reintervention at 6 months after EVAR rather than surveillance of nonexpanding aneurysms with PT2. Another limitation is that preoperative and postoperative anatomic variables were based on CT scan imaging, and the accurate identification of patency of small aortic branch vessels or type 2 endoleaks, or both, can sometimes be difficult. Others have shown that color duplex ultrasound imaging provides hemodynamic information not available by CT and may be superior in detecting type 2 endoleaks.23
24 Abularrage et al
Finally, the composite HR risks were obtained by mathematical calculation based on the available data set. Further studies based on independent data sets are needed to validate these results. CONCLUSIONS The current study represents the largest series to date, to our knowledge, of patients presenting with PT2 after EVAR with the longest follow-up. Preoperative demographic and anatomic factors increasing the risk that a PT2 would develop include patent IMA, increasing maximum luminal diameter, number of patent lumbar arteries, and age. The combination of these readily identifiable preoperative variables can identify a cohort of patients that should be considered for perioperative interventions to preclude PT2. AUTHOR CONTRIBUTIONS Conception and design: CA, RS, MC, CK, DB, RC, GL Analysis and interpretation: CA, MC, HL, GL Data collection: CA Writing the article: CA, GL Critical revision of the article: CA, RS, MC, HL, CK, DB, RC, GL Final approval of the article: CA, RS, MC, HL, CK, DB, RC, GL Statistical analysis: CA, RC, MC, HL Obtained funding: Not applicable Overall responsibility: CA REFERENCES 1. Lee WA, Carter JW, Upchurch G, Seeger JM, Huber TS. Perioperative outcomes after open and endovascular repair of intact abdominal aortic aneurysms in the United States during 2001. J Vasc Surg 2004;39: 491-6. 2. Greenhalgh RM, Brown LC, Kwong GP, Powell JT, Thompson SG. Comparison of endovascular aneurysm repair with open repair in patients with abdominal aortic aneurysm (EVAR trial 1), 30-day operative mortality results: randomised controlled trial. Lancet 2004;364:843-8. 3. Brewster DC, Jones JE, Chung TK, Lamuraglia GM, Kwolek CJ, Watkins MT, et al. Long-term outcomes after endovascular abdominal aortic aneurysm repair: the first decade. Ann Surg 2006;244:426-38. 4. Conrad MF, Adams AB, Guest JM, Paruchuri V, Brewster DC, LaMuraglia GM, et al. Secondary intervention after endovascular abdominal aortic aneurysm repair. Ann Surg 2009;250:383-9. 5. van Marrewijk CJ, Fransen G, Laheij RJ, Harris PL, Buth J; EUROSTAR Collaborators. Is a type II endoleak after EVAR a harbinger of risk? Causes and outcome of open conversion and aneurysm rupture during follow-up. Eur J Vasc Endovasc Surg 2004;27:128-37. 6. Silverberg D, Baril DT, Ellozy SH, Carroccio A, Greyrose SE, Lookstein RA, et al. An 8-year experience with type II endoleaks: natural history suggests selective intervention is a safe approach. J Vasc Surg 2006;44:453-9. 7. Jones JE, Atkins MD, Brewster DC, Chung TK, Kwolek CJ, LaMuraglia GM, et al. Persistent type 2 endoleak after endovascular repair of abdominal aortic aneurysm is associated with adverse late outcomes. J Vasc Surg 2007;46:1-8. 8. Parry DJ, Kessel DO, Robertson I, Denton L, Patel JV, Berridge DC, et al. Type II endoleaks: predictable, preventable, and sometimes treatable? J Vasc Surg 2002;36:105-10.
JOURNAL OF VASCULAR SURGERY July 2010
9. Zanchetta M, Faresin F, Pedon L, Ronsivalle S. Intraoperative intrasac thrombin injection to prevent type II endoleak after endovascular abdominal aortic aneurysm repair. J Endovasc Ther 2007;14:176-83. 10. Klahr S. The modification of diet in renal disease study. N Engl J Med 1989;320:864-6. 11. National Kidney Foundation. K/DOQI clinical practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Am J Kidney Dis 2002;39(2 suppl 1):S1-266. 12. Sampaio SM, Panneton JM, Mozes GI, Andrews JC, Bower TC, Kalra M, et al. Aneurysm sac thrombus load predicts type II endoleaks after endovascular aneurysm repair. Ann Vasc Surg 2005;19:302-9. 13. Sheehan MK, Ouriel K, Greenberg R, McCann R, Murphy M, Fillinger M, et al. Are type II endoleaks after endovascular aneurysm repair endograft dependent? J Vasc Surg 2006;43:657-61. 14. Veith FJ, Baum RA, Ohki T, Amor M, Adiseshiah M, Blankensteijn JD, et al. Nature and significance of endoleaks and endotension: summary of opinions expressed at an international conference. J Vasc Surg 2002;35:1029-35. 15. Gelfand DV, White GH, Wilson SE. Clinical significance of type II endoleak after endovascular repair of abdominal aortic aneurysm. Ann Vasc Surg 2006;20:69-74. 16. Velazquez OC, Baum RA, Carpenter JP, Golden MA, Cohn M, Pyeron A, et al. Relationship between preoperative patency of the inferior mesenteric artery and subsequent occurrence of type II endoleak in patients undergoing endovascular repair of abdominal aortic aneurysms. J Vasc Surg 2000;32:777-88. 17. Warrier R, Miller R, Bond R, Robertson IK, Hewitt P, Scott A. Risk factors for type II endoleaks after endovascular repair of abdominal aortic aneurysms. ANZ J Surg 2008;78:61-3. 18. Fritz GA, Deutschmann HA, Schoellnast H, Stessel U, Sorantin E, Portugaller HR, et al. Frequency and significance of lumbar and inferior mesenteric artery perfusion after endovascular repair of abdominal aortic aneurysms. J Endovasc Ther 2004;11:649-58. 19. Arko FR, Filis KA, Siedel SA, Johnson BL, Drake AR, Fogarty TJ, et al. Intrasac flow velocities predict sealing of type II endoleaks after endovascular abdominal aortic aneurysm repair. J Vasc Surg 2003;37:8-15. 20. Fan CM, Rafferty EA, Geller SC, Kaufman JA, Brewster DC, Cambria RP, et al. Endovascular stent-graft in abdominal aortic aneurysms: the relationship between patent vessels that arise from the aneurysmal sac and early endoleak. Radiology 2001;218:176-82. 21. Timaran CH, Ohki T, Rhee SJ, Veith FJ, Gargiulo NJ 3rd, Toriumi H, et al. Predicting aneurysm enlargement in patients with persistent type II endoleaks. J Vasc Surg 2004;39:1157-62. 22. Steinmetz E, Rubin BG, Sanchez LA, Choi ET, Geraghty PJ, Baty J, et al. Type II endoleak after endovascular abdominal aortic aneurysm repair: a conservative approach with selective intervention is safe and cost-effective. J Vasc Surg 2004;39:306-13. 23. Parent FN, Meier GH, Godziachvili V, LeSar CJ, Parker FM, Carter KA, et al. The incidence and natural history of type I and II endoleak: a 5-year follow-up assessment with color duplex ultrasound scan. J Vasc Surg 2002;35:474-81. 24. Bonvini R, Alerci M, Antonucci F, Tutta P, Wyttenbach R, Bogen M, et al. Preoperative embolization of collateral side branches: a valid means to reduce type II endoleaks after endovascular AAA repair. J Endovasc Ther 2003;10:227-32. 25. Axelrod DJ, Lookstein RA, Guller J, Nowakowski FS, Ellozy S, Carroccio A, et al. Inferior mesenteric artery embolization before endovascular aneurysm repair: technique and initial results. J Vasc Interv Radiol 2004;15:1263-7. 26. Muthu C, Maani J, Plank LD, Holden A, Hill A. Strategies to reduce the rate of type II endoleaks: routine intraoperative embolization of the inferior mesenteric artery and thrombin injection into the aneurysm sac. J Endovasc Ther 2007;14:661-8.
Submitted Oct 15, 2009; accepted Feb 7, 2010.