Follow-up of abdominal aortic aneurysm after endovascular aortic repair: Comparison of volumetric and diametric measurement

Materials and methods 73 consecutive patients (2 females, 71 males; age 38–84 years; mean age, 69.1 ± 8 years) with AAA were treated with percutaneous EVAR in a single institution. For follow-up, CTA was performed periodically after EVAR. Images were evaluated for maximal diameter in consensus by two experienced radiologists. Using OsirixTM, volumetric measurements were done by one radiologist, including the entire infrarenal abdominal aorta. Results In 73 patients 220 CTA examinations were performed after EVAR with a mean follow-up of 17.3 months (range, 1.8–42.7 months). The mean postinterventional volume of aneurysm was 165.63 ml ± 93.29 ml (range, 47.94–565.67 ml). The mean maximal postinterventional diameter was 5.91 ± 1.52 cm (range, 3.72–13.82 cm). At large over the entire observation period a slight, non-significant decrease of 1.6% (2.58 ml ± 69.05 ml, range 82.82–201.92 ml) in volumes and a 9.3% (mean 0.55 cm ± 1.22 cm, range 2.85–1.93 cm) in diameters were observed. For all examinations a high correlation of volume and diameter was calculated ( r = 0.813–0.905; α < 0.01). Conclusion For follow-up of abdominal EVAR using CTA there is a high correlation between volumetric and diametric measurements of aneurysm. Based on a daily clinical routine setting, measurements of maximal diameters in cross sectional imaging of AAA after EVAR seems to be sufficient to exclude post interventional enlargement. Keywords Angiography EVAR CTA Volume Diameter 1 Introduction Abdominal aortic aneurysm (AAA) is a common disease particularly in male smokers with hypertension older than 65 years [1,2] . An increasing incidence can be attributed to improving non-invasive diagnostic modalities, like ultrasonography (US) and computed tomography angiography (CTA), as well as to demographic aging. Since the natural course of untreated AAA is aneurysm enlargement with an increasing risk of rupture [3] , a timely indication for treatment is mandatory. Endovascular aortic aneurysm repair (EVAR) became a well-established alternative to open surgical repair (OSR) for patients with AAA in the last decade and was introduced by Parodi et al. in 1991 and [4,5] . During the procedure, a covered endograft is placed in the aorta and anchored to the proximal and distal necks of the non-aneurysmatic portions. By excluding the aneurysm sac from systemic blood pressure, the conduit prevents aneurysm enlargement and rupture. High technical success rates, low perioperative morbidity and mortality compared to OSR particularly in elderly and multimorbid patients have been demonstrated in several studies (4–7). Nonetheless, there is an increasing body of evidence of mid- and long-term complications such as persisting endoleaks and endograft failure with frequent need for re-interventions. Therefore, a lifelong surveillance is advocated [6,7] . Successful aneurysm exclusion is expected to lead to shrinkage of the aneurysm sac ( Fig. 1 ). Vice versa, with stable or expanding volumes of the aneurysm sac, a persisting endograft failure can be suspected. In postinterventional monitoring, measurement of maximal orthogonal aneurysm diameter is a simple method proposed in former studies and used in many clinical settings [3] . More recently, the method of three-dimensional volume assessment has been introduced as an alternative, claiming earlier definition of success and better identification of patients requiring closer surveillance or earlier re-intervention [8,9] . The purpose of our retrospective study was (I.) to correlate maximal orthogonal diameter measurements to volumetric measurements of AAA after EVAR using CTA and (II.) to investigate which method of measurement is superior in monitoring patients during follow-up. 2 Materials and methods The local ethics committee approved this retrospective single centre study. The study included 73 consecutive patients (2 female, 71 male) with non-ruptured AAA who underwent EVAR between January 2006 and October 2009 at our institution. The mean age of patients was 69.1 ± 8 years (range 38–84 years). Three different stent-grafts were used (Endurant™, Medtronic Inc., Minneapolis; Talent™, Medtronic Inc., Minneapolis; Excluder™, Gore and Associates Inc., Arizona). The choice of the graft in each case was based on patient anatomy. Prior to interventions, all patients gave their written informed consent. The procedure was performed under general anaesthesia in an angiography suite equipped with standard fluoroscopy (Integris Allura, Philips Medical Systems, Best, The Netherlands) by a team including an experienced interventional radiologist, a senior radiology intern and one radiographer. The patient was in supine position. 5000 units of non-fractionated heparin were administered intraarterially before insertion of the delivery sheath. Either bilateral surgical arteriotomy or, more often, percutaneous access of the common femoral artery was chosen and standard EVAR procedure was performed ( Fig. 2 ). Imaging follow-up was performed by computed tomography angiography (Brilliance 64, Philips Medical Systems, Best, The Netherlands) before discharge (<3 days postinterventional), at three months, six months, twelve months and annually thereafter, unless closer intervals became necessary. If required, patients underwent digital subtraction angiography and were attended to adequate treatment. CTA was performed from the celiac artery to the common femoral arteries before and after intravenous contrast administration (Imeron 300, Bracco Imaging, Konstanz, Germany). A delayed phase was acquired at a dose of 120 ml with a flow rate of 4.0 ml/s, scan delay was 90 s. Acquisition parameters were 1.5 mm collimation, 3.0 mm reconstruction spacing and variable pitch. Images were processed with a dedicated software package on an independent picture archiving and communication system (PACS) workstation equipped with 2K-monitors. All scans were assessed for evidence of stent fracture, migration, aneurysm expansion and endoleak. On each occasion, images were evaluated for maximal diameter in consensus by two experienced radiologists. Using Osirix™, volumetric measurements were done by one radiologist, including the entire infrarenal aorta. The data were acquired with Excel™ (Microsoft™, Redmond, USA). Statistical analysis was performed with PASW Statistics 18™ (SPSS Inc., Chicago, USA) comparing the following observational periods: (1) postinterventional period (<3 days after intervention), (2) within first year, (3) within second year and (4) after second year following EVAR. The postinterventional CTA control (1) represented our baseline for evaluating diameter and volume changes. Descriptive statistics (proportions, means, standard deviations) were calculated for patient characteristics. Moreover, the presence or absence of an endoleak was noted at each investigation. Primary endoleaks were diagnosed when found in the postinterventional control, secondary endoleaks when occurring later at follow-up. If present, the type of endoleak was described according to the classification of White and colleagues [10] . Additionally endotension was defined as aneurysm sac increase without evidence of persistent endoleak [11] . Volume changes were expressed percental to the postinterventional volume ( V 1) at referral, using the following formula: [( V 1 − Vx )/V1] × 100. Absolute changes in volume and diameter were calculated as ( V 1 − Vx ). Therefore, a positive value indicated AAA shrinkage, whereas a negative value indicated volume increase. The analogue formula was used to assess diameter changes. Correlation between volumes and maximal diameters was assessed using the Pearson's correlation coefficient, and a value of r > 0.8 was considered to represent a high correlation. In single-tests, the level of significance was defined at p < 0.05. Group differences were calculated with either chi square test or analysis of variance (ANOVA) and, if at hand, inhomogenity was controlled by using Tamhane or Scheffe as post hoc tests. 3 Results In all the patients procedures were carried out to completion without major complications or conversion to open surgery. No patient died within the in-hospital period, but there were five (6.7%) late deaths, not related to the AAA. Two patients died from malignancy, two from myocardial infarction and one patient from severe stroke. Five patients failed to follow-up. The mean follow-up period in the remaining 68 patients was 17.3 months (range 1.8–42.7 months) including visits at the treating and the local referring hospital. A total of 220 CTA examinations were performed after EVAR. The mean postinterventional volume of aneurysm was 165.63 ml ± 93.29 ml (range 47.94–565.67 ml). The mean maximal postinterventional diameter was 5.91 cm ± 1.52 cm (range 3.72–13.82 cm). Within the entire observational period, for all patients, a slight, non-significant of 1.6% decrease in volumes and a 9.3% in diameters were observed (mean volume decrease 2.58 ml ± 69.05 ml, range 82.82–201.92 ml; mean diameter 0.55 cm ± 1.22 cm, range 2.85–1.93 cm). For all examinations a high correlation of volume and diameter was calculated ( r = 0.813–0.905; α < 0.01) ( Table 1 ). There were no patients with increasing aneurysm volume in whom a decrease of aneurysm diameter was noted. Only in the third and fourth observational interval, there was a low underestimation of aneurysm volume by diameter measurement ( Table 1 ). Endoleaks were detected at CTA in 37 (54.4%) patients. 34 endoleaks (91.9%) were primary and detected in the postinterventional controls, whereas the remaining three cases were secondary endoleaks (8.1%). Three cases of type I leak were depicted at the proximal ( n = 2) or distal ( n = 1) neck. The remaining endoleaks were classified as type II ( n = 34) ( Fig. 3 ). 16 type II endoleaks (47.1%) resolved spontaneously without intervention. Additionally, in twelve patients (17.6%) we observed endotension. In patients without evidence of endoleaks, a mean decrease in volume of 24.2% (mean absolute volume decrease 40.01 ml) and in diameter of 23.5% (mean absolute diameter decrease 1.39 cm) was observed. The presence and type of endoleak was associated with volume increase and, to a lesser extent, with diameter increase. In type I endoleaks, an increase of aneurysm dimension was observed with a mean volume increase of −34.6% (mean absolute increase of volume −107.65 ml) and in mean diameter of −16.4% (mean absolute increase of diameter −0.97 cm), respectively. In case of type II endoleaks, an overall mean volume increase of −12.4% (mean absolute increase of volume −20.58 ml) and a mean increase in diameter of −8.1% (mean absolute increase of diameter −0.48 cm), were observed ( Table 2 ). Thereby, we distinguished between type II endoleaks with (41.2%, n = 14) and without (58.8%, n = 20) progression. Four patients (5.9%) required secondary intervention. In one patient, thrombotic occlusion of an iliac limb was detected, which was treated by transarterial thrombectomy and a graft extension of the concerned limb. Due to persistent type Ia endoleaks, two patients required proximal balloon dilatation to configure the sealing zones and make them more secure. Although both endoleaks were not completely sealed, further aneurysm increase was slight by −4.3 and −10.1%, respectively, during follow up of more than one year. The fourth patient was referred to our emergency room with an acute aortic syndrome caused by a covered aneurysm rupture. He was immediately endovascularly revised and is now well and free from symptoms. 4 Discussion EVAR has become a minimal-invasive option to open surgery repair in patients with infrarenal abdominal aortic aneurysms. With CTA being the imaging modality of choice, diameter and volume measurements of the aneurysm sac have been proposed as alternative methods in follow-up after EVAR [12,13] . The purpose of our retrospective study was to compare diameter and volume measurements of the aneurysm sac after EVAR. Theoretically, volume measurements should have a higher sensitivity for detection of gradual changes of the aneurysm sac size in comparison to diameter measurements. Changes in morphology are known to alter the aneurysm sac dimensions at multiple levels, whereas diameter measurement is the expression of aneurysm dimension in only a single cross-section. In line with this argument, volume measurements have been reported to be superior in early detection of endoleaks where only minor changes are expected [8,14] . However, volume measurement of the aneurysm sac is time consuming and its clinical relevance unclear. In our study, a close correlation between volumes and diameters measurements of the aneurysm sac was found ( Table 1 ). There were no patients with simultaneously increasing aneurysm volumes and decreasing aneurysm diameters, or vice versa, neither in early nor in late endoleaks. Only in the third and fourth measurement intervals, there was a low underestimation by diameter measurement, but without clinical relevance. In our series, 37 endoleaks (54.4%) were detected over the entire follow-up period. There was a difference in trend of aneurysm size in patients with or without endoleaks. Persisting leaks were significantly associated with volume and diameter increase. A mean volume reduction of 24.2% as well as in diameter of 23.5% in patients without evidence of a persisting leak confirms the general statement, that complete aneurysm sac exclusion is essential for predicting therapy success. Moreover, we observed clearly marked differences in volume and diameter modification rates by comparing different types of endoleaks, conforming with Bley et al. [15] . They observed a significant progression in so-called high-flow-endoleaks (type I and III) compared to low-flow-endoleaks (type II). In our study, type I endoleaks showed rapid increase of aneurysm dimension in mean volume of −65.0% and in mean diameter of −16.4%, respectively. In case of type II endoleaks, an overall mean volume increase of −12.4% and a mean increase in diameter of −8.1%, respectively, were observed. In type II endoleaks, we furthermore distinguished cases with (41.2%) and without (58.8%) progression concordant in both volume and diameter measurements. Twelve cases of aneurysm increase with no evidence of endoleaks at CTA were observed and classified as endotension. Due to excessive deviation of data, a statistically quantified volume increase could not be calculated. There is still a controversial debate about its pathogenesis [11,16] . In fact, endotension might be the expression of a non-visualized endoleak. We observed endotension in four patients following spontaneously sealed endoleaks. According to Napoli et al., occult leaks were suspected. Alternative imaging modalities have been proposed as more sensitive tools in detecting endoleaks, such as magnetic resonance angiography and contrast-enhanced US [16] , which has not been used in our study. Due to complications, such as thrombotic occlusion, endoleaks and covered aneurysm rupture, 5.9% of our patients required endovascular re-intervention. None needed second revision. In the entire follow-up period complication was associated with neither procedure or graft materials nor the need for conversion. No patient died within the in-hospital period (perioperative mortality 0.0%) but there were five (6.7%) late deaths, not related to the AAA (aneurysm associated mortality 0.0%). One rupture of AAA occurred in the entire period, a rate of 1.0% per year of follow-up. 4.1 Study limitations The present study is limited by its retrospective design. Nonetheless, volume analysis as well as diameter measurements were performed without knowledge of patient follow-up, outcome or the presence of complications at CTA. Furthermore, only a number of 68 patients were included in the present study. Hence, coincidental influences were neither eliminated nor avoided. Also gender distribution was widely in favour for men. In this study, we only recorded personal data (age, gender) and morphological data from CT. No clinical data was collected. Hence, we were not able to evaluate a possible correlation with co-morbidities. We used the presence or absence of an endoleak as a determining factor to evaluate the significance of aneurysm changes. Probably, other parameters, such as the intrasac pressure measurement, would be more appropriate [17] . At last, the inability to use a contrast medium (because of renal failure or contrast agent allergy) prevents endoleak detection. But contrast-enhanced CT is not necessary for volumetric or diametric measurement. The declared aim of EVAR is to prevent the patient from continuing aneurysm enlargement and rupture. Although other studies have questioned the reliability of maximal orthogonal diameter measurement, proposed more detailed three-dimensional reconstruction methods and volume assessment proved to be more accurate than diameter in the early detection of aneurysm growth [8,14,15,18] . Our findings are substantially in agreement with previous studies [14] . For follow-up of abdominal EVAR using CTA there is a high correlation between volumetric measurements and the maximal orthogonal diameter of AAA after EVAR using cross-sectional imaging (level of significance: a < 0.01). However, the dimension of aneurysms was slightly underestimated within and after second year following EVAR. Despite that small aberrance, measurement of maximal orthogonal diameter seems to be sufficient to exclude post interventional enlargement in a daily clinical routine setting. Time-consuming volumetric measurements should be additionally used in complicated or ambiguous cases. 5 Conclusion Based on the presented data, CTA measurement of the maximal orthogonal diameters seems to be as feasible as time-consuming volumetric measurements to exclude post interventional enlargement for follow-up of abdominal EVAR Conflict of interest None References [1] J. Golledge J. Muller A. Daugherty P. Norman Abdominal aortic aneurysm: pathogenesis and implications for management Arteriosclerosis, Thrombosis and Vascular Biology 26 12 2006 2605 2613 [2] K.A. Vardulaki N.M. Walker N.E. Day S.W. Duffy H.A. Ashton R.A. Scott Quantifying the risks of hypertension, age, sex and smoking in patients with abdominal aortic aneurysm British Journal of Surgery 87 2000 195 200 [3] Y.G. Wolf B.B. Hill G.D. Rubin T.J. Fogarty C.K. Zarins Rate of change in abdominal aortic aneurysm diameter after EVAR Journal of Vascular Surgery 32 2000 108 115 [4] J.C. Parodi J.C. Palmaz H.D. Barone Transfemoral intraluminal graft implantation for abdominal aortic aneurysms Annals of Vascular Surgery 5 1991 491 499 [5] G.R. Upchurch J.L. Eliason J.E. Rectenwald G. Escobar L. Kabbani E. Criado Endovascular abdominal aortic aneurysm repair versus open repair: why and why not? Perspectives in Vascular Surgery and Endovascular Therapy 21 2008 48 53 [6] C. van Marrewijk J. Buth P.L. Harris L. Norgren A. Nevelsteen M.G. Wyatt Significance of endoleaks after endovascular repair of abdominal aortic aneurysms. The EUROSTAR experience Journal of Vascular Surgery 35 2002 461 473 [7] tp. EVAR EVAR vs. open repair in patients with abdominal aortic aneurysm (EVAR trial 1): randomized controlled trial The Lancet 365 2005 2179 2186 [8] I. Bargellini R. Cioni P. Petruzzi A. Pratali V. Napoli Endovascular repair of abdominal aortic aneurysms: analysis of aneurysm volumetric changes at mid-term follow-up Cardio Vascular and Interventional Radiology 28 2005 426 433 [9] J.W. van Keulen J. van Prehn M. Prokop F.L. Moll J.A. van Herwaarden Potential value of aneurysm sac volume measurements in addition to diameter measurements after endovascular aneurysm repair Journal of Endovascular Therapy 16 4 2009 506 513 [10] G.H. White J. May R.C. Waugh X. Chaufour W. Yu Type III and type IV endoleak: toward a complete definition of blood flow in the sac after endoluminal abdominal aortic aneurysms repair Journal of Endovascular Surgery 5 1998 305 309 [11] N. Toya T. Fujita Y. Kanaoka T. Ohki Endotension following endovascular aneurysm repair Vascular Medicine 13 2008 305 311 [12] S.W. Stavropoulus S.R. Charagundla Imaging techniques for detection and management of endoleaks after EVAR Radiology 243 2007 641 655 [13] J.V. Liaw M. Clark R. Gibbs M. Jenkins N. Cheshire M. Hamady Complications and management of infrarenal EVAR European Journal of Radiology 5 2008 1 11 [14] Y.G. Wolf M. Tillich W.A. Lee T.J. Fogarty C.K. Zarins G.D. Rubin Changes in aneurysm volume after endovascular repair of abdominal aortic aneurysm Journal of Vascular Surgery 36 2002 305 309 [15] T.A. Bley P.J. Chase S.B. Reeder C.J. Francois K. Shinki G. Tefera Endovascular abdominal aortic aneurysm repair: nonenhanced volumetric CT for follow-up Radiology 253 2009 253 262 [16] V. Napoli I. Bargellini S.G. Sardella Abdominal aortic aneurysm: contrast enhanced US for missed endoleaks after endoluminal repairs Radiology 233 2004 217 225 [17] J. Hinnen Aneurysm sac pressure after EVAR: The role of endoleak European Journal of Vascular and Endovascular Surgery 34 2007 432 441 [18] J.T. Lee I.N. Aziz J.T. Lee Volume regression of abdominal aortic aneurysms and its relation to successful endoluminal exclusion Journal of Vascular Surgery 38 2003 1254 1263