Incidence and Evolution of Mural Thrombus in Abdominal Aortic Endografts

Article Outline

• Abstract
• Introduction
• Materials and Methods
  • Preoperative Data and Endograft Characteristics
  • Postoperative Follow-up
  • Statistics
• Results
  • Preoperative Study
  • Endograft Characteristics
  • Postoperative Follow-up
• Discussion
• Conclusions
• References
• Copyright

Background

The aims of this study were to analyze the predictive factors for intragraft mural thrombus formation and evolution during follow-up after endovascular treatment of abdominal aortic aneurysms and to evaluate its relationship with the subsequent appearance of complications.

Methods

A retrospective study was performed by selecting those patients who underwent endovascular repair of an abdominal aortoiliac aneurysm between June 1998 and September 2004, with a minimum follow-up of 24 months. Preoperative clinical data, anatomical characteristics of the aneurysm, and endograft type were analyzed. In addition, clinical evaluation and abdominal computed tomography angiography (CT scans) performed at 1, 6, 12, and 24 months after the surgery were reviewed.

Results

Eighty-nine patients were submitted for endovascular aneurysm repair in this period, and 75 completed the 24-month follow-up. Eighteen patients developed intragraft mural thrombus (24% incidence), 13 (72.2%) appearing at 1 month of follow-up, and up to 16 (88.9%) appearing during the first 6 months. Logistic regression analysis showed that the lumen percentage of mural thrombus in the native aorta and the use of aortouniiliac endografts were independent predictors of intragraft mural thrombus formation (odds ratio, 1.065; 95% confidence interval, 1.022-1.110, and odds ratio, 8.014; 95% confidence interval, 1.598-40.181, respectively). No spontaneous regression of the thrombus was observed. The area of intragraft mural thrombus had increased at 12 and 18-24 months after their diagnosis (Wilcoxon signed rank test, p = 0.028 and 0.028, respectively). The presence of intragraft mural thrombus was associated with a greater tendency to endograft body or branch occlusion (5 of 18 cases with intragraft mural thrombus (27.8%) versus 1 of 57 cases without it (1.8%), (p = 0.003).

Conclusion

Intragraft formation of mural thrombus is a common finding during the follow-up of abdominal aortic endografts, particularly in aneurysms with large mural thrombus of the native aorta, and is associated with the use of aortouniiliac endografts. The area occupied by the mural thrombus was shown to gradually increase during follow-up and was associated with a greater tendency for endograft occlusion.

Back to Article Outline

Introduction 

Since the report by Parodi et al.¹ in 1991, endografts have been widely used for endovascular aneurysm repair (EVAR) of the abdominal aorta.² However, these patients still need to be closely followed due to complications of the EVAR procedure itself, such as leaks, migrations, twisting, or thrombosis of the endograft.³ Another common complication after EVAR is the formation of endograft mural thrombosis. It has been described using different terms: circular or semicircular thrombus,⁴ intraluminal thrombus,⁵ mural⁶, ⁷ or parietal⁸ thrombus, and thrombus apposition.⁹ Incidence rates from 3% to 23% after successful EVAR have been reported.⁴, ⁶, ⁹, ¹⁰ Thrombus formation has been identified between 1 week and 20 months after the procedure and its further course seems to be highly variable.⁴, ⁹ Although it has been hypothesized that there is an association between intragraft mural thrombosis and the appearance of stenosis or occlusion of the endograft, this has not been confirmed.⁹ The predictive factors or the long-term significance of this complication are also unknown. The aims of this study were to analyze the predictive factors involved in the formation and evolution of intragraft mural thrombus during follow-up after EVAR and to evaluate its relationship with the appearance of subsequent complications.

Back to Article Outline

Materials and Methods 

A retrospective study was carried out, selecting those patients who underwent EVAR for abdominal aortic or iliac aneurysm between June 1998 and September 2004. To guarantee a sufficient follow-up period, patients with less than 24-month follow-up were excluded. Immediate conversions and device thromboses, as well as reinterventions, were also excluded.

The following variables were analyzed (Table I).

Table I. Variables analyzed in the study

Clinical and demographic characteristics according to the standards of the Society for Vascular Surgery/International Society for Cardiovascular Surgery (SVS/ISCVS).

Anatomy of the aneurysm according to the parameters described by the EUROSTAR study group.¹⁴, ¹⁵

Preoperative Data and Endograft Characteristics 

Clinical and demographic variables were evaluated according to the criteria of the Society for Vascular Surgery/International Society for Cardiovascular Surgery (SVS/ISCVS), as well as the length and diameters of the aneurysm according to the parameters described by the EUROSTAR study group.¹⁴, ¹⁵ The presence of thrombus or angulation in the neck of the endograft and the percentage of the native aortic lumen occupied by mural thrombus were also recorded. The indication for the intervention was the presence of an aneurysm in the aorta in 59 cases (66.3%), aortoiliac in 15 (16.9%), and iliac in 15 (16.9%).

After the procedure, the characteristics of the commercial device, as well as the actual configuration of the endograft (bifurcated or aortouniiliac), endograft fabric, oversizing of the proximal neck, location of the aortic proximal fixation (above or below the renal arteries),¹⁶ and location of distal landing zone (common or external iliac) were specifically recorded in all the patients.

Postoperative Follow-up 

A standardized follow-up was performed. Clinical evaluation, as well as abdominal computed tomography angiography (CT scans) and plain radiogaphs at different angles, were obtained at 1, 6, 12, and 24 months after the procedure. Intragraft mural thrombus was defined as the area of endograft lumen occupied by thrombus at the arterial phase of CT sections¹¹ (Fig. 1). Location and measurement of percentage thrombus area were specifically registered in a database. For this purpose, CT images were digitalized and analyzed with an image processing software (Photoshop 6.0 for Windows). Overall intragraft area and the intragraft area occupied by the mural thrombus (number of pixels) were measured. Thrombus area was expressed as a percentage of total lumen intragraft area.

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 1 

  (A) Preoperative computed tomography. (B) An intragraft mural thrombus is observed at 6-month follow-up. (C, D) The thrombus area is increased at 12- and 24-month reviews.

Medical treatment (simple antiplatelet therapy, double regimen, or systemic anticoagulation) and location of graft thrombosis (endograft main body or branches) were also registered during follow-up.

Statistics 

Descriptive statistics and frequencies were obtained and comparisons were made using the SPSS, Version 15.0 statistical package. The statistical significant difference between groups were evaluated using the Pearson χ² test or the Fisher exact test for the categorical variables and the Student t test and analysis of variance (ANOVA) for continuous variables. A stepwise multivariate logistic regression analysis was performed to identify the independent predictors of intragraft mural thrombosis. The Wilcoxon test was used to analyze the differences in the changes in the mural thrombus during follow-up. A value of p< 0.05 was considered to be statistically significant.

Back to Article Outline

Results 

Eighty-nine patients who underwent EVAR of aortoiliac aneurysm between June 1998 and September 2004 fulfilled the inclusion criteria. Nine patients were lost for follow-up and 5 patients died during this period. Only one death was associated with the procedure, due to mesenteric ischemia and multiorgan failure after infection of a femorofemoral bypass. The rest of the sample (75 patients) completed the 24-month follow-up and their results are described next.

The mean age was 73.5 years, and 71 (95%) were male. Different endograft configurations were used (13 [17.3%] aortouniiliacs and 62 [82.7%] bifurcated endografts), most of them of polyester fabric (40 polyester endografts [53.3%] versus 35 ePTFE endografts [46.7%]). The mean oversizing was 15.57% (95% confidence interval [CI], 13.98%-17.18%).

The commercial models of the endografts used were the Gore Excluder AAA Endoprosthesis (W.L. Gore & Associates, Flagstaff, AZ) in 34 (45.3%) patients, Medtronic Talent Abdominal Stent Graft (Medtronic AVE, Santa Rosa, CA) in 17 (22.7%), Vanguard Stent Graft (Boston Scientific corp., Natick, MA) in 13 (17.3%), AneuRx Stent Graft System (Medtronic AVE, Santa Rosa, CA) in 6 (8.0%), Cook Zenith AAA Endovascular Graft (Cook, Bloomington, IN) in 4 (5.3%), and Endologix's Powerlink System (Endologix Inc, Irvine, CA) in 1 (1.3%).

After analyzing all the postoperative CT scans, some degree of intragraft mural thrombus was observed in 18 cases (incidence at 24 months of follow-up, 24%). There were no cases of spontaneous regression of mural thrombus during the follow-up period. After initial analysis of the sample, data were compared regarding the presence or absence of intragraft mural thrombus during follow-up considered as dependent variable.

Preoperative Study 

No significant differences were observed in the distribution, clinical, and demographic characteristics between both groups (presence or absence of intragraft thrombus). Regarding the anatomic characteristics of the aneurysm, higher values were observed in the D2a diameter (proximal segment of infrarenal aortic neck: 23.09mm versus 24.78mm, p = 0.021) and lumen percentage of mural thrombus in the native aorta (31.78% versus 50.54%), in the group who later developed an intragraft mural thrombus. The rest of the results are given in Table II.

Table II. Comparison of the aneurysm anatomical characteristics considered in the baseline preoperative CT between patients with and without a mural thrombus in the follow-up

Values are mean, 95% confidence interval in parentheses, and statistical significance.

Measurements used according to EUROSTAR¹⁵ classification. Diameters: D1, suprarenal aorta; D2a, infrarenal aorta (proximal segment of the infrarenal neck); D3, maximum diameter of the aneurysm; D3a, patent aortic lumen in the same CT section as D3, D4, end aorta; D5 mean, mean of both common iliac diameters.

H1, distance of proximal neck.

Lumen percentage of mural thrombus: [(D3 – D3a)/D3].

Endograft Characteristics 

The aortouniiliac endografts had a greater tendency to form mural thrombus than the bifurcated ones (7 of 13 aortouniiliac endografts [53.8%] versus 11 of 62 bifurcated ones [17.7%], p = 0.011; Table III), as well as the use of polyester fabric instead of ePTFE (tendencey without significant differences: 5 of 35 ePTFE endografts [14.3%] versus 13 of 40 polyester devices [32.5%], p = 0.065). No significant differences were seen regarding the patency of the internal iliac arteries or the distal landing of the endograft (common or external iliac arteries). In addition, no statistically significant relationships were found between the appearance of intragraft mural thrombus and the rest of variables.

Table III. Endograft characteristics

Endograft characteristics	Type		Intragraft mural thrombus				P
			No (n = 57)		Yes (n = 18)
Configuration	Aortouniiliac		6	46.2%	7	53.8%	0.011
	Bifurcated		51	82.3%	11	17.7%
Proximal fixation	Suprarenal		8	72.7%	3	27.3%	0.783
	Juxtaposed or infrarenal		49	76.6%	15	23.4%
Distal landing	Aortouniiliac	CI	3	60.0%	2	40.0%	0.592
		EI	3	37.5%	5	62.5%
	Bifurcated	BCI	27	84.4%	5	15.6%	0.242
		CI+EI	9	81.8%	2	18.2%
		BEI	15	78.9%	4	21.1%
Fabric material	ePTFE		30	85.7%	5	14.3%	0.065
	Polyester		27	67.5%	13	32.5%
Stent material	Cobalt-chrome alloy		1	100%	0	0.0%	0.397
	Nitinol		54	77.1%	16	22.9%
	Stainless steel		2	50.0%	2	50.0%
Oversizing			15.75% (13.92–17.59)		15.02% (11.42–18.62)		0.698

Values are number, percentage, and 95% confidence interval for the oversizing and statistical significance.

The distal seal of the endografts is defined according to the configuration used: aortouniiliac endograft (CI, common Iliac artery; EI, external iliac artery) and bifurcated (BCI, both common iliacs; CI+EI, common Iliac on one side and external iliac on the contralateral; BEI, both external iliac).

Postoperative Follow-up 

Intragraft thrombus formation was detected in 18 cases. They appeared early: 13 cases (72.2%) in the first month of follow-up, and up to 16 (88.9%) in the first 6 months. The most frequent location was the main body of the endograft (16 cases), usually in its middle third. The clot extended into one of the branches in the remaining 2 patients.

At the time they were detected, the mean area occupied by the mural thrombus was 16.90% of the inner area of the endograft (95% CI, 11.67%-22.13%). No significant variation was observed in the mean thrombus area at 6 months, however, a dramatic increase was observed in the CT evaluations performed at 12 and 18-24 months (19.79%, 33.75% and 33.79%, respectively; Wilcoxon signed rank test, p = 0.722, 0.028, and 0.028, respectively) (Fig. 2).

• 
  • View Large Image
  • Download to PowerPoint

• Fig 2. 

  Mean percentage of the luminal area of the endograft occupied by the mural thrombus, according to the months of follow-up after the appearance of the mural thrombus.

No significant difference was observed regarding the drug therapy during follow-up (single or double antiplatelet therapy) and its relationship with the appearance of an intragraft mural thrombus (p = 0.438).

Complete thrombosis of the main body of the graft, or its branches, was observed in 6 patients during the 24-month follow-up period. Five events occurred in patients with intragraft mural thrombus previously detected in intermediate CT evaluations during the follow-up: two aortouniiliac graft occlusions and three branch occlusions of bifurcated grafts (in one of these latter patients, a mural thrombus was present in the main body and extended to the further occluded branch; clot limited to the main body of the graft was identified in the other two patients). No structural defects, such as stenosis or twisting of the branches, were identified in the endografts of these patients.

Among patients without previously detected intragraft mural thrombus, one branch occlusion occurred during follow-up. The comparison between both groups showed that there was a greater tendency to graft occlusion in the group with a previously detected intragraft mural thrombus (5 of 18 [27.8%] versus 1 of 57 [1.8%], p = 0.003).

Revascularization of the lower limbs, by means of an axillofemoral bypass, was performed in two cases of aortouniiliac graft thrombosis. A cross-over femorofemoral bypass was performed in two of the four cases of thrombosed branch of a bifurcated endograft, and medical treatment was given in the other two patients, due to long distance claudication as verified at the treadmill test.

A logistic regression was performed using preoperative data, characteristics of the endografts, and the outpatient treatment as independent variables and the presence or absence of intragraft mural thrombus as a dependent variable (Table IV). Both the lumen percentage of mural thrombus in the native aorta (odds ratio [OR], 1.065; 95% CI, 1.022-1.110, for each 1% increase in the lumen percentage taken up by the mural thrombus), and the aortouniiliac graft configuration (OR, 8.014; 95% CI, 1.598-40.181) showed to be independent predictors of intragraft mural thrombosis in the follow-up.

Table IV. Study of intragraft mural thrombus predictive factors, using preoperative variables, endograft characteristics and outpatient treatment

Logistic regression; OR, odds ratio; 95% CI, 95% confidence interval with upper and lower limits; p, statistical significance.

Back to Article Outline

Discussion 

In the present series, formation of intragraft mural thrombus was observed in 18 cases at 24-month follow-up (24% incidence). This is a slightly higher rate than that registered in other previously published series by Dorffner et al. (18.5%),⁴ Krauss et al. (20.4%),¹² Schunn et al. (10.8%),⁵ and Wegener et al. (23.2%).⁹ Both the lumen percentage occupied by mural thrombus in the native aorta and the configuration of the graft (aortouniiliac endografts) have been shown to be independent predictors of intragraft formation of mural thrombus during the follow-up. Both factors probably lead to significant variations in the area of the patent graft lumen. Aortouniiliac endograft configuration leads to a sharp change from a large proximal aortic diameter to a smaller distal iliac one. Likewise, an extensive preoperative mural thrombus (associated with a narrower aortic lumen) is likely to make it difficult the complete expansion of the main body of the endograft.

In both situations, abrupt change in the caliber of the conduit may cause turbulent flow, subsequently increasing the risk of endograft mural thrombus formation. Computer simulation of blood flow, based on CT data, would probably give more information on the hemodynamic changes within the endograft and could confirm this hypothesis. Despite these considerations, we have not observed any differences regarding either the location of distal landing of the endograft or the patency of the internal iliac arteries, both factors being known to determine the actual outflow of the endograft.

In this study, ePTFE endografts showed a tendency (without significant differences) to generate mural thrombus apposition than those made of polyester fabrics (5 of 35 ePTFE endografts [14.3%] versus 13 of 40 polyester devices [32.5%], p = 0.065). Preliminary results in our thoracic endograft series suggest a similar behavior of ePTFE grafts in the thoracic aorta. Longer follow-up, controlling other confounding factors, might aid to clarify the relative resistance to local thrombosis of endografts made of different fabrics.

In agreement with Wegener et al.⁹ and Dorffner et al.,⁴ intragraft mural thrombus was mainly located in the main body of the endograft, and it appeared early. Similarly to that observed in these series (maximum clot detection between the first week and 20 months after the intervention), intragraft mural thrombosis was detected in the first 6 months in up to 88.9% of the cases.

Previous studies have referred an inconsistent evolution of the area of the mural thrombus: Dorffner et al.⁴ described an initial regression in the area occupied by the mural thrombus in two of five cases, with a later progression at 14 months in one of them. Wegener et al.⁹ described an increase of the thrombus area in 41% of cases during follow-up but a complete resolution in 18% of them.⁹ Certainly, progression of intragraft thrombus area did not increase in all the patients in the present series. However, overall a significant increase in the mean area occupied by the intragraft thrombus was identified at 12- and 18- to 24-month follow-up after the diagnosis of the intragraft mural thrombus. We did not observe any case of spontaneous regression of the thrombus.

Although the actual causes for progression of the intragraft mural thrombus are not completely understood, this situation was clearly associated with a greater tendency to lead to an occlusion of the main body or branches at 24-month follow-up (5 of 18 cases with intragraft mural thrombus [27.8%] versus 1 of 57 cases without it [1.8%], p = 0.003). Despite not being able to identify other structural defects (stenosis or branch twisting) that could justify occlusion of the graft, we do not know if mural thrombus actually increases thrombogenicity of the graft or simply reflects an initial stage of clot formation. Otherwise, the results in the present series contrast with those referred in other previously published series, showing much lower rates of endograft occlusion in patients with intragraft mural thrombus (0%-5.8%).⁹, ¹²

Although some groups have advised the use of anticoagulant therapy to prevent this complication after the appearance of an intragraft mural thrombus,⁴ we did not realize a lower frequency of intragraft thrombus in those patients previously anticoagulated. At present, available data are insufficient to discern whether anticoagulation decreases the graft thrombosis rate in these patients. Our strategy to prevent this complication is based on a strict follow-up, starting anticoagulation only in those cases with a rapid increase in the area occupied by the mural thrombus.

The follow-up at 2 years in this study has been completed in a reasonably high number of patients (75 of 89 [82%]), compared with higher withdrawal rates in other studies such as the EVAR Trial.¹³ However, the main limitation of the present study has been the reduced size of the sample. Longer-term studies on a greater number of patients could provide more information on the consequences and evolution of the intragraft mural thrombus.

Back to Article Outline

Conclusions 

Intragraft formation of mural thrombus is a common finding (24.0%) and appears early after EVAR. It is more frequent in aneurysms that have large mural thrombus in the native aorta and is associated with the use of aortouniiliac endografts. The area occupied by the intragraft mural thrombus is shown to increase during follow-up and has been associated with a greater tendency toward occlusion of the graft.

Back to Article Outline

References 

1. Parodi JC, Palmaz JC, Barone HD. Transfemoral intraluminal graft implementation for abdominal aortic aneurysms. Ann. Vasc. Surg. 1991;5:491–499
   • View In Article
   • Abstract
   • Full-Text PDF (2990 KB)
   • CrossRef
2. Franks SC, Sutton AJ, Bown MJ, et al. Systematic review and meta-analysis of 12 years of endovascular abdominal aortic aneurysm repair. Eur. J. Vasc. Endovasc. Surg. 2007;33:154–171
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (453 KB)
   • CrossRef
3. Hobo R, Buth J. EUROSTAR Collaborators. Secondary interventions following endovascular abdominal aortic aneurysm repair using current endografts. A EUROSTAR report. J. Vasc.Surg. 2006;43:896–902
   • View In Article
4. Dorffner R, Thurnher S, Polterauer P, et al. Treatment of abdominal aortic aneurysms with transfemoral placement of stent-grafts: complications and secondary radiologic intervention. Radiology. 1997;204:79–86
   • View In Article
   • MEDLINE
5. Schunn CD, Krauss M, Heilberger P, et al. Aortic aneurysm size and graft behavior after endovascular stent-grafting: clinical experiences and observations over 3 years. J. Endovasc. Ther. 2000;7:167–176
   • View In Article
   • MEDLINE
   • CrossRef
6. Sakai T, Dake MD, Semba CP, et al. Descending thoracic aortic aneurysm: thoracic CT findings after endovascular stent-graft placement. Radiology. 1999;212:169–174
   • View In Article
   • MEDLINE
7. Heijmen RH, Schepens MA. Mural thrombosis in thoracic aortic stent-graft. Eur. J. Cardiothorac. Surg. 2002;22:628
   • View In Article
   • Full Text
   • Full-Text PDF (92 KB)
   • CrossRef
8. Tillich M, Hausegger KA, Tiesenhausen K, et al. Helical CT angiography of stent-grafts in abdominal aortic aneurysms: morphologic changes and complications. Radiographics. 1999;19:1573–1583
   • View In Article
   • MEDLINE
9. Wegener M, Görich J, Krämer S, et al. Thrombus formation in aortic endografts. J. Endovasc. Ther. 2001;8:372–379
   • View In Article
   • MEDLINE
   • CrossRef
10. Dorffner R, Thurnher S, Youssefzadeh S, et al. Spiral CT angiography in the assessment of abdominal aortic aneurysms after stent grafting: value of maximum intensity projections. J. Comput. Assist. Tomogr. 1997;21:472–477
    • View In Article
    • MEDLINE
    • CrossRef
11. Mita T, Arita T, Matsunaga N, et al. Complications of endovascular repair for thoracic and abdominal aortic aneurysm: an imaging spectrum. Radiographics. 2000;20:1263–1278
    • View In Article
    • MEDLINE
12. Krauss M, Ritter W, Bär I, et al. Imaging of aortic endoprostheses and their complications. Rofo. 1998;169:388–396
    • View In Article
    • MEDLINE
    • CrossRef
13. The EVAR. Trial Participants. Endovascular aneurysm repair versus open repair in patients with abdominal aortic aneurysm (EVAR Trial 1): randomised controlled trial. Lancet. 2005;365:2179–2186
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (109 KB)
    • CrossRef
14. Harris PL, Buth J, Mialhe C, et al. The need for clinical trials of endovascular abdominal aortic aneurysm stent-graft repair: the EUROSTAR Project. EUROpean collaborators on Stent-graft Techniques for abdominal aortic Aneurysm Repair. J. Endovasc. Surg. 1997;4:72–77
    • View In Article
    • MEDLINE
    • CrossRef
15. Mohan IV, Harris PL, Van Marrewijk CJ, et al. Factors and forces influencing stent-graft migration after endovascular aortic aneurysm repair. J. Endovasc. Ther. 2002;9:748–755
    • View In Article
    • MEDLINE
    • CrossRef
16. Chaikof EL, Blankensteijn JD, Harris PL, et al. Ad Hoc Committee for Standardized Reporting Practices in Vascular Surgery of The Society for Vascular Surgery/American Association for Vascular Surgery. Reporting standards for endovascular aortic aneurysm repair. J. Vasc. Surg. 2002;35:1048–1060
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (102 KB)
    • CrossRef