Endoleaks after endovascular aneurysm repair leadto nonuniform intra-aneurysm sac pressureNuno V. Dias, MD, PhD, Krassi Ivancev, MD, PhD, Timothy A. Resch, MD, PhD,Martin Malina, MD, PhD, and Björn Sonesson, MD, PhD, Malmö, Sweden

Objective: This was a study of intra-aneurysm sac pressures in patients who presented with endoleaks after endovascularrepair (EVAR) of abdominal aortic aneurysms (AAA).Methods: Twenty-five patients (18 men, 7 women) with endoleaks, age (IQR 68 to 80), underwent 31 direct intra-aneurysm sac pressure measurements, DISP at 16 months after EVAR (IQR, 14 to 26 months). Diameter of AAA was 59mm (IQR, 52 to 67 mm). Six patients underwent DISP twice. Tip-pressure sensors were used through direct translumbarpuncture of the AAA except in three patients (transabdominal puncture in 2; endoluminal in 1). Mean pressure index(MPI) was calculated between simultaneously registered intra-aneurysm sac and systemic pressures. Values presented aremedians with interquartile range (IQR).Results: Type I endoleaks (n � 1) showed MPI of 93% in the nidus and 62% in the thrombus. Type II endoleaks wereassociated with lower MPIs in the thrombus (35%; IQR 24% to 38%) when AAAs shrank (n � 4) compared with whenthe AAAs remained unchanged (n � 11; MPI, 78%; IQR, 47% to 85%) or expanded (n � 6; MPI, 74%; IQR, 58% to 87%;P � .019). The nidus of type II endoleaks (MPI, 79%; IQR, 70% to 90%) had higher pressure than the thrombus (45%,IQR, 34% to 85%; P � .047; n � 7). Successful embolization of type II endoleaks led to AAA shrinkage (n � 3; MPIreduction, 13% to 31%) or diameter stability (n � 3; unchanged MPIs, 37% to 44%). Type III endoleaks (n � 3) had MPIsin the thrombus of 33% to 70%.Conclusions: Endoleaks after EVAR pressurize the AAA sac nonuniformly, with higher, near-systemic, pressure in theendoleak nidus compared with the thrombus. Nevertheless, type II endoleaks in shrinking AAAs have lower intra-sacpressure than expanding or stable aneurysms, and successful endoleak embolization reduces pressure. ( J Vasc Surg 2007;

46:197-203.)

Endovascular aneurysm repair (EVAR) of abdominalaortic aneurysms (AAA) is successful when the aneurysmsac is completely excluded.1 Type I and III endoleaks areconsidered treatment failures given their association withadverse clinical outcomes.1 In contrast, the importance oftype II endoleaks is controversial. Although they tend toseal spontaneously and are frequently associated with aneu-rysm stability or shrinkage,2,3 aneurysm expansion, andeven sporadic rupture, has been reported with type IIendoleaks.4,5

Follow-up after EVAR has focused on the evaluation ofendoleaks and AAA size. Imaging techniques allow assess-ment of flow within the aneurysm sac by detecting contrastenhancement and/or Doppler signals. However, intrasacpressure after EVAR is only indirectly evaluated by theseimaging methods by assessing AAA size changes. The intra-aneurysm sac pressure is one of the main determinants ofthe tension applied on the AAA wall. A high or low tensionwill lead to aneurysm expansion or shrinkage, respectively.We have previously verified this relationship in the absenceof endoleaks by performing direct intra-aneurysm sac pres-sure measurements (DISP).6,7 Pressure has also been mea-

From the Department of Vascular Diseases Malmö-Lund and EndovascularCentre, Malmö University Hospital, Lund University.

Competition of interest: none.Correspondence: Nuno V. Dias, Department of Vascular Diseases Malmö-

Lund, Malmö University Hospital, Entrance 59-7th Floor, 205 02Malmö, Sweden (e-mail: nuno.dias@med.lu.se).

0741-5214/$32.00Copyright © 2007 by The Society for Vascular Surgery.

doi:10.1016/j.jvs.2007.04.016

sured in patients with endoleaks, but mostly in type IIendoleaks and shortly after EVAR.8,9

The aim of this study was to evaluate intra-aneurysm sacpressures in patients who presented with endoleaks afterEVAR of AAA

METHODS

Patients. Since 1993, 553 patients have undergoneEVAR of AAA at our tertiary university center. Of these, 25patients (18 men, 7 women); age 76 years old (IQR 68 to80) underwent 31 DISP at 16 months (IQR 14 to 26)months after EVAR (Fig 1). The median duration of DISPwas 174 minutes (IQR 141 to 209). DISP was performedin 21 patients while a type II endoleak was present or hadbeen embolized. Three patients had type III endoleaks atthe time of DISP, and one patient had a type I endoleak.

Inclusion criteria for the study were based on anatomicsuitability for direct sac puncture and the presence of anendoleak after EVAR or the status after endoleak emboli-zation. Patients were grouped according to the type ofendoleak or status after its embolization (Table I). Ourinitial experience with DISP in eight of these patients hasbeen included in a previous report.7

The study was approved by the local ethics committee,and all patients gave informed consent before the proce-dure.

Imaging. The imaging protocols included contrast-enhanced spiral-computed tomography (CT) scans preop-eratively and yearly after EVAR. More frequent CT scans

were performed when indicated. Preoperative CT scans

197

seco

JOURNAL OF VASCULAR SURGERYAugust 2007198 Dias et al

included two spiral scans that were done before and afterintravenous nonionic iodinated contrast injection. Postop-erative CT studies included an additional delayed spiralscan. CT scans were reconstructed with 0.75-mm to 3-mmaxial slices. An additional CT-scan was obtained the monthbefore DISP (pre-DISP CT scan).

The shortest transverse diameter of the AAA was mea-sured at its widest portion on axial CT scans by the sameobserver. Diameter changes were calculated to express thediameter evolution before DISP (diameter at the time ofDISP - initial diameter).7 A 5-mm cutoff was used forgrouping AAAs into those shrinking by �5 mm, un-changed, and expanding by �5 mm.1

At the time of DISP, a digital subtraction aortogram

Fig 1. Flow chart shows patients undergoing direct intthe type of endoleak (EL) and abdominal aortic aneurysmsuccessful embolization of type II endoleaks. *Two patiemeasurement was after AAA diameter shrinkage, and the

Table I. Characteristics of patients

Group Patients (n) Age

EndoleakType I 1 81Type II 19* 77 (68-81)

EL control 6† 74 (70-79)Type III 3 73 (64-75)

AAA, Abdominal aortic aneurysm; EVAR, endovascular aneurysm repair; DValues are shown as median with interquartile range in parenthesis*Two patients of the 19 patients with type II endoleaks were measured twexpanded.†Four of these patients were also measured before the embolization of the ewere measured only after a previous embolization of the endoleak).

was performed. This included selective angiography of the

superior mesenteric and hypogastric arteries whenever theorigin of the type II endoleak was not clearly established byCT scan and nonselective aortography.

Endoleaks were classified according to the recom-mended reporting standards.1 Endoleak nidus was de-fined as the contrast filled area within the aneurysm sac.A pressure measurement was considered within the en-doleak nidus when this location was verified by contrastinjection and free blood flow was obtained from theneedle.

Pressure measurement system and direct intra-an-eurysm sac pressure technique. Anatomic suitability wasassessed in the pre-DISP CT scan and defined as the possi-bility of inserting a pressure sensor in the AAA sac without

eurysm sac pressure measurements (DISP) according toA) diameter change. Six patients underwent DISP afterith type II endoleaks underwent DISP twice; the initialnd occurred once the AAA had expanded.

diameter (mm) Time between EVAR and DISP (months)

55 1562 (53-73) 16 (14-22)58 (53-62) 30 (20-48)55 (52-62) 41 (2-97)

direct intra-aneurysm sac pressure measurement; EL, endoleak.

e first time after diameter shrinkage and thereafter once the diameter had

k and are, therefore, also included in type II endoleak group (ie, 2 patients

ra-an(AA

nts w

AAA

ISP,

ice. Th

ndolea

damaging any viscera or jeopardizing stent graft integrity.

ries (

JOURNAL OF VASCULAR SURGERYVolume 46, Number 2 Dias et al 199

The pressure measurement system and DISP techniquehave been previously described in detail.6,7,10 Wired tip-pressure sensors mounted on 0.014-inch guidewires wereused (PressureWire4, RADI Medical AB, Uppsala, Swe-den). The pressure guidewire used for intra-aneurysm sacpressure measurements had a shorter tip (1 mm instead of 3cm) to allow precise placement of the sensor. Systemicpressure was measured in the lumen of the stent-graft.

Access to the AAA sac was obtained by translumbarpuncture using fluoroscopic guidance in 28 occasions, in-cluding the six late embolization controls. Two DISP weredone through direct transabdominal ultrasound-guidedAAA puncture. During the direct puncture of the AAA sac,efforts were made to pinpoint the endoleak nidus (Fig 2,A). Measurements in the thrombus were made when thesensor was located approximately mid-way between thenidus and the stent graft. On one occasion, the pressureguidewire was passed through a coaxial catheter placed inthe ostium of the inferior mesenteric artery through thesuperior mesenteric artery.

The system recorded systemic and AAA sac pressuressimultaneously. Intra-aneurysm sac pressures were ana-lyzed as systolic, diastolic, mean, and pulse pressures. Meanpressure index (MPI) was calculated as the percentage ofthe mean intra-aneurysmal pressure relative to the simulta-neous mean systemic pressure [MPI � (mean AAA pres-sure/mean systemic pressure) � 100].

Measurements were only considered of good quality ifthe drift in recalibration of the pressure sensor did notexceed 5 mm Hg in the end of the measurements. DISP hasbeen previously validated in patients with a median intraob-server variability of MPI of 0% (IQR �7 to 17%).10 Thesensor was also tested in vitro with an accuracy of betterthan 2 mm Hg.10,11

Endoleak embolization. Embolization of the en-doleak was performed once access to the endoleak nidus or

Fig 2. A, Digital subtraction angiography shows contthrough a translumbar puncture of the abdominal aortimeasurement. Two lumbar arteries are filled (arrows). B(Braun, Tuttlingen, Germany) and Lipiodol (Laboratoendoleak nidus and the proximal part of two lumbar arte

its feeding vessels was achieved (Fig 2, B). The embolizing

materials used were coils, radiopaque glue (Histoacryl,Braun, Tuttlingen, Germany and Lipiodol, LaboratoireGuerbet, Aulnay-Sous-Bois, France), or gel-foam sponge(Spongostan Standard, Johnson & Johnson Medical Ltd,Bracknell Berkshire, United Kingdom). Late embolizationcontrol was defined as DISP being performed on a differentoccasion than the embolization.

Statistical analysis. Normal distribution was not as-sumed. Values are presented as medians and interquartilerange (IQR), if not stated otherwise. Nonparametric exacttests were used for paired and unpaired comparisons. Re-sults were considered significant at P � .05. Statisticalanalysis was done using SPSS 12.0.1 software (SPSS Inc,Chicago, Ill).

RESULTS

Type I endoleak. One patient, an 81-year-old man(AAA diameter, 55 mm) with a proximal type I endoleakand unchanged AAA diameter (0 mm), underwent DISP15 months after EVAR. MPI was 62% and pulse pressurewas 21 mm Hg in the thrombus and 93% and 66 mm Hg inthe nidus, respectively. The AAA expanded 9 mm in the 28months after DISP until the patient consented to a reinter-vention.

Type II endoleaks. Nineteen patients (12 men; Ta-ble I) underwent 21 DISP while a type II endoleak waspresent. Two patients underwent DISP twice, becauseno embolization was performed on the first occasion andthe aneurysms subsequently expanded (6 mm after 8months and 11 mm after 11 months, respectively). Atthe time of the pressure measurement four AAAs haddecreased in size, 11 remained unchanged, and six hadexpanded. MPI, intra-aneurysm mean and pulse pres-sures were significantly lower in the thrombus of shrink-ing AAAs compared with AAAs that expanded or re-mained unchanged in size (P � .019, P � .030 and P �

njection inside a type II endoleak nidus (arrowheads)urysm at the time of direct intra-aneurysm sac pressuretus after embolization of the endoleak with Histoacryluerbet, Aulnay-Sous-Bois, France). The glue fills thearrow).

rast ic ane, Staire G

.019, respectively; Table II).

re to

JOURNAL OF VASCULAR SURGERYAugust 2007200 Dias et al

Pressure in the thrombus and nidus of type IIendoleaks. Pressure was measured within the thrombusand inside the endoleak nidus during the same puncture in7 patients (Fig 3). Pressure was significantly higher insidethe endoleak nidus, with MPI at 79% (IQR, 70% to 90%)and pulse pressure of 22 mm Hg (IQR, 18 to 57 mm Hg)than within the thrombus, with MPI at 45% (IQR, 34% to85%) and pulse pressure of 11 mm Hg (IQR, 7 to 21 mmHg; P � .043 and P � .028, respectively). The median MPIdifference between the nidus and the thrombus was 32%(IQR, 4% to 44%).

Type II endoleak embolization. Thirteen patientsunderwent embolization of type II endoleaks at the time ofDISP. Glue was used in 12 cases and coils and gel-foamparticles in one. Twelve of the 13 procedures were consid-ered successful, and the AAA diameter changed 0 mm(IQR, �7 to 2 mm) during the following 13 months (IQR,6 to 37 months). In five of these patients, pressure wasmeasured immediately after the embolization at the same

Table II. Intra-aneurysm sac pressure in the presence of t

Diameter changebefore DISP N

Follow-up beforeDISP (mo)

�AAA diame(mm)

Shrinking 4 19 (7-43) �6 (�9, �5Unchanged 11 15 (13-16) 2 (0, 3)Expanding 6 22 (20-31) 7 (6,10)

DISP, Direct intra-aneurysm sac pressure measurement; N, number of DISGrouping was done according to the AAA diameter change.Values are shown as median with interquartile range in parenthesis.*P � .019.†P � .030.‡P � .019.

Fig 3. Recording from direct intra-aneurysm pressureMeasurement in the endoleak nidus. Red curves repintra-aneurysm sac pressure. Each figure has three tracingrecording where the aortic and intra-aneurysm sac preduring 10 heart cycles is gathered separately in two differight) and the other intra-aneurysm sac pressures (lower smof the systolic, diastolic, and mean pressures for both theIn the smaller tracings the scale is adjusted by the softwa

procedure. MPI changed from 70% (IQR, 57% to 84%)

before the embolization to 53% (IQR, 30% to 79%) imme-diately after (P � .05). Intra-aneurysm sac mean and pulsepressure changed from 86 mm Hg (IQR, 56 to 98 mm Hg)and 13 mm Hg (IQR, 12 to 19 mm Hg) before theembolization to, respectively, 56 mm Hg (IQR, 31 to 111mm Hg) and 12 mm Hg (IQR, 8 to 40 mm Hg) after theembolization (P � .05). There was no association betweenthe final MPI and the AAA diameter change afterwards.

One embolization was unsuccessful, and the endoleakpersisted at the end of the procedure. This patient wasconsidered unfit for any further reintervention and theaneurysm continued to expand (19 mm during the follow-ing 26 months).

Late control of type II endoleak embolization. Sixpatients (5 men, 1 women) aged 74 years (IQR, 70 to 79years; Table I) underwent DISP at 16 months (IQR, 6 to31 months) after successful embolization of type II en-doleaks. Three patients with MPIs of �31% at the lateembolization control showed AAA shrinkage. The MPIs

I endoleaks

AAA mean pressure(mm Hg)

AAA pulse pressure(mm Hg) MPI (%)

40 (24-47)* 7 (6-10)† 35 (25-38)‡

86 (53-97)* 21 (12-22)† 78 (47-85)‡

76 (59-109)* 17 (11-31)† 74 (58-87)‡

ormed; AAA, abdominal aortic aneurysm; MPI, mean pressure index.

surement. A, Measurement within the thrombus. B,the aortic pressure, and green curves represent the

e large one to the left is the tracing obtained during eachare registered simultaneously. The pressure recorded

tracings, one for the aortic (upper smaller tracing to thetracing to the right). The system calculates also the meanand intra-sac pressures (values below the small tracings).the amplitude of the curves.

ype I

ter

)

P perf

mearesents. Thssuresrentaller

aortic

were lower after the embolization (Table III). Three pa-

fter th

JOURNAL OF VASCULAR SURGERYVolume 46, Number 2 Dias et al 201

tients with unchanged AAA diameter after the emboliza-tion of the type II endoleak had MPIs between 37% and44% at the late embolization control. The MPIs had notchanged significantly at the late control compared withbefore the embolization (Table III).

Type III endoleaks. Three patients (3 men, Table I)underwent DISP in the presence of a type III endoleak.One patient with a small type III endoleak persisting afterEVAR underwent DISP 2 months after the EVAR. MPIsranged from 70% close to the endoleak to 42% in theperiphery of the AAA sac. The endoleak sealed spontane-ously afterwards and the AAA diameter shrank by 12 mm.

The other two patients underwent DISP when late typeIII endoleaks were identified (one partial separation of thecomponents of an aortouniiliac system and one small fabricdisruption), and the AAAs remained unchanged in diame-ter (�2 and �1 mm, respectively). Intra-sac pressurewithin the thrombus decreased with increasing distance tothe endoleak (MPIs range, 57% to 35% and 44% to 33%,respectively). No measurements were performed in theendoleak nidus. One patient refused any further reinterven-tions and died within 1 month of unknown cause. Theother patient showed a 6-mm expansion of the AAA in the4 months after DISP and was converted to open repair.

DISCUSSION

The aim of EVAR is the exclusion of the aneurysm sacfrom blood flow and pressure. AAA size and endoleakstatus have been the key points in the evaluation of thetreatment success. However, intra-aneurysm sac pressurehas, until recently, been mostly indirectly assessed. The useof tip pressure sensors for direct intra-aneurysm sac pressuremeasurement after EVAR has been previously validat-ed.6,7,10 Shrinking AAAs without endoleaks were shown tohave lower intra-aneurysm sac pressure within the throm-bus compared with expanding aneurysms.7 Furthermore,DISP was able to predict future AAA diameter changes inthe absence of endoleaks.

The effect of endoleaks after EVAR in intra-sac pressureis incompletely understood, however, especially consider-ing the varied clinical outcome associated with the differenttypes of endoleaks. The results of this study show that theendoleak nidus (channel) has consistently higher pressure

Table III. Pressure in six patients undergoing late endole

PatientMPI before

embolization (%)MPI after

embolization (%)

Intra-ane(

Systolic

1 70 31 342 NA 13 223 67 19 254 NA 42 535 45 44 456 39 37 60

MPI, Mean pressure index; AAA, abdominal aortic aneurysm.The follow-up after the embolization also includes the imaging follow-up a

than the intra-sac thrombus.

Type II endoleaks seem to be associated with varyingintra-sac pressures in the thrombus relating to the AAAdiameter changes. Moreover, type II endoleaks appear tobe dynamic because the variation can occur within the samepatient at different time points. Successful embolization ofa type II endoleak leads to depressurization of the AAA sac,with shrinkage or stabilization of the aneurysm diameter.Type I and III endoleaks usually lead to a pressurization ofthe AAA sac, as it was anticipated.

DISP in one patient with a type I endoleak showed ahigh pulsatile intra-aneurysm sac pressure in the thrombusand even higher and near-systemic pressure in the nidus.This was followed by AAA expansion and reinforces theperception of these endoleaks as treatment failures.

Type II endoleaks were associated with varying degreesof intrasac pressurization. The previously reported differ-ence in intra-sac pressure between shrinking and expandingAAAs without endoleaks7 was also verified in type II en-doleaks. However, AAA shrinkage with a type II endoleakwas associated with higher intrasac pressures in the throm-bus than what was previously seen in AAA shrinkage with-out endoleaks (MPI of 35% and 19%, respectively).7

Published reports have consistently shown near-systemicpressure in patients with type II endoleaks in the early fol-low-up after EVAR.8,9 These measurements, however, wereperformed mainly within the endoleak nidus, and as ourresults show, there is a pressure difference between the en-doleak nidus and the AAA thrombus. The difference in pres-sure between the nidus and the thrombus seems to depend onseveral factors, including the size of the endoleak nidus,12,13

its flow and perfusion pressure,14-16 and thrombus composi-tion. In our experience, a discrepancy in the endoleak size wasseen in some patients on the CT scan compared with theaneurysmogram with contrast injection into the nidus.

The composition and structural properties of thethrombus, and their dependency on the proximity of bloodcirculation, may also influence the transmission of pres-sure.17 Furthermore, the amplitude of the pressure gradi-ent measured between the nidus and the thrombus mayalso be influenced by the distance between the measure-ment locations. This varying distribution of sac pressure inpatients with endoleaks, although consistently higher in

bolization control

sac pressureg)

Follow-up afterembolization (mo)

�AAA diameterafter embolization (mm)olic Mean

30 44 �720 72 �1821 50 �645 39 �140 11 158 9 �1

e second direct intra-aneurysm sac pressure measurement.

ak em

urysmmm H

Diast

281918403857

expanding AAAs, may question the reliability of systems

JOURNAL OF VASCULAR SURGERYAugust 2007202 Dias et al

based on pressure measurements in a single spot, such aswith implantable pressure sensors.18

The use of implantable devices has recently been re-ported in the intra-operative identification of endoleaks19

and in canine models of type II endoleaks.20 However, thelocation of the pressure transducer in relation to the en-doleak nidus may influence the measurements obtainedwith those devices. For instance, if the pressure transduceris placed within the endoleak nidus, measurements can behigh without necessarily implying AAA expansion.

Type II endoleaks seem to be a dynamic entity, becausemany seal spontaneously early after EVAR. This dynamiccharacter of type II endoleaks, the difference in intra-sacpressure between patients with and without endoleaks, andthe pressure gradient between the nidus and thrombussuggest the need for a cautious interpretation of intrasacpressure in patients with endoleaks.

Direct percutaneous puncture of the AAA allows notonly intrasac pressure measurement but also embolizationof type II endoleaks. This is a significant advantage, becausetranslumbar embolization has been shown to be moreefficient9,21-23 than endoluminal methods.24-27 Successfultype II endoleak embolization was associated with shrink-age or stabilization of the aneurysm diameter. The diameterstability may be partially explained by the incompressibilityof the embolic material injected into the sac. Intra-aneu-rysm sac pressure appears, nevertheless, to decrease uponsuccessful embolization. In contrast to others,8 we did notfind this pressure reduction directly after the embolizationbut only later during follow-up. It is possible that byperforming our initial pressure measurement later in thefollow-up after EVAR, we have measured a more selectedgroup of patients where the endoleaks that sealed sponta-neously in the early follow-up had been excluded.

CONCLUSION

The endoleak nidus had high, near-systemic pressure,but the degree of thrombus pressurization varied. Thus,pressure is not uniformly distributed through the AAA sac.Shrinking AAAs with type II endoleaks had lower pressurethan AAAs with an expanding or unchanged diameter.Successful endoleak embolization leads to a delayed depres-surization of the AAA sac. Type I and III endoleaks areassociated with high intrasac pressures.

We are very grateful to Kristina Lindholm, PhD andChristel Ekberg-Jönsson for their assistance.

AUTHOR CONTRIBUTIONS

Conception and design: ND, KI, BSAnalysis and interpretation: ND, TR, BSData collection: ND, KI, MM, BSWriting the article: ND, TR, BSCritical revision of the article: KI, MMFinal approval of the article: ND, KI, TR, MM, BSStatistical analysis: NDObtained funding: ND, KI, BS

Overall responsibility: ND, BS

REFERENCES

1. Chaikof EL, Blankensteijn JD, Harris PL, White GH, Zarins CK,Bernhard VM, et al. Reporting standards for endovascular aortic aneu-rysm repair. J Vasc Surg 2002;35:1048-60.

2. Resch T, Ivancev K, Lindh M, Nyman U, Brunkwall J, Malina M, et al.Persistent collateral perfusion of abdominal aortic aneurysm after endo-vascular repair does not lead to progressive change in aneurysm diame-ter. J Vasc Surg 1998;28:242-9.

3. Rhee RY, Eskandari MK, Zajko AB, Makaroun MS. Long-term fate ofthe aneurysmal sac after endoluminal exclusion of abdominal aorticaneurysms. J Vasc Surg 2000;32:689-96.

4. White RA, Donayre C, Walot I, Stewart M. Abdominal aortic aneurysmrupture following endoluminal graft deployment: report of a predict-able event. J Endovasc Ther 2000;7:257-62.

5. van Marrewijk CJ, Fransen G, Laheij RJ, Harris PL, Buth J. Is a type IIendoleak after EVAR a harbinger of risk? Causes and outcome of openconversion and aneurysm rupture during follow-up. Eur J Vasc Endo-vasc Surg 2004;27:128-37.

6. Sonesson B, Dias N, Malina M, Olofsson P, Griffin D, Lindblad B,Ivancev K. Intra-aneurysm pressure measurements in successfully ex-cluded abdominal aortic aneurysm after endovascular repair. J Vasc Surg2003;37:733-8.

7. Dias NV, Ivancev K, Malina M, Resch T, Lindblad B, Sonesson B.Intra-aneurysm sac pressure measurements after endovascular aneurysmrepair: differences between shrinking, unchanged, and expanding aneu-rysms with and without endoleaks. J Vasc Surg 2004;39:1229-35.

8. Baum RA, Carpenter JP, Cope C, Golden MA, Velazquez OC, NeschisDG, et al. Aneurysm sac pressure measurements after endovascularrepair of abdominal aortic aneurysms. J Vasc Surg 2001;33:32-41.

9. Rial R, Serrano Fj F, Vega M, Rodriguez R, Martin A, Mendez J, et al.Treatment of type II endoleaks after endovascular repair of abdominalaortic aneurysms: translumbar puncture and injection of thrombin intothe aneurysm sac. Eur J Vasc Endovasc Surg 2004;27:333-5.

10. Dias NV, Ivancev K, Malina M, Hinnen JW, Visser M, Lindblad B, SonessonB. Direct intra-aneurysm sac pressure measurement using tip-pressure sensors:in vivo and in vitro evaluation. J Vasc Surg 2004;40:711-6.

11. Kälvesten E, Smith L, Tenerz L, Stemme G. The first surface microma-chined pressure sensor for cardiovascular pressure measurements. Pro-ceedings of the 11th Annual International Workshop on Micro-Elec-tro-Mechanical Systems, Heidelberg, Germany; 1998:574-9.

12. Timaran CH, Ohki T, Rhee SJ, Veith FJ, Gargiulo NJ 3rd, Toriumi H,et al. Predicting aneurysm enlargement in patients with persistent typeII endoleaks. J Vasc Surg 2004;39:1157-62.

13. Timaran CH, Ohki T, Veith FJ, Lipsitz EC, Gargiulo NJ 3rd, Rhee SJ, etal. Influence of type II endoleak volume on aneurysm wall pressure anddistribution in an experimental model. J Vasc Surg 2005;41:657-63.

14. Arko FR, Filis KA, Siedel SA, Johnson BL, Drake AR, Fogarty TJ, et al.Intrasac flow velocities predict sealing of type II endoleaks after endo-vascular abdominal aortic aneurysm repair. J Vasc Surg 2003;37:8-15.

15. Parent FN, Meier GH, Godziachvili V, LeSar CJ, Parker FM, CarterKA, et al. The incidence and natural history of type I and II endoleak: a5-year follow-up assessment with color duplex ultrasound scan. J VascSurg 2002;35:474-81.

16. Li Z, Kleinstreuer C. Computational analysis of type II endoleaks in astented abdominal aortic aneurysm model. J Biomech 2006;39:2573-82.

17. Vallabhaneni SR, Gilling-Smith GL, Brennan JA, Heyes RR, Hunt JA,How TV, et al. Can intrasac pressure monitoring reliably predict failureof endovascular aneurysm repair? J Endovasc Ther 2003;10:524-30.

18. Ellozy SH, Carroccio A, Lookstein RA, Jacobs TS, Addis MD, Teodor-escu VJ, et al. Abdominal aortic aneurysm sac shrinkage after endovas-cular aneurysm repair: correlation with chronic sac pressure measure-ment. J Vasc Surg 2006;43:2-7.

19. Ohki T, Ouriel K, Silveira PG, Katzen B, White R, Criado F, et al. Initialresults of wireless pressure sensing for endovascular aneurysm repair:The APEX Trial--Acute Pressure Measurement to Confirm AneurysmSac EXclusion. J Vasc Surg 2007;45:236-42.

20. Chaer RA, Trocciola S, Derubertis B, Hynecek R, Xu Q, Lam R, et al.Evaluation of the accuracy of a wireless pressure sensor in a caninemodel of retrograde-collateral (type II) endoleak and correlation with

histologic analysis. J Vasc Surg 2006;44:1306-13.

JOURNAL OF VASCULAR SURGERYVolume 46, Number 2 Carpenter 203

21. Baum RA, Carpenter JP, Golden MA, Velazquez OC, Clark TW,Stavropoulous SW, et al. Treatment of type 2 endoleaks after endovas-cular repair of abdominal aortic aneurysms: comparison of transarterialand translumbar techniques. J Vasc Surg 2002;35:23-9.

22. Martin ML, Dolmatch BL, Fry PD, Machan LS. Treatment of type IIendoleaks with Onyx. J Vasc Interv Radiol 2001;12:629-32.

23. Schmid R, Gurke L, Aschwanden M, Stierli P, Jacob AL. CT-guidedpercutaneous embolization of a lumbar artery maintaining a type IIendoleak. J Endovasc Ther 2002;9:198-202.

24. Kasirajan K, Matteson B, Marek JM, Langsfeld M. Technique andresults of transfemoral superselective coil embolization of type II lum-

bar endoleak. J Vasc Surg 2003;38:61-6.

sac by means of implanted remote pressure sensors has been

25. Haulon S, Tyazi A, Willoteaux S, Koussa M, Lions C, Beregi JP.Embolization of type II endoleaks after aortic stent-graft implanta-tion: technique and immediate results. J Vasc Surg 2001;34:600-5.

26. Karch LA, Henretta JP, Hodgson KJ, Mattos MA, Ramsey DE, McLaf-ferty RB, et al. Algorithm for the diagnosis and treatment of endoleaks.Am J Surg 1999;178:225-31.

27. Chuter TA, Faruqi RM, Sawhney R, Reilly LM, Kerlan RB, Canto CJ,et al. Endoleak after endovascular repair of abdominal aortic aneurysm.J Vasc Surg 2001;34:98-105.

Submitted Feb 14, 2007; accepted Apr 3, 2007.

INVITED COMMENTARY

Jeffrey P. Carpenter, MD, Philadelphia, Pa

Rupture of aneurysms after endovascular aneurysm repair(EVAR) happens at an alarming rate of 1% per year according toboth European Collaborators on Stent-Graft Techniques for AAAand Thoracic Aortic Aneurysm and Dissection Repair (EURO-STAR) and Lifeline Registry data. Continued, careful monitoringof patients after EVAR is essential and clearly mandated by thissobering statistic. How to do so, however, remains problematic.Repeated contrast exposure (and radiation exposure) is detrimen-tal, as evidenced by the more precipitous decline in renal functionnoted in EVAR cohorts compared with open surgical aneurysmrepair cohorts. Duplex scanning as an alternative method of mon-itoring patients for endoleak has performed well in the hands ofsome, but not all.

Direct puncture of aneurysm sacs has allowed us to directlyaccess endoleaks, ascertain their significance, and treat them. Bythis method we have learned that even seemingly small and innoc-uous type II endoleaks may transmit systemic pressure to theaneurysm wall, placing patients at risk for fatal rupture of theiraneurysm. This invasive method, however, is not practical as ascreening or monitoring test.

Continuous measurement of pressure within the aneurysm

proposed as another alternative method of monitoring patientsafter EVAR. This technique is attractive in that it gets to the“bottom line,” the measurement of pressure within the sacitself. One of the challenges to this technology, however, is thephenomenon of differential distributions of pressure spreadthroughout the thrombus-filled aneurysm sac. This observationhas been made in animal models of aortic aneurysms and nowdirectly in human subjects.

Pressure sensor technology is in its infancy. In addition tomalfunctions of the sensors, false-negative study results have beenreported. The report of Dias et al raises further concerns about thea priori ability of a pressure sensor to accurately detect an endoleakunless it is in direct communication with that leak. Sensors locatedwithin the thrombus will suffer progressive dampening in propor-tion to their distance from the endoleak and may fail to detect aleak that could be transmitting significant pressure to the aneurysmwall elsewhere. The future of pressure sensor technology for EVARmonitoring is as yet unclear, and a long-term observational trial hasyet to be performed. The differential distribution of pressureswithin a thrombus filled aneurysm sac will need to be an important

consideration in the future development of this technology.