What is the clinical utility of a 6-month computed tomography in the follow-up of endovascular aneurysm repair patients?
Article Outline
Objective
A drawback of endovascular aneurysm repair (EVAR) is the need for ongoing surveillance. Follow-up schedules including 1-, 6-, and 12-month computed tomography (CT) established by regulatory trials have been carried into clinical practice without critical assessment. The utility of a 6-month CT, with its associated radiation exposure and contrast toxicity, obtained after a normal result at 1-month CT has not been established.
Methods
All EVAR patients from 1996 to 2004 at one institution with complete local 1-year follow-up were reviewed for clinically significant CT findings at 1, 6, and 12 months. Before 2000, all patients underwent 1-, 6-, and 12-month CT. In 2000, a policy of omitting the 6-month CT in patients who had a normal result on the 1-month scan was adopted.
Results
During the study period, 573 patients underwent EVAR, and 376 patients who had complete local 1-year follow-up were included in this review. All had a 1-month CT scan and the result was abnormal in 40 (10.6%): five had type 1 leaks (1.3%), 34 had type 2 leaks (9.0%), and one had a type 3 leak (0.3%); all were followed with 6-month CT. The 1-month CT scan result was normal for 336 (89.4%) patients. Of these, group I (130 patients, 67 treated after 2000) underwent routine 6-month CT, with only two abnormalities noted (1.5%); both were type 2 endoleaks not associated with sac growth. No 6-month CT in this group demonstrated findings warranting intervention. The 6-month CT was omitted in group II (206 patients, all treated after 2000), and follow-up was only at 1 year. In this group, no patient's management would have been altered by findings on a 6-month CT. No patient in either group experienced aneurysm sac growth by 1 year. Clinical complications occurred in three group I patients (2.3%): seroma, limb occlusion, and main body thrombosis. Only one group II patient (0.5%) experienced a complication ≤1 year, a limb occlusion at 9 months.
Conclusions
After EVAR, a 6-month CT after a normal 1-month CT result does not identify any clinically significant findings warranting intervention and can be omitted safely from the follow-up schedule.
Unlike standard open repair, endovascular aneurysm repair (EVAR) is associated with a unique set of complications, including endoleak, device migration, and material fatigue, that mandate ongoing postoperative surveillance.1, 2, 3, 4 The early advantage of EVAR in mortality and morbidity is clearly diminished by the frequency and expense of continued long-term testing and reinterventions.
Computed tomography (CT) with intravenous contrast remains the most common modality currently used at most institutions in surveillance after EVAR. This regimen is not, however, without drawbacks. Radiation exposure after repeated scans can be significant.5 Often these patients have coexisting risk factors for renal insufficiency that exacerbate contrast nephrotoxicity, and contrast nephropathy can affect from 7% to 12% of patients after CT.6 Patients or families must often miss work and travel long distances to maintain their follow-up appointments. Finally, physicians, their office staffs, and the entire health care system are strained by the excessive use of surveillance resources.2
The original EVAR regulatory trials, registries, and device manufacturers recommended conservatively chosen follow-up regimens, including CT scans with intravenous contrast every 6 months, and sometimes every 3 months, even if the result on the initial postoperative scan was normal. These surveillance regimens including 1-, 6-, and 12-month CT in the first year have been carried into many clinical practices without critical assessment, resulting in a large number of scans that may not affect clinical decision making and may not be necessary for appropriate patient care.
Thresholds for intervention on abnormalities, such as type 2 endoleaks, are changing, and so must our attempts to identify them evolve.7, 8 As the first step in an ongoing effort to safely minimize the intensity and risk of EVAR surveillance, we hypothesized that although a 6-month CT may be indicated to monitor abnormalities identified on a 1-month CT, it can be omitted safely from the follow-up schedule if the early CT scan result not identify any clinically significant findings.
Methods
The conduct of this study was approved by the University of Pittsburgh Institutional Review Board. A retrospective review of patients who underwent successful elective EVAR at one University of Pittsburgh Medical Center institution between 1996 and 2004 and who had complete local 1-year follow-up was performed. Study variables collected included age, sex, size of aneurysm at repair, type of endograft used, imaging at 1-, 6-, and 12-month follow-up as applicable, and endograft-related complications up to 1 year out from surgery. All CT scans were re-reviewed for this study to measure aneurysm sac size using the minor axis method9 and to identify the presence of any complication including endoleak, device migration, or other graft abnormalities.
Before 2000, all patients underwent 1-, 6-, and 12-month visits with routine CT scanning in conjunction with abdominal radiographs. In 2000, a policy of omitting the 6-month return visit as well as the CT scan in patients who had a normal result on the 1-month CT scan was adopted. This policy was applied mostly for postmarketing EVAR procedures but not for patients enrolled in protocols that continued to dictate a more frequent follow-up. Patients with abnormal findings at 1 month all returned for a 6-month follow-up and CT imaging. Two groups of patients with normal findings at 1 month were compared. Group I included those who underwent routine 6-month follow-up, and group II patients omitted the 6-month evaluation and returned for a 1-year visit (Fig 1). Mean age, sex, mean aneurysm size, and device types used were identified in each group. Computed tomography abnormalities found at each time point, device-related complications up to 1-year follow-up, and reinterventions up to 1-year follow-up were compared between the two groups.
Fig 1.
Protocol for computed tomography (CT) surveillance after endovascular aneurysm repair (EVAR), including number and types of endoleak at each follow-up patients in group I, group II, and those with abnormal results on 1-month CT scans.
Results
During the study period, 573 patients underwent EVAR at one institution. Seven patients were excluded because they died from non-aneurysm-related causes <1 year, another 29 were lost to follow-up by 1 year, and 161 patients underwent follow-up CT at other institutions and were excluded because those scans could not be re-reviewed for this study.
The remaining 376 patients (66%) with complete local 1-year follow-up were included in this report. The proportion of included patients was the same before and after 2000. Group I had 130 patients (67 after 2000) and group II had 206 patients (all after 2000). The mean age of the study patients was 74 years (range, 52-91 years), and 85% were men. Mean minor axis diameter of the aneurysms was 56 mm (range, 40-93 mm). The baseline distribution among group I, group II, and the patients with abnormal 1-month CT scan results is reported in the Table. Although all three were similar in terms of age, sex, and aneurysm size, there was heterogeneity as far as the types of endografts used in each group. Three different endografts were equally used during their phase II trials in group I, but the Ancure (Guidant, Indianapolis, Ind) device was over-represented in group II subsequent to its postmarket release.
Table. Distribution of age, sex, aneurysm size, and device type used among group I, group II, and the group of patients with abnormal 1-month computed tomography scans
| Variable | Group I (n = 130) | Group II (n = 206) | Pa | Patients with abnormal 1-month CT scans (n = 40) |
|---|---|---|---|---|
| Age, mean years | 73.0 | 74.0 | 0.22 | 74.2 |
| Male, % | 83.1% | 86.9% | 0.71 | 87.5% |
| Minor axis aneurysm diameter, mean mm | 54.3 | 54.0 | 0.75 | 55.0 |
| No. (%) repaired with each device type | ||||
 Ancureb | 48(36.9) | 173(83.9) | .0076 | 26(65.0) |
| Excluderc | 37(28.5) | 5(2.4) | <.0001 | 2(5.0) |
| Lifepathd | 28(21.5) | 1(0.5) | <.0001 | 4(10.0) |
| AneuRxe | 2(1.5) | 18(8.7) | .007 | 3(7.5) |
| Zenithf | 2(1.5) | 9(4.4) | 0.21 | 4(10.0) |
| TriVascularg | 6(4.6) | 0(0.0) | .003 | 0(0.0) |
| Quantumh | 4(3.0) | 0(0.0) | .02 | 1(2.5) |
| Talente | 1(0.7) | 0(0.0) | <.0001 | 0(0.0) |
aValues refer to comparisons between group I and group II. P value < 0.05 was considered significant. |
bGuidant, Indianapolis, Ind. |
cW. L. Gore and Assoc., Flagstaff, Ariz. |
dEdwards Lifesciences, Irvine, Calif. |
eMedtronic, Minneapolis, Minn. |
fCook, Bloomington, Ind. |
gTriVascular, Santa Rosa, Calif. |
hCordis, Miami Lakes, Fla. |
Of the 376 total EVAR patients, 40 (10.6%) had an abnormality on their 1-month CT scan (Fig 1). Five patients had type 1 endoleaks (1.3%), 34 had type 2 (9%), and one patient had a type 3 endoleak (0.3%). Of those patients with an abnormal CT result at 1 month, 64% also had an abnormal result at 6 months.
The major focus of this study, however, is the 336 patients (89.4%) with a normal 1-month CT result. Of these, 130 (group I) underwent routine 6-month follow-up with a CT scan, whereas 206 (group II) omitted the 6-month visit and CT and were followed up only at 1 year (Fig 1). Fifty two percent of the group I patients were treated after 2000 and were generally trial patients whose protocols mandated a 6-month CT. All group II patients were treated postmarketing after 2000.
Group I
Only two of the 130 patients in group I (1.5%) had an abnormal CT result at 6 months (Fig 1). Both were new type 2 endoleaks, and neither was associated with sac growth. There were no instances of device migration, disconnection, or other device failure. More important, no interventions were performed on any of these patients as a result of the findings of the 6-month follow-up.
At 1 year, all patients returned for follow-up, but only 126 of the 130 group I patients had a CT scan. Four were not scanned for a variety of reasons, including new contrast allergy, renal insufficiency, and scheduling conflicts. At 1 year, four patients had abnormal CT results (Fig 1). One patient had a proximal type 1 endoleak that was not evident at 6 months and was repaired successfully with the placement of a proximal extension cuff. Images of the attachment site from this patient's 1-, 6-, and 12-month CT scans are shown in Fig 2. The three other abnormalities were type 2 endoleaks, one persistent from 6 months and two newly diagnosed. None were associated with sac growth. Of the two type 2 endoleaks noted at 6 months, one had apparently resolved spontaneously; again, no device migration or other failure had occurred. The repair of the type 1 leak represented the only intervention performed in this group on the basis of the 12-month CT.
Fig 2.
Proximal attachment site in the group I patient who developed a type 1 leak at (A) 1 month, (B) 6 months, and (C) at 1 year when the leak developed.
Three clinically significant device- or procedure-related complications occurred in group I by 1-year follow-up (2.3%). One patient had a groin seroma that became clinically evident at 6 months and was drained because it became symptomatic. A second patient presented with thrombosis of the main body of her endograft at 7 months, which was treated with an axillary–bifemoral bypass. This patient had a TriVascular (Trivascular, Santa Rosa, Calif) device and had no evidence of kinking or stenosis in the device at 6 months (Fig 3). A considerable amount of sac shrinkage was noted, however, with the minor axis decreasing from 54 to 38 mm. It is not clear if this extensive remodeling continued beyond 6 months and contributed to the thrombosis. In the last patient, thrombosis developed in one limb of his endograft at 1 year requiring a cross-femoral bypass, again with no abnormalities at the 6-month assessment (Fig 4).
Fig 3.
The 6-month computed tomography scan of the group I patient who developed main body thrombosis showed no evidence of stenosis at the narrowest portion of the aorta.
Fig 4.
The 6-month computed tomography scan of the group I patient who developed limb thrombosis showed no evidence of stenosis at the narrowest portion of the aorta.
Group II
Group II consisted of the 206 patients who had normal 1-month CT scans and proceeded to 12 months before their next surveillance imaging. At 12 months in this group, seven abnormal CT scans were identified (Fig 1), all of which were new type 2 endoleaks. None were associated with sac growth, and there were no instances of device migration or other failure. As dictated by treatment policies at the time, which have since changed, two of these type 2 endoleaks were coiled, one successfully, and the rest of the endoleaks all eventually resolved spontaneously.
Only one patient in group II had a clinically significant complication by 1 year (0.5%). This patient had a unilateral unsupported Ancure limb thrombosis at 9 months that was treated with lysis and then placement of a Wallstent (Boston Scientific, Natick, Mass) at an area of stenosis in the affected limb.
Discussion
Late device failures and poor aneurysm exclusion in EVAR patients may lead to ruptures and late complications. Follow-up schedules should be tailored both in method and frequency to maximize the yield of clinically significant findings while reducing risks, costs, and collection of information that does not affect patient management. Initial abnormalities are typically noted in 10% to 15% of patients,10 mostly type 2 endoleaks, and were present in only 11% of our patients. Clearly, closer follow-up of this minority of patients may be justifiable, but the policy of extending the same close scrutiny to all patients has not been evaluated until this review. The yield of abnormalities on a 6-month CT after a normal 1-month CT result is only 1.5%, and none were clinically significant. Omitting the 6-month CT in group II did not result in any untoward effect in >200 patients, casting serious doubt on the appropriateness of such routine testing.
Some caveats to our results are important to highlight given the design of this study. Clearly, inclusion in group I or group II was not based not on randomization but on the surveillance practice of our institution at the time. Furthermore, many group I patients were part of ongoing EVAR trials, whereas all group II patients were by necessity treated in the postmarketing phase. Both of these factors may introduce a certain amount of heterogeneity between the two groups as far as indications for stent grafting, anatomic morphology, and patient comorbidities. However, if anything, EVAR trial patients were over-represented in group I; with more recent patients, after market release, forming all of group II. This suggests that even with more aggressive application of EVAR, the group II patients did not suffer from omission of the 6-month CT.
The variety of EVAR devices used in our patients and the lack of uniform distribution between groups may be pertinent as well. The over-representation of the Ancure device in group II was due to the early United States Food and Drug Administration (FDA) approval process of this device. This may lead to an interpretation that the omission of the 6-month scan is only safe in patients with an Ancure device; however, the paucity of findings in all device types at the 6-month interval after a negative 1-month CT in group I suggests otherwise. We have eliminated the 6-month follow-up evaluation from all other devices since with similar results.
The recommended practice of following up EVAR patients with CT scans at 1, 6, and 12 months in the first year is part of the instructions for use of all marketed devices, carried through from the regulatory trials. Limited data have been published on the efficacy of this regimen. In fact, after examining the European Collaborators on Stent/Graft Techniques for Aortic Aneurysm Repair (EUROSTAR) registry, Leurs et al11 concluded that EVAR patients who were compliant with a high-intensity follow-up regimen had more EVAR complications than those who were less compliant with their surveillance regimen. Although such an analysis of this registry may be contaminated by other variables, the relationship between intensity of surveillance and risk of EVAR complications remains largely unknown. In fact, it may be that the optimal surveillance regimen is patient- and device-specific.
Fairman et al12 reported that the presence of any endoleak is a predictor for aneurysm sac expansion by the time of 6-month follow-up, highlighting the importance of the 6-month CT in this population. However, most of these are type 2 endoleaks, which follow a relatively benign course when not associated with sac growth, suggesting that a modified follow-up schedule may be suitable even in this population.7, 8
In group II, the 6-month follow-up visit as well as the CT scan was omitted. In the entire study population, four clinically significant complications occurred ≤1 year. A 6-month visit was of value only in detecting the seroma in group I, which was a prominent complaint of the patient and would have probably been reported in the absence of a routine visit. Most recent patients are now treated completely percutaneously, which should decrease the incidence of this complication. A 6-month visit may have been of value in detecting the one limb occlusion that occurred at 9 months in group II had it been performed. However, a 6-month visit was performed before the two device occlusions occurred in group I and was obviously of no value in preventing those complications. Both patients in group I who had device occlusions had palpable femoral pulses and no symptoms of lower extremity insufficiency at their 6-month visit.
This study focused on the first year of follow-up only, but it is interesting and not surprising to note that a small but significant number of complications occurred >1 year. The occurrence of these late complications suggests that although we have shown that elimination of a 6-month CT in these patients is safe, it remains mandatory to continue some form of indefinite surveillance, although the optimal intensity of late follow-up remains unknown.
Although most centers still rely on CT angiography for the follow-up of their EVAR patients, novel modalities for surveillance are emerging. The utility of duplex ultrasound with and without contrast enhancement for the detection of endoleaks has been studied by several groups and promises to provide a nephrotoxin-free method for detecting EVAR complications.13 It may ultimately prove to be even more sensitive than CT, given its ability to detect subtle flow characteristics within an aneurysm sac, which with CT requires multiple scans at different contrast times to be detected. We and others have previously described ultrasound imaging to be an accurate method to detect sac expansion, perhaps the most important marker for ongoing risk of rupture.14 Despite our original skepticism regarding the effective detection of endoleaks in hospital laboratories, we have updated our ultrasound equipment and moved the studies into an office-based laboratory, relying more on this modality than in previous reports.
Although its long-term utility is still being investigated, measurement of intrasac pressure using an implantable transducer has shown promise in identifying endoleaks and may provide a marker that is easier to interrogate more frequently.15
Conclusion
Advantages of EVAR compared with traditional open aneurysm repair in terms of patient quality of life and cost-effectiveness may be abrogated by intensive surveillance regimens. Furthermore, CT angiography as a follow-up modality incurs risk of contrast toxicity and radiation exposure. This study suggests that elimination of a routine 6-month CT in the follow-up of most EVAR patients is safe because it does not identify any significant clinical findings. Further study of alternative surveillance modalities and the natural history of late EVAR complications may identify safer, less frequent follow-up regimens for some subgroups of EVAR patients.
Author contributions
References
- . 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;364:843–848
- . Endovascular aneurysm repair versus open repair in patients with abdominal aortic aneurysm (EVAR trial 1): randomised controlled trial. Lancet. 2005;365:2179–2186
- . Quality of life endovascular and open AAA repair (Results of a randomised trial). Eur J Vasc Endovasc Surg. 2004;27:121–127
- . Follow-up costs increase the cost disparity between endovascular and open abdominal aortic aneurysm repair. J Vasc Surg. 2005;42:912–918
- Radiation risk: Directorate-General for the Environment of the European Commission. Referral guidelines for imaging. European Commission, Radiation Protection Report 118. 2000.
- Contrast material-induced renal failure in patients with diabetes mellitus, renal insufficiency, or both (A prospective controlled study). N Engl J Med. 1989;320:143–149
- Type II endoleak after endovascular abdominal aortic aneurysm repair: a conservative approach with selective intervention is safe and cost-effective. J Vasc Surg. 2004;39:306–313
- An 8-year experience with type II endoleaks: natural history suggests selective intervention is a safe approach. J Vasc Surg. 2006;44:453–459
- . Two-dimensional versus three-dimensional CT scan for aortic measurement. J Endovasc Ther. 2003;10:531–538
- Are type II endoleaks after endovascular aneurysm repair endograft dependent?. J Vasc Surg. 2006;43:657–661
- . What determines and are the consequences of surveillance intensity after endovascular abdominal aortic aneurysm repair?. Ann Vasc Surg. 2005;19:868–875
- . Factors predictive of early or late aneurysm sac size change following endovascular repair. J Vasc Surg. 2006;43:649–656
- Contrast-enhanced Duplex surveillance after endovascular abdominal aortic aneurysm repair: improved efficacy using a continuous infusion technique. J Vasc Surg. 2006;43:259–264
- . Color-flow duplex ultrasound scan versus computed tomographic scan in the surveillance of endovascular aneurysm repair. J Vasc Surg. 2003;38:645–651
- Initial results of wireless pressure sensing for endovascular aneurysm repair: the APEX Trial–Acute Pressure Measurement to Confirm Aneurysm Sac EXclusion. J Vasc Surg. 2007;45:236–242
Competition of interest: none.
CME article
PII: S0741-5214(08)00194-8
doi:10.1016/j.jvs.2008.01.056
© 2008 The Society for Vascular Surgery. Published by Elsevier Inc. All rights reserved.