- Email: [email protected]
Contrast Enhanced Ultrasound can Replace Computed Tomography Angiography for Surveillance After Endovascular Aortic Aneurysm Repair
Eur J Vasc Endovasc Surg (2016)
-,
1e7
Contrast Enhanced Ultrasound can Replace Computed Tomography Angiography for Surveillance After Endovascular Aortic Aneurysm Repair K.K. Bredahl
a,*
, M. Taudorf b, L. Lönn
a,b,c
, K.C. Vogt a, H. Sillesen
a,c
, J.P. Eiberg
a,c,d
a
Department of Vascular Surgery, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark Department of Interventional Radiology, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark d Copenhagen Academy for Medical Education and Simulation, Capital Region of Denmark, Copenhagen, Denmark b c
WHAT THIS PAPER ADDS Previous studies have been conducted either as duplex ultrasound (DUS) versus computed tomography angiography (CTA), or contrast enhanced (CEUS) versus CTA for endoleak detection after endovascular aneurysm repair (EVAR). In this study the performance of DUS and CEUS in post-EVAR endoleak detection was evaluated in 278 consecutive patients and the long-term clinical consequences of endoleaks not demonstrated by the two ultrasonic modalities were observed. Compared with endoleaks missed by CEUS, this study showed that endoleaks missed by DUS were more prone to have significant clinical implications. The study shows that future EVAR surveillance protocols can be carried out safely with CEUS, but the role of DUS remains limited to cases with stable residual sacs and freedom from endoleak on prior CEUS or CTA.
Objective/Background: Surveillance after endovascular aortic aneurysm repair (EVAR) is mandatory and computed tomography angiography (CTA) is considered the standard imaging modality, although patients are exposed to ionizing radiation and nephrotoxic contrast medium. The primary aim of this study was to determine the diagnostic efficacy of duplex ultrasound (DUS) and contrast enhanced ultrasound (CEUS) using CTA as the gold standard. The secondary aim was to determine the clinical consequences of endoleaks missed by DUS and CEUS, or CTA. Methods: All patients with EVAR for an aorto-iliac aneurysm between 1 August 2011 and 31 October 2014 were prospectively and consecutively enrolled. CEUS was added to the existing surveillance protocol, which included DUS, plain abdominal X-ray, and CTA at 3 and 12 months after stent implantation. Results: In 278 patients, endoleaks were detected in 68, 69, and 46 cases by CTA, CEUS, and DUS, respectively. The sensitivity and specificity of DUS and CEUS were 46% and 93%, and 85% and 95%, respectively. CEUS and CTA were diagnostically equivalent, as opposed to DUS and CTA (p ¼ .002). Endoleaks detected by CTA led to reintervention in 11 (4%) patients. These endoleaks were also detected by CEUS; however, three out of 11 patients were missed by DUS and underwent re-intervention: limb extension, re-cuff, and attempt to coil lumbar leaks. Endoleaks missed by CEUS or CTA were type II endoleaks without sac expansion. Conclusion: In surveillance programs after EVAR a diagnostic CEUS examination may replace CTA. Ó 2016 European Society for Vascular Surgery. Published by Elsevier Ltd. All rights reserved. Article history: Received 29 February 2016, Accepted 6 July 2016, Available online XXX Keywords: Aortic aneurysm, Abdominal, Ultrasonography, Imaging, Contrast enhanced, Tomography, X-ray computed, Endoleak
INTRODUCTION Lifelong surveillance after endovascular aneurysm repair (EVAR) is necessary to detect endoleak and prevent rupture.1 Thus far, computed tomography angiography (CTA) has been the preferred imaging modality for EVAR surveillance; however, it has significant drawbacks: risk of * Corresponding author. Department of Vascular Surgery RK3111, Rigshospitalet, Blegdamsvej 9, 2100 Copenhagen, Denmark. E-mail address: [email protected] (K.K. Bredahl). 1078-5884/Ó 2016 European Society for Vascular Surgery. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ejvs.2016.07.007
contrast induced nephropathy, stochastic risk of radiation induced cancer, and cost.2e4 Conversely, duplex ultrasound (DUS) is harmless but criticized for inferior endoleak detection compared with CTA.5 Ultrasound contrast agents increase the signal to noise ratio and thus the ability to discriminate persistent blood flow in the residual sac and thereby the conclusiveness of the ultrasound examination. The ability of contrast enhanced ultrasound (CEUS) to detect endoleak and direct re-intervention has been confirmed in a number of studies.6e8 However, studies comparing DUS with CEUS in detecting endoleaks are sparse.9 Moreover, the long-term clinical consequences of
Please cite this article in press as: Bredahl KK, et al., Contrast Enhanced Ultrasound can Replace Computed Tomography Angiography for Surveillance After Endovascular Aortic Aneurysm Repair, European Journal of Vascular and Endovascular Surgery (2016), http://dx.doi.org/10.1016/j.ejvs.2016.07.007
2
endoleaks missed by DUS or CEUS have not been addressed previously. The aims of this study were to compare the ability of DUS and CEUS to detect endoleaks using CTA as the gold standard, and to observe the course and consequence of undetected endoleaks in the clinical setting. MATERIALS AND METHODS Study design and patients In a cross sectional and single center design, the ability to detect endoleaks with DUS and CEUS was compared using CTA as the gold standard. In order to determine the clinical implications of endoleaks missed by DUS, CEUS, and CTA, data from clinical follow up were prospectively recorded, and retrospectively analyzed in patients in whom an endoleak was seen with one imaging modality but not mutually confirmed by the other. The post-EVAR surveillance program comprised plain abdominal X-ray, DUS, and CTA 3 and 12 months after EVAR. Standard surveillance thereafter did not include CTA and patients were therefore not recruited at time points beyond 12 months after EVAR (Fig. 1).1 CTA showing a type I or type II endoleak with sac expansion > 5 mm indicated reintervention was required. Between 1 August 2011 and 30 October 2014, CEUS was added to the standard post-EVAR surveillance at 3 or 12
K.K. Bredahl et al.
months after stent implantation. The comparisons of DUS versus CTA and CEUS versus CTA were only performed once, at the patient’s first visit. All patients were prospectively and consecutively recruited, and patients with endoleaks missed by one or two modalities were followed from recruitment until 30 October 2015, and their data were analyzed retrospectively (Fig. 1). Patients who underwent EVAR for abdominal aortic aneurysm, or aorto-iliac or iliac aneurysms were eligible. Patients were excluded if they died between EVAR and the control visit, underwent post-EVAR surveillance at another institution, had impaired renal function and were unfit for CTA, were unwilling to give written informed consent, or were seen with complete residual sac resolution on DUS (Fig. 1). CTA and ultrasound investigations were considered concurrent if they were done within 7 days. Commercially available EVAR devices (Zenith FlexÒ; Cook Medical, Bloomington, IN, USA) were inserted in all cases. The study was approved by the local ethics committee of Copenhagen (H2-2011-016). DUS One investigator (K.K.B.) with 1 year’s experience in vascular DUS, blinded to the results of CTA, performed all the DUS and CEUS investigations using a diagnostic ultrasound system
Figure 1. Study design and flow chart of patient selection. Gray circles: pre-operative scan. Green circles: 3 month standard surveillance (computed tomographic angiography [CTA], duplex ultrasound [DUS], X-ray, and contrast enhanced ultrasound [CEUS]). Blue circles: 1 year standard surveillance (CTA, DUS, X-ray, and CEUS). Yellow circles: standard surveillance beyond 1 year (follow up without CTA unless type II endoleak was seen at 1 year follow up with stable sac size). White circles: data from the last follow up visit were recorded for patients in whom CTA, CEUS, and DUS results at recruitment were in disagreement with regard to diagnosis of endoleak. Note. a Only the patient’s first visit within the recruitment period was used for the comparative analysis in Table 3. b Follow up was specifically dedicated to patients with missed endoleaks on DUS that were confirmed on CTA, and the follow up period was extended to 1 year. c Died between endovascular aneurysm repair (EVAR) and scheduled date of control. d Patients were followed at another institution. e Patients with false negative DUS endoleak diagnosis at their first visit were followed for clinical consequences. f Of 278 patients, 215 (77%) patients were enrolled at the 3 month visit, and 63 (23%) patients at their 12 month visit as the latter group had their 3 month visit before study start. g Withdrawn owing to complete residual sac resolution on DUS. Please cite this article in press as: Bredahl KK, et al., Contrast Enhanced Ultrasound can Replace Computed Tomography Angiography for Surveillance After Endovascular Aortic Aneurysm Repair, European Journal of Vascular and Endovascular Surgery (2016), http://dx.doi.org/10.1016/j.ejvs.2016.07.007
Contrast-enhanced Ultrasound in EVAR-surveillance
(iU22; Philips Medical Systems, Bothell, WA, US) and a 5 MHz curved array transducer (C5-1; Philips Medical Systems). Patients were not instructed to fast prior to examination but had at least 10 minutes of rest before DUS. Gray scale imaging was performed from the renal arteries to the common femoral artery. The maximum diameter of the residual sac was measured from leading edge adventitia of the anterior wall to the leading edge of adventitia on the posterior wall.10 Compared with the pre-operative measurement, sac expansion was defined as a 5 mm increase in the maximal diameter. The in-built wall filters for low flow on the ultrasound machine were used and pulse repetition frequency and color gain were adjusted to complete filling of the graft lumen without excessive color artifact outside. Anechoic areas in the residual sac and modular conjunctions were scrutinized. Endoleak was defined as the presence of reproducible flow between the residual sac wall and stent graft, and was confirmed by spectral Doppler assessment. CEUS A dedicated contrast application, taking advantages of a combined technique of pulse inversion harmonic imaging with power modulation at a low mechanical index of 0.05 was used. Refract dosages of 1.0 mL ultrasound contrast agent (SonoVue; Bracco, Milan, Italy) were administered into an antecubital vein through an 18 gauge needle followed by 5 mL saline flush. Another 1.0 mL contrast was added when the contrast enhancement vanished after, typically, 2 minutes. The focal zone was placed beneath the residual sac to prolong enhancement. Endoleak was defined as the presence of contrast enhancement in the residual sac outside the endograft. Contrast enhanced CTA Biphasic CTA (unenhanced and contrast enhanced, fixed delay of 55 s) was performed using a helical 64-slice CT scanner (Toshiba Medical Systems, Crawley, UK). Detector configuration was 0.5 64 mm (collimation ¼ 32 mm) with a pitch of 0.8. A bolus dose of 80 mL non-ionic iodinated contrast medium (iohexol 350 mgI/mL [Omnipaque; GE Healthcare Denmark A/S, Copenhagen, DK]) was injected into an antecubital vein at a rate of 3 mL/second. Scan reconstructions were made with a slice thickness and increment of 3 mm. One interventional radiologist (M.T.) blinded to the result of DUS and CEUS assessed multiplanar reconstructions of all CTAs, using a PACS system (Agfa Impax 5.2; Agfa-Gevaert NV, Mortsel, Belgium). The presence of endoleak was defined by attenuation inside the residual sac only seen on the contrast enhanced sequence. Maximum diameter was measured from outer to outer circumferential wall in any direction perpendicular to the direction of tortuosity.11 Statistics Demographic data are presented as mean SD. McNemar’s chi-square test was used to identify the differing abilities of
3 Table 1. Baseline characteristics of patients included in analysis (n ¼ 278). Demographics Mean SD aneurysm diameter (cm) 5.6 1.2 Mean SD age (y) 74 7 Mean SD BMI (kg/m2) 26 4 Male/female (n) 243/35 Indication for EVAR Asymptomatic (elective) 246 (88) Symptomatic (subacute) 32 (12) Rupture 0 (0) AAA 203 (73) Aorto-iliac aneurysm 42 (15) Iliac aneurysm 24 (9) Re-intervention 9 (3) EVAR procedures Trifab (bifurcated) 222 (80) Iliac branched 14 (5) Aorto-uni-iliac 17 (6) Uni-iliac 15 (5) Other a 10 (4) Note. Data are n (%) unless otherwise indicated. a Fenestrated graft (n ¼ 1), Palmaz stent due to type Ia endoleak (n ¼ 5), limb extension due to type Ib endoleak (n ¼ 3), coils due to type II endoleak (n ¼ 1).
CTA, DUS, and CEUS to detect endoleaks. A p-value .05 indicated statistical significance. Statistical analysis was carried out in SAS 9.3 (SAS Institute, Cary, NC, USA). RESULTS Patients During the study period 359 patients scheduled for postEVAR surveillance were investigated consecutively (Fig. 1). In total, 69 patients met the exclusion criteria and five cases were excluded owing to poor insonation related to obesity and mesh inlay, leaving a technical success rate of ultrasound of 98% (n ¼ 285/290) (Fig. 1). An additional seven patients were withdrawn as CEUS was not performed owing to complete residual sac shrinkage, leaving 278 patients for comparative analysis (Table 1). In 131 out of 278 patients, CEUS was performed at both 3 and 12 months post-EVAR. No adverse reactions to ultrasound contrast medium or non-ionic iodinated contrast medium were observed. Endoleak detection The gold standard, CTA, identified 68 (24%) patients with endoleaks: seven with a type I endoleaks; four with type II endoleaks with aneurysm sac expansion; and 57 with type II endoleaks without aneurysm sac expansion. Hence, CTA revealed a requirement for re-intervention in 11 patients with endoleaks (type I and type II with sac expansion), all of which were detected by CEUS, but only eight of which were detected by DUS (Table 2). Endoleaks missed by CTA and CEUS were all type II endoleaks, and nine out of 10 endoleaks missed by CEUS were localized posterior to the stent graft, using CTA as gold standard.
Please cite this article in press as: Bredahl KK, et al., Contrast Enhanced Ultrasound can Replace Computed Tomography Angiography for Surveillance After Endovascular Aortic Aneurysm Repair, European Journal of Vascular and Endovascular Surgery (2016), http://dx.doi.org/10.1016/j.ejvs.2016.07.007
4
K.K. Bredahl et al.
Table 2. Patients (n ¼ 11) directed by computed tomographic angiography (CTA) for elective re-intervention and their corresponding endoleak diagnosis at enrollment. Patients (n) Type of endoleaks Cases with discrepancy (n ¼ 3) 2 I 1 II Cases without discrepancy (n ¼ 8) 5 I
Seen by DUS
Seen by CEUS
Treatment
No No
Yes Yes
Limb extension (n ¼ 1), second cuff (n ¼ 1) DSA negative (n ¼ 1) b
a
Cuff extension (n ¼ 2), limb extension (n ¼ 1), Trifab (n ¼ 1) c, coils (n ¼ 1) d 3 II Yes Yes Thrombin injection (n ¼ 1) e, conservative (n ¼ 2) f, g Note. Patients planned for re-intervention based on CTA and the corresponding duplex ultrasound and contrast enhanced ultrasound. a Abdominal aortic aneurysm with persistent type Ia endoleak previously treated with cuff. b Patient having the same lumbar leaks coiled 12 months ago had a negative digital subtraction angiography (DSA) during new coiling attempt. c Uni-iliac stent graft in iliac aneurysm was changed to an aorto-bi-iliac system (Zenith Trifab; Cook Medical, Bloomington, IN, USA). d Iliac-branched stent graft with retrograde flow in internal iliac aneurysm. e Retrograde flow in internal iliac aneurysm at index procedure treated with coils. f No treatment owing to newly diagnosed esophageal cancer. g Retrograde flow in internal iliac aneurysm deemed impossible to coil. Yes
Yes
In 278 patients, McNemar’s chi-square test rejected the hypothesis that CEUS and CTA were diagnostically equivalent (p ¼ .002.), as opposed to CEUS compared with CTA (p ¼ .827). When DUS was replaced by CEUS, the sensitivity of endoleak detection improved from 45.6% (95% CI 33.5e 58.1%) to 85.3% (95% CI 74.6e92.7%) (p < .001) (Table 3). In the 131 patients having CTA at both 3 and 12 months, endoleak frequency decreased from 25% (54/215) to 19% (25/131) (p ¼ .195). The corresponding sensitivity of CEUS was 87.0% (95% CI 75.1e94.6%) and 84.0% (95% CI 63.9e 95.5%) at 3 and 12 months (p ¼ .717), respectively, and 48.2% (95% CI 34.3e62.2%) and 56.0% (95% CI 34.9e 75.6%) (p ¼ .516), respectively, when using DUS. Long-term results of missed endoleaks In order to evaluate the clinical consequences of a postEVAR surveillance protocol based on the performance of DUS, the 37 patients with endoleak invisible on DUS but demonstrated by CTA were analyzed. Three patients, described in the previous section, were directed for elective repair and underwent re-intervention within 30 days, all with endoleak detected by CEUS and CTA (Table 2). In the follow up period, another two patients (rupture and symptomatic) underwent open repair 147 and 863 days post-EVAR, respectively (Table 4). Notably, the concurrent CTA and CEUS showed contrast enhancement in the residual sac interpreted as type II endoleak without sac expansion; consequently, none of these patients was called for re-
intervention. Thus, these were interpreted as DUS missed endoleaks (n ¼ 2/37). Another four asymptomatic patients (n ¼ 4/37) underwent elective endovascular re-intervention during the follow up period, after evident sac expansion seen on DUS as part of the standard surveillance program 283 (9 months), 1,090 (36 months), 1,321 (44 months), and 1,372 days (46 months) post-EVAR, respectively. The remaining patients (n ¼ 28/37) had insignificant type II endoleaks (Table 4). The clinical consequence of a post-EVAR surveillance protocol based solely on CEUS without CTA was likewise analyzed. In 10 patients, CTA demonstrated endoleaks (all type II) invisible on both DUS and CEUS. During a median follow up period of 876 days (range 615e1,476) days, none of these patients required re-intervention (Table 4). In 11 cases, CEUS demonstrated type II endoleak without sac expansion that could not be confirmed by CTA (Table 3). During a median surveillance of 708 days (range 262e1,272 days), re-intervention was not required in any of these 11 patients as the standard surveillance program showed persistent type II endoleaks without sac expansion (n ¼ 3) or neither sac expansion nor endoleak (n ¼ 8). In 15 cases, DUS demonstrated type II endoleaks without sac expansion that were not confirmed by CTA. During a median surveillance of 634 days (range 317e1,370 days) days neither sac expansion nor endoleak was observed in 13 patients, and one of these patients died of mesenteric ischemia of unknown cause.
Table 3. Endoleak detection by ultrasound and contrast enhanced ultrasound compared with computed tomographic angiography. Method USþ USe Sensitivity (95% CI) % Specificity (95% CI) % NPV (95% CI) % Note. NPV ¼ negative predictive
CTþ 31 37 45.6 (33.5e58.1) 92.6 (88.5e96.0) 84.9 (79.7e89.3) value.
CTe 15 195
Method CEUSþ CEUSe Sensitivity (95% CI) % Specificity (95% CI) % NPV (95% CI) %
CTþ 58 10 85.3 (74.6e92.7) 94.8 (90.8e97.4) 95.2 (91.4e97.6)
CTe 11 199
Please cite this article in press as: Bredahl KK, et al., Contrast Enhanced Ultrasound can Replace Computed Tomography Angiography for Surveillance After Endovascular Aortic Aneurysm Repair, European Journal of Vascular and Endovascular Surgery (2016), http://dx.doi.org/10.1016/j.ejvs.2016.07.007
Contrast-enhanced Ultrasound in EVAR-surveillance Table 4. Clinical follow up of patients (n ¼ 37) in whom endoleak was undetected by duplex ultrasound (DUS) on the first control visit but was demonstrated on computed tomographic angiography (CTA). Symptomatic cases (n ¼ 2) n Rupture (147 d) a 1 Sub-acute open repair, clinical impending 1 rupture and type II endoleak (863 d) a Detected by CEUS (n ¼ 25) 3 Re-intervention as a result of CTA at recruitment (< 30 d) b 1 Type Ia endoleak, Palmaz stent (1,372 d) c Lumbar leak, coil embolized 3 (1,321, 1,090, 283 d) c Persisting type II endoleak without expansion 8 Resolved type II endoleak without sac expansion 8 Lost to follow up owing to cancer and death 2 of unknown cause Not detected by CEUS (n ¼ 10) Persisting type II endoleak without sac expansion 4 Resolved type II endoleak without sac expansion 6 Note. Clinical data of patients with initially undetected endoleaks by DUS, with a median follow up of 672 days (range 147e1,372 days), are presented. Days after recruitment are presented in brackets. a At enrolment, CTA and contrast enhanced ultrasound (CEUS) results were interpreted as type II endoleaks without sac expansion. b At the first visit, three patients were directed for re-intervention owing to endoleaks confirmed on CTA and CEUS but undetected on DUS (see also Table 2). c Patients directed to elective endovascular re-intervention in the follow up interval incited by the standard surveillance program.
DISCUSSION In 278 patients CEUS and CTA were diagnostically equivalent in detecting endoleaks in the first year after stent implantation. All endoleaks directed for re-intervention by CTA were also detected by CEUS. Endoleaks detected by CTA but missed by CEUS either resolved or were insignificant in provoking sac expansion or re-intervention. However, DUS missed endoleaks in nearly one third of patients who, according to CTA results, required re-intervention. It appears that CTA in EVAR surveillance can be replaced by CEUS, while DUS alone must be considered unreliable in endoleak detection within the first year. In total, 37 patients had endoleaks demonstrated by CTA and CEUS that were not detected by DUS. This number could be reduced significantly (to 10 patients) when DUS was supplemented by CEUS. Notably, long-term clinical follow up showed that type II endoleaks missed by CEUS did not lead to re-intervention, indicating that these endoleaks did not cause sac expansion in the long term, as opposed to type II endoleaks missed by DUS. The results are similar to those of other studies where type II endoleaks were associated with inferior sac regression and rupture.12,13 However, three of four type II endoleaks missed by DUS were insignificant, few cases (n ¼ 4) were directed for delayed re-intervention, and all cases with sac
5
expansion were seen on regular DUS control. In addition, it should be kept in mind that the clinical effect of type II endoleak treatment remains controversial, and some advocate conservative management even when the endoleak provokes sac expansion.14e16 Two cases with initially undetected endoleaks by DUS led to rupture and subacute open repair. However, these patients were (not even by CTA) directed for re-intervention, and the patient with rupture had a rather complicated history and was treated with a second extension cuff. Thus, as DUS is able to detect sac expansion reliably, EVAR surveillance based on DUS could be sufficient in cases where CTA or CEUS have shown freedom from endoleak and a stable residual sac.6,7,17 The results affect the choice of future EVAR surveillance as they dispute the findings of other studies implying that both CEUS and DUS are equally effective in detecting significant endoleaks.18 EVAR surveillance protocols using CTA as the first choice must, accordingly, be reconsidered and reserved for inconclusive cases only, where re-intervention is expected and when the patient is unsuitable for insonation. However, in certain non-European countries CEUS has not been approved for vascular imaging.19 Like other investigators, it was found herein that DUS and CEUS could be performed with a high technical success rate, but it is recognized that obesity and mesh inlay will interfere with ultrasound investigations, irrespective of ultrasound contrast availability.7 All ultrasound examinations, DUS and CEUS, were performed by the same operator with 1 year of clinical vascular ultrasound experience. Determination of a learning curve for CEUS and inter-operator variability assessment were not the purposes of this study. However, operator dependency is the Achilles heel of ultrasonography and may limit implementation in non-tertiary and low volume hospitals, and the implementation issue must be recognized as requiring future investigation. When new diagnostic methods are to be tested, a population as homogeneous as possible is preferable to be able to exclude confounders and better evaluate the effect of the method. In the present study, all patients were treated at the same institution during a relative short recruitment period, using the same graft and with consistent indications for EVAR. The surveillance was performed using the same CTA, CEUS, and DUS protocol, and all ultrasound investigations were performed by the same operator, who also interpreted the scans and used the same high end ultrasound equipment. Nevertheless, the results describing a very homogeneous population with implantation of the same device are comparable with others using different non-fenestrated endografts.6,7,20 Therefore, the device itself is probably not that important compared with the suprarenal area being a pivot point for endoleak detection. Even for highly experienced sonographers, the suprarenal area remains challenging, and ultrasonography for endoleak detection, be it DUS or CEUS, of suprarenal and fenestrated endografts must be considered questionable. CTA, as performed in this study, was probably not the optimal gold standard for endoleak detection as highlighted by the
Please cite this article in press as: Bredahl KK, et al., Contrast Enhanced Ultrasound can Replace Computed Tomography Angiography for Surveillance After Endovascular Aortic Aneurysm Repair, European Journal of Vascular and Endovascular Surgery (2016), http://dx.doi.org/10.1016/j.ejvs.2016.07.007
6
number of false positives using CEUS. Only one injected phase was performed, which could make it difficult to differentiate between types of endoleak. Hence, as the scanning time with CEUS was continued beyond the CT delay of 55 s, it is probable that more low flow endoleaks were detected with CEUS. Finally, increment and slice thickness of 3 mm instead of 1 mm were non-optimal. However, the CTA protocol used in this study is the diagnostic standard and has, to date, been found appropriate in detecting the most clinically important endoleaks and was not changed in the present research setting in order to minimize the radiation. Another possible drawback of the study design was the lack of mutual blindness between DUS and CEUS, as the same operator performed both examinations always in the same order: DUS followed by CEUS. The potential role of this order was not evaluated, but observing an endoleak on the first investigation may have biased the second towards rediscovering the leak. However, the use of two operators would have introduced operator dependency and the “DUS before CEUS set up” was similar to common clinical practice. Finally, this study outlined the clinical implications of an EVAR surveillance algorithm based on ultrasound. A DUS protocol would have missed a few patients that needed reintervention within the first year, and endoleaks would have been missed in half of the cases, and thus the efficacy of the surveillance protocol would be confined to the ability to measure sac changes. From this study, only indirect conclusions can be drawn as the long-term evaluation was retrospective in nature, and the results of the concurrent CTAs were unblinded. Ideally, future study should be conducted using a randomized design. In conclusion, this study showed that CEUS is comparable with CTA in detecting endoleak after EVAR and that DUS is less reliable than both CTA and CEUS in the first year of EVAR surveillance. CONFLICT OF INTEREST H.S. has received research grants from Philips Ultrasound. FUNDING The Danish Heart Foundation funded this project. REFERENCES 1 Moll FL, Powell JT, Fraedrich G, Verzini F, Haulon S, Waltham M, et al. Management of abdominal aortic aneurysms clinical practice guidelines of the European Society for Vascular Surgery. Eur J Vasc Endovasc Surg 2011;41(Suppl. 1):S1e58. 2 Karthikesalingam A, Page AA, Pettengell C, Hinchliffe RJ, Loftus IM, Thompson MM, et al. Heterogeneity in surveillance after endovascular aneurysm repair in the UK. Eur J Vasc Endovasc Surg 2011;42:585e90. 3 Mitchell AM, Jones AE, Tumlin J, Kline J. Incidence of contrastinduced nephropathy after contrast enhanced computed tomography in the outpatient setting. Clin J Am Soc Nephrol 2010;5:4e9.
K.K. Bredahl et al. 4 Jones C, Badger S, Boyd CS, Soong CV. The impact of radiation dose exposure during endovascular aneurysm repair on patient safety. J Vasc Surg 2010;52:298e302. 5 Mirza T, Karthikesalingam A, Jackson D, Walsh SR, Holt PJ, Hayes PD, et al. Duplex ultrasound and contrast enhanced ultrasound versus computed tomography for the detection of endoleak after EVAR: systematic review and bivariate metaanalysis. Eur J Vasc Endovasc Surg 2010;39:418e28. 6 Perini P, Sediri I, Midulla M, Delsart P, Mouton S, Gautier C, et al. Single centre prospective comparison between contrast enhanced ultrasound and computed tomography angiography after EVAR. Eur J Vasc Endovasc Surg 2011;42:797e802. 7 Ten Bosch J, Rouwet EV, Peters CTH, Jansen L, Verhagen HJM, Prins MH, et al. Contrast enhanced ultrasound versus computed tomographic angiography for surveillance of endovascular abdominal aortic aneurysm repair. J Vasc Interv Radiol 2010;21:638e43. 8 Gürtler VM, Sommer WH, Meimarakis G, Kopp R, Weidenhagen R, Reiser MF, et al. A comparison between contrast enhanced ultrasound imaging and multislice computed tomography in detecting and classifying endoleaks in the follow up after endovascular aneurysm repair. J Vasc Surg 2013;58:340e5. 9 Iezzi R, Basilico R, Giancristofaro D, Pascali D, Cotroneo AR, Storto ML. Contrast enhanced ultrasound versus color duplex ultrasound imaging in the follow up of patients after endovascular abdominal aortic aneurysm repair. J Vasc Surg 2009;49:552e60. 10 Bredahl K, Eldrup N, Meyer C, Eiberg JE, Sillesen H. Reproducibility of ECG-gated ultrasound diameter assessment of small abdominal aortic aneurysms. Eur J Vasc Endovasc Surg 2013;45:235e40. 11 Chaikof EL, Blankensteijn JD, Harris PL, White GH, Zarins CK, Bernhard VM, et al. Reporting standards for endovascular aortic aneurysm repair. J Vasc Surg 2002;35:1048e60. 12 Lal BK, Zhou W, Li Z, Kyriakides T, Matsumura J. From the Society for Vascular Surgery Predictors and outcomes of endoleaks in the Veterans Affairs Open Versus Endovascular Repair (OVER) Trial of Abdominal Aortic Aneurysms. J Vasc Surg 2016;62:1394e404. 13 Kray J, Kirk S, Franko J, Chew DK. Role of type II endoleak in sac regression after endovascular repair of infrarenal abdominal aortic aneurysms. J Vasc Surg 2015;61:869e74. 14 Sarac TP, Gibbons C, Vargas L, Liu J, Srivastava S, Bena J, et al. Long-term follow up of type II endoleak embolization reveals the need for close surveillance. J Vasc Surg 2012;55:33e40. 15 Sidloff DA, Gokani V, Stather PW, Choke E, Bown MJ, Sayers RD. Editor’s choice e type II endoleak: conservative management is a safe strategy. Eur J Vasc Endovasc Surg 2014;48:391e9. 16 Walker J, Tucker L-Y, Goodney P, Candell L, Hua H, Okuhn S, et al. Type II endoleak with or without intervention after endovascular aortic aneurysm repair does not change aneurysm related outcomes despite sac growth. J Vasc Surg 2015;62:551e61. 17 Bredahl K, Taudorf M, Long A, Lönn L, Rouet L, Ardon R, et al. Three dimensional ultrasound improves the accuracy of diameter measurement of the residual sac in EVAR patients. Eur J Vasc Endovasc Surg 2013;46:525e32. 18 Karthikesalingam A, Al-Jundi W, Jackson D, Boyle JR, Beard JD, Holt PJE, et al. Systematic review and meta-analysis of duplex ultrasonography, contrast enhanced ultrasonography or computed tomography for surveillance after endovascular aneurysm repair. Br J Surg 2012;99:1514e23.
Please cite this article in press as: Bredahl KK, et al., Contrast Enhanced Ultrasound can Replace Computed Tomography Angiography for Surveillance After Endovascular Aortic Aneurysm Repair, European Journal of Vascular and Endovascular Surgery (2016), http://dx.doi.org/10.1016/j.ejvs.2016.07.007
Contrast-enhanced Ultrasound in EVAR-surveillance 19 Appis AW, Facc MJT, Feinstein SB, Fesc F. Update on the safety and efficacy of commercial ultrasound contrast agents in cardiac applications. Echo Res Pract 2015;2:R55e62. 20 Cantisani V, Ricci P, Grazhdani H, Napoli A, Fanelli F, Catalano C, et al. Prospective comparative analysis of colour-Doppler
7 ultrasound, contrast enhanced ultrasound, computed tomography and magnetic resonance in detecting endoleak after endovascular abdominal aortic aneurysm repair. Eur J Vasc Endovasc Surg 2011;41:186e92.
Please cite this article in press as: Bredahl KK, et al., Contrast Enhanced Ultrasound can Replace Computed Tomography Angiography for Surveillance After Endovascular Aortic Aneurysm Repair, European Journal of Vascular and Endovascular Surgery (2016), http://dx.doi.org/10.1016/j.ejvs.2016.07.007
-,
1e7
Contrast Enhanced Ultrasound can Replace Computed Tomography Angiography for Surveillance After Endovascular Aortic Aneurysm Repair K.K. Bredahl
a,*
, M. Taudorf b, L. Lönn
a,b,c
, K.C. Vogt a, H. Sillesen
a,c
, J.P. Eiberg
a,c,d
a
Department of Vascular Surgery, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark Department of Interventional Radiology, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark d Copenhagen Academy for Medical Education and Simulation, Capital Region of Denmark, Copenhagen, Denmark b c
WHAT THIS PAPER ADDS Previous studies have been conducted either as duplex ultrasound (DUS) versus computed tomography angiography (CTA), or contrast enhanced (CEUS) versus CTA for endoleak detection after endovascular aneurysm repair (EVAR). In this study the performance of DUS and CEUS in post-EVAR endoleak detection was evaluated in 278 consecutive patients and the long-term clinical consequences of endoleaks not demonstrated by the two ultrasonic modalities were observed. Compared with endoleaks missed by CEUS, this study showed that endoleaks missed by DUS were more prone to have significant clinical implications. The study shows that future EVAR surveillance protocols can be carried out safely with CEUS, but the role of DUS remains limited to cases with stable residual sacs and freedom from endoleak on prior CEUS or CTA.
Objective/Background: Surveillance after endovascular aortic aneurysm repair (EVAR) is mandatory and computed tomography angiography (CTA) is considered the standard imaging modality, although patients are exposed to ionizing radiation and nephrotoxic contrast medium. The primary aim of this study was to determine the diagnostic efficacy of duplex ultrasound (DUS) and contrast enhanced ultrasound (CEUS) using CTA as the gold standard. The secondary aim was to determine the clinical consequences of endoleaks missed by DUS and CEUS, or CTA. Methods: All patients with EVAR for an aorto-iliac aneurysm between 1 August 2011 and 31 October 2014 were prospectively and consecutively enrolled. CEUS was added to the existing surveillance protocol, which included DUS, plain abdominal X-ray, and CTA at 3 and 12 months after stent implantation. Results: In 278 patients, endoleaks were detected in 68, 69, and 46 cases by CTA, CEUS, and DUS, respectively. The sensitivity and specificity of DUS and CEUS were 46% and 93%, and 85% and 95%, respectively. CEUS and CTA were diagnostically equivalent, as opposed to DUS and CTA (p ¼ .002). Endoleaks detected by CTA led to reintervention in 11 (4%) patients. These endoleaks were also detected by CEUS; however, three out of 11 patients were missed by DUS and underwent re-intervention: limb extension, re-cuff, and attempt to coil lumbar leaks. Endoleaks missed by CEUS or CTA were type II endoleaks without sac expansion. Conclusion: In surveillance programs after EVAR a diagnostic CEUS examination may replace CTA. Ó 2016 European Society for Vascular Surgery. Published by Elsevier Ltd. All rights reserved. Article history: Received 29 February 2016, Accepted 6 July 2016, Available online XXX Keywords: Aortic aneurysm, Abdominal, Ultrasonography, Imaging, Contrast enhanced, Tomography, X-ray computed, Endoleak
INTRODUCTION Lifelong surveillance after endovascular aneurysm repair (EVAR) is necessary to detect endoleak and prevent rupture.1 Thus far, computed tomography angiography (CTA) has been the preferred imaging modality for EVAR surveillance; however, it has significant drawbacks: risk of * Corresponding author. Department of Vascular Surgery RK3111, Rigshospitalet, Blegdamsvej 9, 2100 Copenhagen, Denmark. E-mail address: [email protected] (K.K. Bredahl). 1078-5884/Ó 2016 European Society for Vascular Surgery. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ejvs.2016.07.007
contrast induced nephropathy, stochastic risk of radiation induced cancer, and cost.2e4 Conversely, duplex ultrasound (DUS) is harmless but criticized for inferior endoleak detection compared with CTA.5 Ultrasound contrast agents increase the signal to noise ratio and thus the ability to discriminate persistent blood flow in the residual sac and thereby the conclusiveness of the ultrasound examination. The ability of contrast enhanced ultrasound (CEUS) to detect endoleak and direct re-intervention has been confirmed in a number of studies.6e8 However, studies comparing DUS with CEUS in detecting endoleaks are sparse.9 Moreover, the long-term clinical consequences of
Please cite this article in press as: Bredahl KK, et al., Contrast Enhanced Ultrasound can Replace Computed Tomography Angiography for Surveillance After Endovascular Aortic Aneurysm Repair, European Journal of Vascular and Endovascular Surgery (2016), http://dx.doi.org/10.1016/j.ejvs.2016.07.007
2
endoleaks missed by DUS or CEUS have not been addressed previously. The aims of this study were to compare the ability of DUS and CEUS to detect endoleaks using CTA as the gold standard, and to observe the course and consequence of undetected endoleaks in the clinical setting. MATERIALS AND METHODS Study design and patients In a cross sectional and single center design, the ability to detect endoleaks with DUS and CEUS was compared using CTA as the gold standard. In order to determine the clinical implications of endoleaks missed by DUS, CEUS, and CTA, data from clinical follow up were prospectively recorded, and retrospectively analyzed in patients in whom an endoleak was seen with one imaging modality but not mutually confirmed by the other. The post-EVAR surveillance program comprised plain abdominal X-ray, DUS, and CTA 3 and 12 months after EVAR. Standard surveillance thereafter did not include CTA and patients were therefore not recruited at time points beyond 12 months after EVAR (Fig. 1).1 CTA showing a type I or type II endoleak with sac expansion > 5 mm indicated reintervention was required. Between 1 August 2011 and 30 October 2014, CEUS was added to the standard post-EVAR surveillance at 3 or 12
K.K. Bredahl et al.
months after stent implantation. The comparisons of DUS versus CTA and CEUS versus CTA were only performed once, at the patient’s first visit. All patients were prospectively and consecutively recruited, and patients with endoleaks missed by one or two modalities were followed from recruitment until 30 October 2015, and their data were analyzed retrospectively (Fig. 1). Patients who underwent EVAR for abdominal aortic aneurysm, or aorto-iliac or iliac aneurysms were eligible. Patients were excluded if they died between EVAR and the control visit, underwent post-EVAR surveillance at another institution, had impaired renal function and were unfit for CTA, were unwilling to give written informed consent, or were seen with complete residual sac resolution on DUS (Fig. 1). CTA and ultrasound investigations were considered concurrent if they were done within 7 days. Commercially available EVAR devices (Zenith FlexÒ; Cook Medical, Bloomington, IN, USA) were inserted in all cases. The study was approved by the local ethics committee of Copenhagen (H2-2011-016). DUS One investigator (K.K.B.) with 1 year’s experience in vascular DUS, blinded to the results of CTA, performed all the DUS and CEUS investigations using a diagnostic ultrasound system
Figure 1. Study design and flow chart of patient selection. Gray circles: pre-operative scan. Green circles: 3 month standard surveillance (computed tomographic angiography [CTA], duplex ultrasound [DUS], X-ray, and contrast enhanced ultrasound [CEUS]). Blue circles: 1 year standard surveillance (CTA, DUS, X-ray, and CEUS). Yellow circles: standard surveillance beyond 1 year (follow up without CTA unless type II endoleak was seen at 1 year follow up with stable sac size). White circles: data from the last follow up visit were recorded for patients in whom CTA, CEUS, and DUS results at recruitment were in disagreement with regard to diagnosis of endoleak. Note. a Only the patient’s first visit within the recruitment period was used for the comparative analysis in Table 3. b Follow up was specifically dedicated to patients with missed endoleaks on DUS that were confirmed on CTA, and the follow up period was extended to 1 year. c Died between endovascular aneurysm repair (EVAR) and scheduled date of control. d Patients were followed at another institution. e Patients with false negative DUS endoleak diagnosis at their first visit were followed for clinical consequences. f Of 278 patients, 215 (77%) patients were enrolled at the 3 month visit, and 63 (23%) patients at their 12 month visit as the latter group had their 3 month visit before study start. g Withdrawn owing to complete residual sac resolution on DUS. Please cite this article in press as: Bredahl KK, et al., Contrast Enhanced Ultrasound can Replace Computed Tomography Angiography for Surveillance After Endovascular Aortic Aneurysm Repair, European Journal of Vascular and Endovascular Surgery (2016), http://dx.doi.org/10.1016/j.ejvs.2016.07.007
Contrast-enhanced Ultrasound in EVAR-surveillance
(iU22; Philips Medical Systems, Bothell, WA, US) and a 5 MHz curved array transducer (C5-1; Philips Medical Systems). Patients were not instructed to fast prior to examination but had at least 10 minutes of rest before DUS. Gray scale imaging was performed from the renal arteries to the common femoral artery. The maximum diameter of the residual sac was measured from leading edge adventitia of the anterior wall to the leading edge of adventitia on the posterior wall.10 Compared with the pre-operative measurement, sac expansion was defined as a 5 mm increase in the maximal diameter. The in-built wall filters for low flow on the ultrasound machine were used and pulse repetition frequency and color gain were adjusted to complete filling of the graft lumen without excessive color artifact outside. Anechoic areas in the residual sac and modular conjunctions were scrutinized. Endoleak was defined as the presence of reproducible flow between the residual sac wall and stent graft, and was confirmed by spectral Doppler assessment. CEUS A dedicated contrast application, taking advantages of a combined technique of pulse inversion harmonic imaging with power modulation at a low mechanical index of 0.05 was used. Refract dosages of 1.0 mL ultrasound contrast agent (SonoVue; Bracco, Milan, Italy) were administered into an antecubital vein through an 18 gauge needle followed by 5 mL saline flush. Another 1.0 mL contrast was added when the contrast enhancement vanished after, typically, 2 minutes. The focal zone was placed beneath the residual sac to prolong enhancement. Endoleak was defined as the presence of contrast enhancement in the residual sac outside the endograft. Contrast enhanced CTA Biphasic CTA (unenhanced and contrast enhanced, fixed delay of 55 s) was performed using a helical 64-slice CT scanner (Toshiba Medical Systems, Crawley, UK). Detector configuration was 0.5 64 mm (collimation ¼ 32 mm) with a pitch of 0.8. A bolus dose of 80 mL non-ionic iodinated contrast medium (iohexol 350 mgI/mL [Omnipaque; GE Healthcare Denmark A/S, Copenhagen, DK]) was injected into an antecubital vein at a rate of 3 mL/second. Scan reconstructions were made with a slice thickness and increment of 3 mm. One interventional radiologist (M.T.) blinded to the result of DUS and CEUS assessed multiplanar reconstructions of all CTAs, using a PACS system (Agfa Impax 5.2; Agfa-Gevaert NV, Mortsel, Belgium). The presence of endoleak was defined by attenuation inside the residual sac only seen on the contrast enhanced sequence. Maximum diameter was measured from outer to outer circumferential wall in any direction perpendicular to the direction of tortuosity.11 Statistics Demographic data are presented as mean SD. McNemar’s chi-square test was used to identify the differing abilities of
3 Table 1. Baseline characteristics of patients included in analysis (n ¼ 278). Demographics Mean SD aneurysm diameter (cm) 5.6 1.2 Mean SD age (y) 74 7 Mean SD BMI (kg/m2) 26 4 Male/female (n) 243/35 Indication for EVAR Asymptomatic (elective) 246 (88) Symptomatic (subacute) 32 (12) Rupture 0 (0) AAA 203 (73) Aorto-iliac aneurysm 42 (15) Iliac aneurysm 24 (9) Re-intervention 9 (3) EVAR procedures Trifab (bifurcated) 222 (80) Iliac branched 14 (5) Aorto-uni-iliac 17 (6) Uni-iliac 15 (5) Other a 10 (4) Note. Data are n (%) unless otherwise indicated. a Fenestrated graft (n ¼ 1), Palmaz stent due to type Ia endoleak (n ¼ 5), limb extension due to type Ib endoleak (n ¼ 3), coils due to type II endoleak (n ¼ 1).
CTA, DUS, and CEUS to detect endoleaks. A p-value .05 indicated statistical significance. Statistical analysis was carried out in SAS 9.3 (SAS Institute, Cary, NC, USA). RESULTS Patients During the study period 359 patients scheduled for postEVAR surveillance were investigated consecutively (Fig. 1). In total, 69 patients met the exclusion criteria and five cases were excluded owing to poor insonation related to obesity and mesh inlay, leaving a technical success rate of ultrasound of 98% (n ¼ 285/290) (Fig. 1). An additional seven patients were withdrawn as CEUS was not performed owing to complete residual sac shrinkage, leaving 278 patients for comparative analysis (Table 1). In 131 out of 278 patients, CEUS was performed at both 3 and 12 months post-EVAR. No adverse reactions to ultrasound contrast medium or non-ionic iodinated contrast medium were observed. Endoleak detection The gold standard, CTA, identified 68 (24%) patients with endoleaks: seven with a type I endoleaks; four with type II endoleaks with aneurysm sac expansion; and 57 with type II endoleaks without aneurysm sac expansion. Hence, CTA revealed a requirement for re-intervention in 11 patients with endoleaks (type I and type II with sac expansion), all of which were detected by CEUS, but only eight of which were detected by DUS (Table 2). Endoleaks missed by CTA and CEUS were all type II endoleaks, and nine out of 10 endoleaks missed by CEUS were localized posterior to the stent graft, using CTA as gold standard.
Please cite this article in press as: Bredahl KK, et al., Contrast Enhanced Ultrasound can Replace Computed Tomography Angiography for Surveillance After Endovascular Aortic Aneurysm Repair, European Journal of Vascular and Endovascular Surgery (2016), http://dx.doi.org/10.1016/j.ejvs.2016.07.007
4
K.K. Bredahl et al.
Table 2. Patients (n ¼ 11) directed by computed tomographic angiography (CTA) for elective re-intervention and their corresponding endoleak diagnosis at enrollment. Patients (n) Type of endoleaks Cases with discrepancy (n ¼ 3) 2 I 1 II Cases without discrepancy (n ¼ 8) 5 I
Seen by DUS
Seen by CEUS
Treatment
No No
Yes Yes
Limb extension (n ¼ 1), second cuff (n ¼ 1) DSA negative (n ¼ 1) b
a
Cuff extension (n ¼ 2), limb extension (n ¼ 1), Trifab (n ¼ 1) c, coils (n ¼ 1) d 3 II Yes Yes Thrombin injection (n ¼ 1) e, conservative (n ¼ 2) f, g Note. Patients planned for re-intervention based on CTA and the corresponding duplex ultrasound and contrast enhanced ultrasound. a Abdominal aortic aneurysm with persistent type Ia endoleak previously treated with cuff. b Patient having the same lumbar leaks coiled 12 months ago had a negative digital subtraction angiography (DSA) during new coiling attempt. c Uni-iliac stent graft in iliac aneurysm was changed to an aorto-bi-iliac system (Zenith Trifab; Cook Medical, Bloomington, IN, USA). d Iliac-branched stent graft with retrograde flow in internal iliac aneurysm. e Retrograde flow in internal iliac aneurysm at index procedure treated with coils. f No treatment owing to newly diagnosed esophageal cancer. g Retrograde flow in internal iliac aneurysm deemed impossible to coil. Yes
Yes
In 278 patients, McNemar’s chi-square test rejected the hypothesis that CEUS and CTA were diagnostically equivalent (p ¼ .002.), as opposed to CEUS compared with CTA (p ¼ .827). When DUS was replaced by CEUS, the sensitivity of endoleak detection improved from 45.6% (95% CI 33.5e 58.1%) to 85.3% (95% CI 74.6e92.7%) (p < .001) (Table 3). In the 131 patients having CTA at both 3 and 12 months, endoleak frequency decreased from 25% (54/215) to 19% (25/131) (p ¼ .195). The corresponding sensitivity of CEUS was 87.0% (95% CI 75.1e94.6%) and 84.0% (95% CI 63.9e 95.5%) at 3 and 12 months (p ¼ .717), respectively, and 48.2% (95% CI 34.3e62.2%) and 56.0% (95% CI 34.9e 75.6%) (p ¼ .516), respectively, when using DUS. Long-term results of missed endoleaks In order to evaluate the clinical consequences of a postEVAR surveillance protocol based on the performance of DUS, the 37 patients with endoleak invisible on DUS but demonstrated by CTA were analyzed. Three patients, described in the previous section, were directed for elective repair and underwent re-intervention within 30 days, all with endoleak detected by CEUS and CTA (Table 2). In the follow up period, another two patients (rupture and symptomatic) underwent open repair 147 and 863 days post-EVAR, respectively (Table 4). Notably, the concurrent CTA and CEUS showed contrast enhancement in the residual sac interpreted as type II endoleak without sac expansion; consequently, none of these patients was called for re-
intervention. Thus, these were interpreted as DUS missed endoleaks (n ¼ 2/37). Another four asymptomatic patients (n ¼ 4/37) underwent elective endovascular re-intervention during the follow up period, after evident sac expansion seen on DUS as part of the standard surveillance program 283 (9 months), 1,090 (36 months), 1,321 (44 months), and 1,372 days (46 months) post-EVAR, respectively. The remaining patients (n ¼ 28/37) had insignificant type II endoleaks (Table 4). The clinical consequence of a post-EVAR surveillance protocol based solely on CEUS without CTA was likewise analyzed. In 10 patients, CTA demonstrated endoleaks (all type II) invisible on both DUS and CEUS. During a median follow up period of 876 days (range 615e1,476) days, none of these patients required re-intervention (Table 4). In 11 cases, CEUS demonstrated type II endoleak without sac expansion that could not be confirmed by CTA (Table 3). During a median surveillance of 708 days (range 262e1,272 days), re-intervention was not required in any of these 11 patients as the standard surveillance program showed persistent type II endoleaks without sac expansion (n ¼ 3) or neither sac expansion nor endoleak (n ¼ 8). In 15 cases, DUS demonstrated type II endoleaks without sac expansion that were not confirmed by CTA. During a median surveillance of 634 days (range 317e1,370 days) days neither sac expansion nor endoleak was observed in 13 patients, and one of these patients died of mesenteric ischemia of unknown cause.
Table 3. Endoleak detection by ultrasound and contrast enhanced ultrasound compared with computed tomographic angiography. Method USþ USe Sensitivity (95% CI) % Specificity (95% CI) % NPV (95% CI) % Note. NPV ¼ negative predictive
CTþ 31 37 45.6 (33.5e58.1) 92.6 (88.5e96.0) 84.9 (79.7e89.3) value.
CTe 15 195
Method CEUSþ CEUSe Sensitivity (95% CI) % Specificity (95% CI) % NPV (95% CI) %
CTþ 58 10 85.3 (74.6e92.7) 94.8 (90.8e97.4) 95.2 (91.4e97.6)
CTe 11 199
Please cite this article in press as: Bredahl KK, et al., Contrast Enhanced Ultrasound can Replace Computed Tomography Angiography for Surveillance After Endovascular Aortic Aneurysm Repair, European Journal of Vascular and Endovascular Surgery (2016), http://dx.doi.org/10.1016/j.ejvs.2016.07.007
Contrast-enhanced Ultrasound in EVAR-surveillance Table 4. Clinical follow up of patients (n ¼ 37) in whom endoleak was undetected by duplex ultrasound (DUS) on the first control visit but was demonstrated on computed tomographic angiography (CTA). Symptomatic cases (n ¼ 2) n Rupture (147 d) a 1 Sub-acute open repair, clinical impending 1 rupture and type II endoleak (863 d) a Detected by CEUS (n ¼ 25) 3 Re-intervention as a result of CTA at recruitment (< 30 d) b 1 Type Ia endoleak, Palmaz stent (1,372 d) c Lumbar leak, coil embolized 3 (1,321, 1,090, 283 d) c Persisting type II endoleak without expansion 8 Resolved type II endoleak without sac expansion 8 Lost to follow up owing to cancer and death 2 of unknown cause Not detected by CEUS (n ¼ 10) Persisting type II endoleak without sac expansion 4 Resolved type II endoleak without sac expansion 6 Note. Clinical data of patients with initially undetected endoleaks by DUS, with a median follow up of 672 days (range 147e1,372 days), are presented. Days after recruitment are presented in brackets. a At enrolment, CTA and contrast enhanced ultrasound (CEUS) results were interpreted as type II endoleaks without sac expansion. b At the first visit, three patients were directed for re-intervention owing to endoleaks confirmed on CTA and CEUS but undetected on DUS (see also Table 2). c Patients directed to elective endovascular re-intervention in the follow up interval incited by the standard surveillance program.
DISCUSSION In 278 patients CEUS and CTA were diagnostically equivalent in detecting endoleaks in the first year after stent implantation. All endoleaks directed for re-intervention by CTA were also detected by CEUS. Endoleaks detected by CTA but missed by CEUS either resolved or were insignificant in provoking sac expansion or re-intervention. However, DUS missed endoleaks in nearly one third of patients who, according to CTA results, required re-intervention. It appears that CTA in EVAR surveillance can be replaced by CEUS, while DUS alone must be considered unreliable in endoleak detection within the first year. In total, 37 patients had endoleaks demonstrated by CTA and CEUS that were not detected by DUS. This number could be reduced significantly (to 10 patients) when DUS was supplemented by CEUS. Notably, long-term clinical follow up showed that type II endoleaks missed by CEUS did not lead to re-intervention, indicating that these endoleaks did not cause sac expansion in the long term, as opposed to type II endoleaks missed by DUS. The results are similar to those of other studies where type II endoleaks were associated with inferior sac regression and rupture.12,13 However, three of four type II endoleaks missed by DUS were insignificant, few cases (n ¼ 4) were directed for delayed re-intervention, and all cases with sac
5
expansion were seen on regular DUS control. In addition, it should be kept in mind that the clinical effect of type II endoleak treatment remains controversial, and some advocate conservative management even when the endoleak provokes sac expansion.14e16 Two cases with initially undetected endoleaks by DUS led to rupture and subacute open repair. However, these patients were (not even by CTA) directed for re-intervention, and the patient with rupture had a rather complicated history and was treated with a second extension cuff. Thus, as DUS is able to detect sac expansion reliably, EVAR surveillance based on DUS could be sufficient in cases where CTA or CEUS have shown freedom from endoleak and a stable residual sac.6,7,17 The results affect the choice of future EVAR surveillance as they dispute the findings of other studies implying that both CEUS and DUS are equally effective in detecting significant endoleaks.18 EVAR surveillance protocols using CTA as the first choice must, accordingly, be reconsidered and reserved for inconclusive cases only, where re-intervention is expected and when the patient is unsuitable for insonation. However, in certain non-European countries CEUS has not been approved for vascular imaging.19 Like other investigators, it was found herein that DUS and CEUS could be performed with a high technical success rate, but it is recognized that obesity and mesh inlay will interfere with ultrasound investigations, irrespective of ultrasound contrast availability.7 All ultrasound examinations, DUS and CEUS, were performed by the same operator with 1 year of clinical vascular ultrasound experience. Determination of a learning curve for CEUS and inter-operator variability assessment were not the purposes of this study. However, operator dependency is the Achilles heel of ultrasonography and may limit implementation in non-tertiary and low volume hospitals, and the implementation issue must be recognized as requiring future investigation. When new diagnostic methods are to be tested, a population as homogeneous as possible is preferable to be able to exclude confounders and better evaluate the effect of the method. In the present study, all patients were treated at the same institution during a relative short recruitment period, using the same graft and with consistent indications for EVAR. The surveillance was performed using the same CTA, CEUS, and DUS protocol, and all ultrasound investigations were performed by the same operator, who also interpreted the scans and used the same high end ultrasound equipment. Nevertheless, the results describing a very homogeneous population with implantation of the same device are comparable with others using different non-fenestrated endografts.6,7,20 Therefore, the device itself is probably not that important compared with the suprarenal area being a pivot point for endoleak detection. Even for highly experienced sonographers, the suprarenal area remains challenging, and ultrasonography for endoleak detection, be it DUS or CEUS, of suprarenal and fenestrated endografts must be considered questionable. CTA, as performed in this study, was probably not the optimal gold standard for endoleak detection as highlighted by the
Please cite this article in press as: Bredahl KK, et al., Contrast Enhanced Ultrasound can Replace Computed Tomography Angiography for Surveillance After Endovascular Aortic Aneurysm Repair, European Journal of Vascular and Endovascular Surgery (2016), http://dx.doi.org/10.1016/j.ejvs.2016.07.007
6
number of false positives using CEUS. Only one injected phase was performed, which could make it difficult to differentiate between types of endoleak. Hence, as the scanning time with CEUS was continued beyond the CT delay of 55 s, it is probable that more low flow endoleaks were detected with CEUS. Finally, increment and slice thickness of 3 mm instead of 1 mm were non-optimal. However, the CTA protocol used in this study is the diagnostic standard and has, to date, been found appropriate in detecting the most clinically important endoleaks and was not changed in the present research setting in order to minimize the radiation. Another possible drawback of the study design was the lack of mutual blindness between DUS and CEUS, as the same operator performed both examinations always in the same order: DUS followed by CEUS. The potential role of this order was not evaluated, but observing an endoleak on the first investigation may have biased the second towards rediscovering the leak. However, the use of two operators would have introduced operator dependency and the “DUS before CEUS set up” was similar to common clinical practice. Finally, this study outlined the clinical implications of an EVAR surveillance algorithm based on ultrasound. A DUS protocol would have missed a few patients that needed reintervention within the first year, and endoleaks would have been missed in half of the cases, and thus the efficacy of the surveillance protocol would be confined to the ability to measure sac changes. From this study, only indirect conclusions can be drawn as the long-term evaluation was retrospective in nature, and the results of the concurrent CTAs were unblinded. Ideally, future study should be conducted using a randomized design. In conclusion, this study showed that CEUS is comparable with CTA in detecting endoleak after EVAR and that DUS is less reliable than both CTA and CEUS in the first year of EVAR surveillance. CONFLICT OF INTEREST H.S. has received research grants from Philips Ultrasound. FUNDING The Danish Heart Foundation funded this project. REFERENCES 1 Moll FL, Powell JT, Fraedrich G, Verzini F, Haulon S, Waltham M, et al. Management of abdominal aortic aneurysms clinical practice guidelines of the European Society for Vascular Surgery. Eur J Vasc Endovasc Surg 2011;41(Suppl. 1):S1e58. 2 Karthikesalingam A, Page AA, Pettengell C, Hinchliffe RJ, Loftus IM, Thompson MM, et al. Heterogeneity in surveillance after endovascular aneurysm repair in the UK. Eur J Vasc Endovasc Surg 2011;42:585e90. 3 Mitchell AM, Jones AE, Tumlin J, Kline J. Incidence of contrastinduced nephropathy after contrast enhanced computed tomography in the outpatient setting. Clin J Am Soc Nephrol 2010;5:4e9.
K.K. Bredahl et al. 4 Jones C, Badger S, Boyd CS, Soong CV. The impact of radiation dose exposure during endovascular aneurysm repair on patient safety. J Vasc Surg 2010;52:298e302. 5 Mirza T, Karthikesalingam A, Jackson D, Walsh SR, Holt PJ, Hayes PD, et al. Duplex ultrasound and contrast enhanced ultrasound versus computed tomography for the detection of endoleak after EVAR: systematic review and bivariate metaanalysis. Eur J Vasc Endovasc Surg 2010;39:418e28. 6 Perini P, Sediri I, Midulla M, Delsart P, Mouton S, Gautier C, et al. Single centre prospective comparison between contrast enhanced ultrasound and computed tomography angiography after EVAR. Eur J Vasc Endovasc Surg 2011;42:797e802. 7 Ten Bosch J, Rouwet EV, Peters CTH, Jansen L, Verhagen HJM, Prins MH, et al. Contrast enhanced ultrasound versus computed tomographic angiography for surveillance of endovascular abdominal aortic aneurysm repair. J Vasc Interv Radiol 2010;21:638e43. 8 Gürtler VM, Sommer WH, Meimarakis G, Kopp R, Weidenhagen R, Reiser MF, et al. A comparison between contrast enhanced ultrasound imaging and multislice computed tomography in detecting and classifying endoleaks in the follow up after endovascular aneurysm repair. J Vasc Surg 2013;58:340e5. 9 Iezzi R, Basilico R, Giancristofaro D, Pascali D, Cotroneo AR, Storto ML. Contrast enhanced ultrasound versus color duplex ultrasound imaging in the follow up of patients after endovascular abdominal aortic aneurysm repair. J Vasc Surg 2009;49:552e60. 10 Bredahl K, Eldrup N, Meyer C, Eiberg JE, Sillesen H. Reproducibility of ECG-gated ultrasound diameter assessment of small abdominal aortic aneurysms. Eur J Vasc Endovasc Surg 2013;45:235e40. 11 Chaikof EL, Blankensteijn JD, Harris PL, White GH, Zarins CK, Bernhard VM, et al. Reporting standards for endovascular aortic aneurysm repair. J Vasc Surg 2002;35:1048e60. 12 Lal BK, Zhou W, Li Z, Kyriakides T, Matsumura J. From the Society for Vascular Surgery Predictors and outcomes of endoleaks in the Veterans Affairs Open Versus Endovascular Repair (OVER) Trial of Abdominal Aortic Aneurysms. J Vasc Surg 2016;62:1394e404. 13 Kray J, Kirk S, Franko J, Chew DK. Role of type II endoleak in sac regression after endovascular repair of infrarenal abdominal aortic aneurysms. J Vasc Surg 2015;61:869e74. 14 Sarac TP, Gibbons C, Vargas L, Liu J, Srivastava S, Bena J, et al. Long-term follow up of type II endoleak embolization reveals the need for close surveillance. J Vasc Surg 2012;55:33e40. 15 Sidloff DA, Gokani V, Stather PW, Choke E, Bown MJ, Sayers RD. Editor’s choice e type II endoleak: conservative management is a safe strategy. Eur J Vasc Endovasc Surg 2014;48:391e9. 16 Walker J, Tucker L-Y, Goodney P, Candell L, Hua H, Okuhn S, et al. Type II endoleak with or without intervention after endovascular aortic aneurysm repair does not change aneurysm related outcomes despite sac growth. J Vasc Surg 2015;62:551e61. 17 Bredahl K, Taudorf M, Long A, Lönn L, Rouet L, Ardon R, et al. Three dimensional ultrasound improves the accuracy of diameter measurement of the residual sac in EVAR patients. Eur J Vasc Endovasc Surg 2013;46:525e32. 18 Karthikesalingam A, Al-Jundi W, Jackson D, Boyle JR, Beard JD, Holt PJE, et al. Systematic review and meta-analysis of duplex ultrasonography, contrast enhanced ultrasonography or computed tomography for surveillance after endovascular aneurysm repair. Br J Surg 2012;99:1514e23.
Please cite this article in press as: Bredahl KK, et al., Contrast Enhanced Ultrasound can Replace Computed Tomography Angiography for Surveillance After Endovascular Aortic Aneurysm Repair, European Journal of Vascular and Endovascular Surgery (2016), http://dx.doi.org/10.1016/j.ejvs.2016.07.007
Contrast-enhanced Ultrasound in EVAR-surveillance 19 Appis AW, Facc MJT, Feinstein SB, Fesc F. Update on the safety and efficacy of commercial ultrasound contrast agents in cardiac applications. Echo Res Pract 2015;2:R55e62. 20 Cantisani V, Ricci P, Grazhdani H, Napoli A, Fanelli F, Catalano C, et al. Prospective comparative analysis of colour-Doppler
7 ultrasound, contrast enhanced ultrasound, computed tomography and magnetic resonance in detecting endoleak after endovascular abdominal aortic aneurysm repair. Eur J Vasc Endovasc Surg 2011;41:186e92.
Please cite this article in press as: Bredahl KK, et al., Contrast Enhanced Ultrasound can Replace Computed Tomography Angiography for Surveillance After Endovascular Aortic Aneurysm Repair, European Journal of Vascular and Endovascular Surgery (2016), http://dx.doi.org/10.1016/j.ejvs.2016.07.007