Intra-aneurysmal pressure after incomplete endovascular exclusion

CLINICAL RESEARCH STUDIES

Intra-aneurysmal pressure after incomplete endovascular exclusion Juan Carlos Parodi, MD,a Ramón Berguer, MD, PhD,b Luis Mariano Ferreira, MD,a Ricardo La Mura, MD,a and Mark L. Schermerhorn, MD,c Buenos Aires, Argentina Purpose: An endoleak results from the incomplete endovascular exclusion of an aneurysm. We developed an experimental model to analyze hemodynamic changes within the aneurysm sac in the presence of an endoleak, with and without a simulated open collateral branch. Methods: With a latex aneurysm model connected to a pulsatile pump, pressures were measured simultaneously within the system (systemic pressure) and the aneurysm sac (intrasac pressure). The experiments were performed without endoleak (control group) and after creating a 3.5-mm (group 1), 4.5-mm (group 2), and 6-mm (group 3) diameter orifice in the endograft, simulating an endoleak. Pressures were also registered with and without a patent aneurysm side branch. Results: In each endoleak group, the intrasac diastolic pressure (DP) and mean pressure (MP) were significantly higher than the systemic DP and MP (P = .01, P = .006, and P = .001, respectively), although the pressure curve was damped. The presence of an open side branch significantly reduced the intrasac DP and MP. Conclusion: In this model, intrasac pressures were significantly higher than systemic pressures in the presence of all endoleaks, even the smallest ones. Intrasac pressures higher than systemic pressure may pose a high risk for aneurysm rupture. Although patent side branches significantly reduce these pressures, the aggressive management of an endoleak should be pursued. (J Vasc Surg 2001;33:909-14.)

Our definition of initial technical success with endovascular techniques in the management of aneurysms includes complete exclusion of the sac, reduction of the intraaneurysm pressure, restoration of normal blood flow, and, ultimately, prevention of rupture. The fixation points at the ends of a stented-graft must be in complete apposition with the normal vessel wall without thrombus interposition to achieve these goals. If this is not achieved, sealing at the fixation points will be incomplete or temporary. The persistence of blood flow outside the graft into the sac is defined as an endoleak. An endoleak involves the failure of complete exclusion of the aneurysm with the persistence of elevated pressures within the aneurysm sac. This, at least in theory, may result in a failure to reduce the risk of aneurysm rupture, which was suggested by several series in which incompletely excluded aneurysms continued to expand.1-4

From the Service of Vascular Surgery, Instituto Cardiovascular de Buenos Airesa; Service of Vascular Surgery, Harper University Hospital, Wayne State University; b and the Section of Vascular Surgery, Dartmouth Hitchcock Medical Center.c Competition of interest: nil. Reprint requests: J.C. Parodi, MD, Instituto Cardiovascular de Buenos Aires, Blanco Encalada 1543, Buenos Aires, 1428, Argentina. Copyright © 2001 by The Society for Vascular Surgery and The American Association for Vascular Surgery. 0741-5214/2001/$35.00 + 0 24/1/119038 doi:10.1067/mva.2001.119038

The hemodynamics of endoleaks are not completely understood. We hypothesized that an endoleak will lead to a situation of significantly increased intra-aneurysm pressure because of the lack of appropriate outflow from the aneurysm sac, which could significantly increase the risk of rupture unless adequately treated. The aim of this study was to assess the hemodynamic changes associated with complete and incomplete exclusion of an aneurysm by using an ex vivo model of aneurysm. METHODS Description of the model. The model was constructed as an in vitro circulatory system. The model consisted of three components: (1) a pulsatile pump, (2) the aneurysm model connected via a silastic polymeric silicone tubing system, and (3) a collecting system (Fig 1). A section of rubber tubing was used between connections to make the system more compliant and to adjust the peripheral resistance comparable with a physiologic value. The aneurysm was made of latex tubing, simulating a fusiform aneurysm with a maximum transverse diameter of 6 cm. Two small side ports came out from the widest middle segment of the sac, one simulating a collateral branch and the other to be used for pressure measurements. The collateral artery (2 mm diameter) had a rubber cuff at the end to modify its resistance and drained into an open reservoir. The perfusion fluid was a solution of glycerol with a viscosity similar to plasma. The aneurysm was excluded with an 909

JOURNAL OF VASCULAR SURGERY November 2001

910 Parodi et al

Fig 1. Schematic of ex vivo circulation. P, Pneumatic pump; A, aneurysm model; CB, collateral branch; WP, working port; C, blood pressure cuff; OR, open reservoir.

expanded polytetrafluoroethylene tube graft (Gore-Tex, Flagstaff, Ariz) that was precoagulated with human blood to eliminate leakage through the interstices of prosthesis. Both ends were glued with cyanocrylate and fixed with sealed straps at the proximal and distal ends of the aneurysm. The systemic and intrasac pressures were measured simultaneously. For flow measurements, an extracorporeal electromagnetic flow transducer was used (Biomedicus Transducer model TX40, Medtronic, Minneapolis, Minn). Baseline flow dynamics. After exclusion of the aneurysm sac, intrasac and systemic systolic pressure (SP), diastolic pressure (DP), and mean pressure (MP) baseline measurements were made. Changes were then induced in the pump output by modifying the rate, the system pressure, the proportion of time between systole and diastole, and the peripheral resistance. We varied systemic pressure from 80/30 mm Hg to 215/110 mm Hg by changing the resistance with a rubber cuff applied around the latex tubing and the systolic driving pressure. To achieve this, we changed the stroke volume and peripheral resistance to between 1.90 and 3.9 L/min and 1421 and 2884 dyne/s/cm–5, respectively. These measurements were performed with and without an open aneurysm side branch. Endoleak flow dynamics. The experiments were performed in the absence of endoleak (control) and in the presence of small, medium, and large endoleaks. To simulate an endoleak, a hole was created in the expanded polytetrafluoroethylene graft with an aortic-vein punch (Scanlan, St Paul, Minn) of different sizes. Endoleak hole diameters tested were 3.5 mm (group 1), 4.5 mm (group 2), and 6 mm (group 3). Systemic and intrasac pressures were recorded after this protocol for each group. Pressure measurements were again made with and without an open side branch. We used only one model, in which every observation was carried out. Each of the pressure values given in Tables

I through IV is an average of 30 measurements after hemodynamic, endoleak size, and sac outflow changes were established. Statistical analysis. Continuous variables are expressed as the means ± SD. The Wilcoxon rank sum test was used as a means of analyzing the data derived from pressure measurements made with and without endoleak. A P value less than .05 was considered to indicate statistical significance. The experiment was conceived, planned, and evaluated by the senior author (J.C.P.) and was performed in the Wayne State University experimental laboratory under the direction of Dr Berguer, who designed the model according to the initial plan. Drs Ferreira and La Mura performed the experiments and collected the data, and Dr Schermerhorn independently analyzed this data and made statistical analyses. All five authors reviewed the results and discussed conclusions. METHODS Baseline flow dynamics. Data on systemic and intrasac pressure before making the endoleak is presented in Table I. In the control group (absence of endoleak), the systemic SP remained higher (191.9 ± 26.4 mm Hg) than the intrasac SP (44.1 ± 5.8 mm Hg). The intrasac DP in the control group was lower than the systemic DP (35.4 ± 2.7 vs 66.4 ± 18.1 mm Hg, P = .001). The intrasac pressure curve was damped compared with the systemic pressure curve. Pressure measurements after introduction of an endoleak. After the creation of a 3.5-mm endoleak, SP remained higher in the systemic compartment (116.4 ± 30.8 mm Hg) than within the aneurysm sac (91.4 ± 19.1 mm Hg, P = .001). DP however, increased significantly within the aneurysm sac after the creation of an endoleak (86.1 ± 19.1 mm Hg) and became greater than the systemic DP (54.1 ± 18.5, P < .001). Similarly, the MP after endoleak increased significantly within the aneurysm sac (87.9 ± 19.3 mm Hg) and again became greater than the systemic MP (74.2 ± 21.3 mm Hg). Similar effects on all pressure measurements were noted with larger endoleaks. Intrasac MP increased significantly as the size of the endoleak increased (Table II). The waveform of intrasac pressure was damped compared with the systemic pressure curve, with higher DP but lower SP (Fig 2). The use of lower systemic pressures during this part of the experiment was necessary because of a significant increase in the sac pressure after the endoleak was created, which otherwise would cause disengagement of connections and even an abrupt increase in the sac volume that could rupture the device. Pressure after patent collateral branch. In the presence of a 3.5-mm endoleak, with a 350 mL/min collateral flow, the systemic SP (112.4 ± 13.1 mm Hg) was higher than the intrasac SP (62.7 ± 8.1 mm Hg). However, DP remained significantly increased within the aneurysm sac (40.3 ± 6.1 mm Hg vs 60.91 ± 9.2 mm Hg, P = .001).

JOURNAL OF VASCULAR SURGERY Volume 34, Number 5

Parodi et al

911

Fig 2. Curve of systemic and intrasac pressure in the presence of an endoleak that is 3.5 mm in diameter.

Table I. Pressure in the aneurysm sac (mm Hg) in the absence of endoleak Location Systemic circulation Aneurysm sac P value

Systolic pressure

Diastolic pressure

Mean pressure

191.9 ± 26.4 44.1 ± 5.8 .001

66.4 ± 18.1 35.4 ± 2.7 .001

108.4 ± 20.6 38.3 ± 3.7 .001

Table II. Mean intrasac pressure (mm Hg) in the presence of endoleaks Location

Mean pressure

Systemic circulation Aneurysm sac P value

108.4 ± 20.6 38.3 ± 3.7 .001

3.5-mm diameter endoleak

4.5-mm diameter endoleak

6-mm diameter endoleak

78.8 ± 14.8 100.8 ± 14.4 .006

80.9 ± 11.3 111.4 ± 17.8 .001

74.2 ± 21.3 87.9 ± 19.3 .001

Table III. Pressures (mm Hg) in the aneurysm sac in the presence of a 3.5-mm diameter endoleak and patent collateral branch Location

Type

Systemic

Systolic Diastolic Mean Systolic Diastolic Mean

Aneurysm sac

350 mL/min 112.4 40.3 64.1 62.7 60.9 62.2

± ± ± ± ± ±

13.1 6.1 12.9 8.1 9.2 7.8

The MP was not significantly different (64.1 ± 12.9 mm Hg vs 62.2 ± 7.8 mm Hg, P = .62). As the collateral flow increased, the intrasac DP and MP decreased. In the presence of 650 mL/min of collateral flow, the MP in the aneurysm sac (41.3 ± 6.5 mm Hg) became significantly lower than the systemic pressure (61.3 ± 10.5 mm Hg, P = .001). A similar effect on pressure was also noted after the initiation of higher collateral flow (900 mL/min) (Table III) and in the presence of larger endoleaks (4.5 mm and 6 mm) (Table IV). DISCUSSION Incomplete exclusion of an aneurysm with an endovascular graft may result in persistent flow within the

650 mL/min 98.8 46.0 61.3 42.2 41.2 41.3

± ± ± ± ± ±

11.2 5.2 10.5 6.4 8.2 6.5

900 mL/min 126.0 52.3 77.2 43.3 40.8 41.3

± ± ± ± ± ±

21.2 5.5 8.2 4.9 6.6 5.3

aneurysm sac (endoleak). This flow may be caused by an incomplete seal at the graft ends or, between segments, by thrombus interposition, by incomplete deployment, or by inappropriate sizing. There may be flow through the graft material itself via interstices, tears, or perforations. Perigraft flow may develop in time on the basis of arterial dilatation or aneurysmal remodeling after shrinkage. Finally, endoleaks that are not graft-related may be seen with retrograde flow from patent lumbar or inferior mesenteric arteries. The presence of endoleaks without enlargement and, conversely, enlargement without demonstrable endoleaks allows us to justify our concept of aneurysm sac pressurization as a real cause of enlargement, ultimately leading to aneurysm rupture. Aneurysm expan-

JOURNAL OF VASCULAR SURGERY November 2001

912 Parodi et al

Table IV. Systemic/intrasac mean pressures (mm Hg) in the presence of different sizes of endoleaks and collateral flow rates Endoleak diameter Collateral flow

3.5 mm

0 350 mL/min 650 mL/min 900 mL/min

74.2/87.9 64.1/62.2 61.3/41.3 77.2/41.3

sion has been reported after technically successful exclusion that is associated with inadequate reduction of intraaneurysmal pressure.5,6 However, not all endoleaks produce aneurysm enlargement.7,8 Therefore, the outcome of patients with endoleaks is not yet completely defined. Some authors have reported delayed rupture in patients “waiting” for endoleak repair,9-11 whereas sac shrinkage in the presence of type I and type II endoleaks has also been described.12 Nevertheless, much of this conflicting information, with the experience of the nonresected treatment of AAAs,13 suggests that endoleaks can lead to aneurysm rupture. As is shown in our study, Chuter et al,14 Criado et al,15 Faries et al,16 and Sanchez et al17 showed a significant aneurysm-to-aorta pressure decrease after endoluminal exclusion in animal models. However, no endoleaks or collateral sac branches were created in these models. Schurink et al18 created an ex vivo model in which intrasac pressure after exclusion was 0 mm Hg. They also demonstrated that every small endoleak was associated with an intrasac DP that was identical to the systemic DP. However, in a clinical observation, Treharne et al19 and Chuter et al20 confirmed a decrease in aneurysm pressure immediately after stentgraft implantation and a damped pulsatile pressure curve after stent-graft placement. In all their cases, the endovascular treatment produced a significant reduction not only in the SP, but also in the DP and MP. In all cases in which the sac pressure failed to fall, an endoleak was demonstrated.15,19 In the study by Malina et al,21 aneurysms with endoleaks showed no diameter decrease, and the postoperative pulsatile wall motion of the aneurysm was 50% higher than that in the excluded cases (Table V). We developed an experimental model that enabled us to register pressures after aneurysm exclusion with either complete sealing or in the presence of an endoleak, with and without lumbar-like collateral outflow. We are aware of the hemodynamic differences among aneurysms in different locations. We merely attempted to develop a simple model that allows the study of basic principles. Thus, we decided to construct a generic model that would apply to any aneurysm, regardless of its location. There are several hemodynamic differences between thoracic and infrarenal aneurysms in vivo. Mainly, thoracic aneurysms have higher flow; as a result of higher elastin content, pulse transmission is more complete in the former; and the presence of reflected pressure waves from

4.5 mm 78.8/100.8 40.4/35.8 66.7/46.5 76.7/47.6

6 mm 80.9/111.4 48.8/66.3 50.0/61.1 55.4/56.4

the iliac arteries is seen in adults older than 50 years because of the impact of the blood flow against posterior plaques. The changes in pressure discussed in this study were not designed to vary with the hemodynamic differences between anatomic aortic segments. Our decision to perform this study was the result of an observation made in a clinical case of an open conversion in a patient with a persistent type I endoleak. Intrasac DP and MP were higher than the systemic pressures. Similar observations have been made by others, who reported this phenomenon anecdotally. The same observation was made by us in type B dissections, in which DP and MP in the false lumen were higher than the systemic pressures. Our experiment clearly shows a decrease in the MP and DP within the sac after complete exclusion. The same residual damped curve was registered. However, although the pressure transducer was not in contact with the graft at any time, the intrasac pressure was never zero. The potential explanation for this finding is that the positive pressure resulted from pressure transmission from the graft acting on a semielastic system (aneurysm). As is the case in patients who undergo endoluminal treatment, the intra-aneurysmal pressure cannot be zero. Two factors produce pressurization inside the sac: (1) pressure inside the endograft that is transmitted by pulsation inside a semirigid container (ie, the aneurysm sac), and (2) the intra-abdominal pressure. The presence of an endoleak causes a significant increase in aneurysm pressure (MP and DP), the extent of which is directly proportional to the endoleak size. Perhaps the most important finding of our study was that even a small size endoleak causes considerable pressure in the aneurysm sac, which in the clinical setting could lead to aneurysm rupture. To explain this, we speculate that the blind end of the aneurysm sac has an inappropriate outflow for the blood that enters the sac during systole. This explanation seems reasonable, because the DP and MP dropped as the outflow of the aneurysm increased. La Place’s law dictates that the wall stress of an artery is proportional to the radius and intraluminal pressure and inversely proportional to wall thickness. This concept predicts that increasing blood pressure or sac diameter should increase wall tension and also the risk of rupture. In the clinical setting, diastolic blood pressure is a more accurate predictor of rupture than systolic blood pressure. In their analysis of patients treated non-operatively, Szilagyi et al22 found a

JOURNAL OF VASCULAR SURGERY Volume 34, Number 5

Parodi et al

913

Table V. Published studies related to intrasac pressure after endoluminal aneurysm exclusion Author

Model

Purpose

Results

Chuter TA14

Animal

Early and late ISP monitoring

Criado E15

Animal

Efficacy of endovascular exclusion ISP measurements, angiography, necropsy, and histologic examination

Faires PL16

Animal

Sanchez LA17

Animal

Schurink GW18

Artificial

Treharne GD19

Human beings

Acute and chronic pressure monitoring and angiographic exclusion Acute and chronic ISP monitoring; CT scan and duplex scan Endoleaking in terms of ISP; CT scan and DSA ISP and SFA flow monitoring (aortouniiliac device)

Chuter TA20

Human beings

ISP monitoring; CT scan

Malina M21

Human beings

PWM; aneurysm sac diameter measurement

Exclusion was associated with reduction of the ISP and reduced aneurysm growth; all controls (without exclusion) died Exclusion was associated with reduction of the ISP and reduced aneurysm growth; all controls (without exclusion) died; all dogs with endoleaks died Exclusion was associated with reduction of the ISP Exclusion was associated with reduction of the ISP In the presence of endoleak, intrasac diastolic pressure was equal to systemic diastolic pressure Exclusion was associated with reduction of the ISP; in the presence of endoleak, failure of sac pressure to fall was registered Exclusion was associated with reduction of the ISP and palpable abdominal pulse Exclusion was associated with reduction of the PWM and aneurysm diameter

CT, Computed tomography; ISP, intrasac pressure; DSA, digital subtraction angiography; SFA, superficial femoral artery; PWM, pulsatile wall motion.

strong relationship between diastolic hypertension and aortic aneurysm rupture. They found that diastolic hypertension was present in 67% to 72% of the patients who experienced rupture, but in only 23% to 29% of the patients without rupture. This correlates with a high risk of rupture in patients with incomplete aneurysm exclusion. After opening the collateral branch, the intrasac MP and DP decreased in proportion to outflow. The patent collateral branch depressurizes the high-pressure state made by the endoleak. The effects of these residual perfusion vessels are unclear. The size, flow, and pressure of the outflow collateral artery in the presence of an endoleak will likely determine its ability to depressurize the aneurysm sac. Several authors studied the effect of patent lumbar arteries as the source of endoleaks.23,24 In some cases, lumbar endoleaks resulted in an increase of the diameter of the aneurysm sac.25 A type II endoleak developed in 17 of our 126 patients who were treated by an industry-made stent-graft. Because of aneurysm enlargement, four of these patients were treated. Several limitations of the study should be noted. The effects of thrombus, patent side branches, or both are unknown and warrant further investigation. The difference in compliance between the human and the experimental aneurysmal wall is also unknown. We elected to use a latex model for our experimental aneurysm, because the silicone model did not show any pulsatility. The thickness of the rubber model was calculated to achieve pulsatility that was as similar as possible to the data published by Schurink et al.18 In this model, the intra-aneurysmal pressure was measured at an acute post-implantation time. Pressure within the excluded aneurysm sac was significantly lower than systemic arterial pressure, maintaining a dampened arterial

waveform. The presence of an endoleak increased the intrasac DP and MP. Incomplete aneurysm exclusion caused by an endoleak results in failure to decrease the intra-aneurysmal pressure and, therefore, confers not only a lack of protection, but may actually predispose the aneurysm to rupture. A variety of devices are currently in clinical use for the endoluminal exclusion of abdominal aortic aneurysms; all have been associated in some cases with persistent perfusion of the aneurysm sac, leading to aneurysm rupture and death. Persistent patency of the aneurysm branches has also been observed. Their effect on intrasac pressure is unclear and is under clinical investigation. Intrasac DP and MP in the presence of an endoleak were higher than the systemic pressures, possibly representing a high risk for aneurysm rupture. Actually, the patient with a type I or type III endoleak conceivably could be in a worse situation than before the aneurysm was treated with an endoluminal graft. The worst scenario would be created when the aneurysm is effectively excluded and, after a variable period, a new type I or type III endoleak develops. Furthermore, because atrophy or hypertrophy of the vessel wall occurs in relation to the variation of the intravessel flow and pressure, atrophy that would lead to weakening of the aneurysm wall can be expected when the aneurysm is isolated. This observation was made in the clinical setting in anecdotal cases of our own. At that stage, lumbar and inferior mesenteric arteries are usually thrombosed. High DP and MP would be present within the sac, and without open outflow vessels, their levels could reach very high values. In addition, atrophy of the wall of the aneurysm after long-term exclusion would further facilitate rupture.

JOURNAL OF VASCULAR SURGERY November 2001

914 Parodi et al

REFERENCES 1. May J, White G, Yu W, Waugh R, Stephen M, Harris J. A prospective study of changes in morphology and dimensions of abdominal aortic aneurysms following endoluminal repair: a preliminary report. J Endovasc Surg 1995;2:343-7. 2. Malina M, Ivancec K, Chuter TAM, Lindh M, Lanne T, Lindblad B. Changing aneurysmal morphology after endovascular grafting: relation to leakage or persistent perfusion. J Endovasc Surg 1997;4:23-30. 3. Zarins CK, White RA, Fogarty TJ. Aneurysm rupture after endovascular repair using the AneuRX stent-graft. J Vasc Surg 2000;31: 960-70. 4. Alimi YS, Chakfe N, Rioval E, Slimane KK, Valeerio N, Riepe G, et al. Rupture of an abdominal aortic aneurysm after endovascular graft placement and aneurysm size reduction. J Vasc Surg 1998;28:178-83. 5. Gilling-Smith G, Brennan J, Harris P, Bakran A, Gould D, McWilliams R. Endotension after endovascular aneurysm repair: definition, classification, and strategies for surveillance and intervention. J Endovasc Surg 1999;6:305-7. 6. White GH, May J, Petrasek P, Waugh R, Stephen M, Harris J. Endotension: an explanation for continued AAA growth after successful endoluminal repair. J Endovasc Surg 1999;6:308-15. 7. Resch T, Ivancev K, Brunkwall J, Nirhov N, Malina M, Lindblad B. Midterm changes in aortic aneurysm morphology after endovascular repair. J Endovasc Ther 2000;7:279-85. 8. Matsumura JS, Pearce WH, McCarthy WJ, Yao JST. Reduction in aortic aneurysm size: early results after endovascular graft placement. J Vasc Surg 1997;25:113-23. 9. Lumsden AB, Allen RC, Chaikot EL, Resnikoff M, Moritz MW, Gerhard H, et al. Delayed rupture of aortic aneurysms following endovascular stent grafting. Am J Surg 1995;170:174-8. 10. Torsello GB, Klenk E, Kasprzak B, Umscheid T. Rupture of abdominal aortic aneurysm previously treated by endograft stent-graft. J Vasc Surg 1998;28:184-7. 11. May J, White GH, Waugh R, et al. Rupture of abdominal aortic aneurysms: concurrent comparison of outcome of those occurring after endoluminal repair versus those occurring de novo. Eur J Vasc Endovasc Surg 1999;18:344-8. 12. White RA, Donayre C, Wallot I, Stewart M. Abdominal aortic aneurysm rupture following endoluminal graft deployment. J Endovasc Ther 2000;7:257-62. 13. Resnikoff M, Darling RC, Chang BB, Lloyd WE, Paty PSK, Leather RP, et al. Fate of excluded abdominal aortic aneurysm sac: long-term follow up of 831 patients. J Vasc Surg 1996;24:851-5.

14. Chuter TA, Viscomi S, Slater JL, et al. Canine model of abdominal aortic aneurysm treated by endovascular graft implantation. Cardiovasc Surg 1997;5:490-6. 15. Criado E, Marston WA, Woosley JET, Ligush J, Chuter TA, Baird C, et al. An aortic aneurysm model for the evaluation of endovascular exclusion prosthesis. J Vasc Surg 1995;22:306-15. 16. Faries PL, Sánchez LA, Marin ML, Parsons RE, Lyon RT, Oliveri S, et al. An experimental model for acute and chronic evaluation of intraaneurysmal pressure. J Endovasc Surg 1997;4:290-7. 17. Sanchez LA, Faries PL, Marin ML, Ohki T, Parsons RE, Marty B, et al. Chronic intraneurysmal pressure measurement: an experimental method for evaluating the effectiveness of endovascular aortic aneurysm exclusion. J Vasc Surg 1997;26:222-30. 18. Schurink GW, Aarts NJ, Wilde J, Baalen JM, Chuter TA, Scholtz Kool LJ, et al. Endoleakage after stent-graft treatment of abdominal aneurysm: implication on pressure and imaging—an in vitro study. J Vasc Surg 1998;28:234-41. 19. Treharne GD, Loftus IM, Thompson MM, Lennard N, Smith J, Fishwick G, et al. Quality control during endovascular aneurysm repair: monitoring aneurysmal sac pressure and superficial femoral artery flow velocity. J Endovasc Surg 1999;6:239-45. 20. Chuter T, Ivancev K, Malina M, Resch T, Brunkwall J, Lindblad B, et al. Aneurysm pressure following endovascular exclusion. Eur J Vasc Endovasc Surg 1997;13:85-7. 21. Malina M, Lanne T, Ivancev K, Lindblad B, Brunkwall J. Reduced pulsatile wall motion of abdominal aortic aneurysms after endovascular repair. J Vasc Surg 1998;27:624-31. 22. Szilagyi DE, Elliott JP, Smith RF. Clinical fate of the patient with asymptomatic abdominal aortic aneurysm and unfit for surgical treatment. Arch Surg 1972;104:600-6. 23. White GH, May J, Waugh R, et al. Type I and type II endoleaks: a more useful classification for reporting results of endoluminal AAA repair [letter]. J Endovasc Surg 1998;5:189-91. 24. White GH, May J, Waugh R, Chaufour X, Yu W. Type III and type IV endoleaks: toward a complete definition of blood flow in the sac after endoluminal AAA repair. J Endovasc Surg 1998;5:305-9. 25. Gorich J, Rilinger N, Sokiranski R, Kramer S, Schutz A, SunderPlassmann L, et al. Embolization of type II endoleaks fed by the inferior mesenteric artery: using the superior mesenteric artery approach. J Endovasc Ther 2000;7:297-301.

Submitted Mar 21 2001; accepted Jul 17, 2001.

IMAGES AND REFLECTIONS A new section in the Journal of Vascular Surgery, Images and Reflections, gives authors the opportunity for reflection by submitting creative writing (prose or poetry), photographs, artwork, and unique aspects of medical history. Submissions must be limited to one journal page. Please contact the Editor before submission.