Can Intrasac Pressure Monitoring Reliably Predict ... - SAGE Journals

524

J ENDOVASC THER 2003;10:524–530

l VASCULAR

l

FELLOWS’ FORUM 2003, SECOND PLACE

Can Intrasac Pressure Monitoring Reliably Predict Failure of Endovascular Aneurysm Repair? Srinivasa R. Vallabhaneni, FRCS; Geoffrey L. Gilling-Smith, MS, FRCS; John A. Brennan, MD, FRCS; Richard R. Heyes, MSc*; John A. Hunt, PhD*; Thien V. How, PhD*; and Peter L. Harris, MD, FRCS Regional Vascular Unit, Royal Liverpool University Hospital and *Department of Clinical Engineering, University of Liverpool, England, UK l

l

Purpose: To determine if pressure measured at a single location within aneurysm sac thrombus accurately reflects the force applied to the aneurysm wall and the risk of rupture by examining (1) if pressure is distributed uniformly within aneurysm thrombus, (2) the pressure transmission through aneurysm thrombus, and (3) the microstructural basis for pressure transmission. Methods: Pressure within aneurysm thrombus was measured by direct puncture through the aneurysm wall at 121 sites in 26 patients during open abdominal aortic aneurysm repair. Measurements were taken prior to cross clamping and compared with intrasac pressure measured at 30 sites in 6 patients without aneurysm thrombus (controls). Transmission of pressure through aneurysm thrombus was further examined ex vivo by subjecting fresh thrombus to a pressure gradient in a custom-made pressure cell. Pressure transmission was correlated with matrix density as determined by light microscopy and image analysis. Results: Mean pressure within aneurysm thrombus was higher than mean systemic pressure in 11 patients, lower in 1, and identical in 9. In 5 patients, the pressure was greater than systemic in some areas of the thrombus but less in others. Sac pressure was identical to systemic pressure at all sites in the controls. In 12 thrombus specimens (6 patients) examined in the pressure cell, pressure transmission varied significantly between specimens, correlating directly with matrix density (R250.747, p50.001). Conclusions: Pressure transmission through aneurysm thrombus is variable and depends upon the microstructure of the thrombus. Pressure measured at a single location may not, therefore, accurately reflect the pressure acting on the aneurysm wall. J Endovasc Ther 2003;10:524–530 Key words: abdominal aortic aneurysm, endovascular repair, endotension, surveillance, thrombus, intra-aneurysm pressure, experimental model l

l

Patients who have undergone endovascular repair of abdominal aortic aneurysm (AAA) require lifelong surveillance to monitor endograft integrity and to ensure that the an-

eurysm remains isolated from the circulation and free from the risk of rupture. Since the risk of rupture depends primarily upon the pressure within the aneurysm sac, the aim of

Sponsorship provided by a grant from The Royal Liverpool University Hospital R&D Support Fund. The Vascular Fellows’ Research Award competition held on February 10, 2003, at International Congress XVI on Endovascular Interventions (Scottsdale, Arizona, USA) evaluated participants on both their oral and written presentations. ISES, co-sponsors of this annual competition, congratulates the 2003 award winners. Address for correspondence and reprints: SR Vallabhaneni, FRCS, Link 8C, Royal Liverpool University Hospital, Prescot Street, Liverpool L7 8XP, England, UK. Fax: 44-151-706-5803; E-mail: [email protected] Q 2003 by the INTERNATIONAL SOCIETY

OF

ENDOVASCULAR SPECIALISTS

Available at www.jevt.org


DISTRIBUTION OF PRESSURE WITHIN AAA Vallabhaneni et al.

525

surveillance must be to determine whether or not the sac remains or is again pressurized (endotension).1,2 The pressure within the aneurysm sac can be determined by percutaneous needle puncture,3 but this is invasive and risks both perforation of the endograft and introduction of infection. There is, therefore, considerable interest in the concept of remote monitoring of intrasac pressure by means of an implantable sensor.4,5 Reliance on such a sensor to determine whether or not the aneurysm is at risk of rupture assumes that pressure is uniformly distributed within the sac such that measurement at a single location reflects the pressure applied to the entire aneurysm wall. There is, in fact, no evidence to support such an assumption, so we investigated the transmission of pressure through aneurysm thrombus both in vivo and ex vivo.

METHODS In Vivo Pressure Distribution Within Aneurysm Thrombus A study protocol approved by the Liverpool Research Ethics Committee was designed to measure pressure within infrarenal aneurysm thrombus during conventional open AAA repair. All patients gave informed consent to participate. Pressure was measured by direct puncture of the aneurysm wall with a 19-G needle attached to a pressure transducer (BD, Franklin Lakes, NJ, USA) via a 100-cm-long, nondistensible extension tube (1-mm internal diameter, 2-mm external diameter). A Silastic sleeve was applied over the needle to ensure that pressure was always measured at a depth of 3 mm (Fig. 1). After calibration of the transducer and before application of the aortic cross clamp, pressure within the aneurysm thrombus was measured and expressed as a percentage of simultaneously recorded radial artery pressure. Measurements were recorded after allowing the pressure readings to stabilize. The pressure transducer included an integrated flushing device that continuously delivered minute volumes of saline through the pressure line and needle, allowing rapid detection of needle occlusion reflected by a progressive

Figure 1 l Measurement of pressure within aneurysm thrombus in vivo (X: sites of measurement).

rise in pressure rather than stabilization. When this occurred, the needle was withdrawn, and no further attempts were made to measure pressure at that location. Pressure was measured at up to 6 different sites at least 1-cm apart in each aneurysm (Fig. 1). Sampling sites were determined from the distribution of thrombus seen on the preoperative computed tomographic (CT) images so that pressure was always measured within thrombus. Once all pressure measurements had been obtained from within aneurysm thrombus, the needle was reinserted at a separate location and advanced into the aortic lumen to measure systemic aortic pressure. Because pressure distribution within an aneurysm may also be affected by turbulent flow within the sac, similar measurements were taken in 6 patients without aneurysm thrombus as controls.

Ex Vivo Pressure Transmission Through Thrombus A specially designed apparatus was constructed with 2 saline-filled pressure cells A

526



Figure 2 l Diagrammatic representation of the pressure cell apparatus used to examine pressure transmission through thrombus ex vivo.

and B (Fig. 2). Aneurysm thrombus removed at open operation was cut into disks and placed in between the 2 cells. The disk of thrombus was supported on one side by a mesh and on the other by a ring to prevent movement or bulging. A constant pressure of 120 mmHg was applied to cell A, while the pressure in the closed cell B was continuously monitored by means of a catheter tip transducer. The gap between the edges of the disk and the pressure cell were sealed with molten 10% gelatin (Type A Porcine Gelatin; Sigma Diagnostics, Poole, UK.) in distilled water so that the only route through which pressure could be transmitted was via the specimen itself. Examining the transmission of pressure through material of known porosity was performed first to validate the reliability and reproducibility of the pressure cell. Open-cell foam disks were tested as positive controls and were found to offer negligible resistance to pressure transmission. When silicone disks were tested as negative controls (expected transmission zero), the pressure within cell B increased by ,5 mmHg over 2 hours. Validation was repeated 3 times to confirm reproducibility of the technique. The pressure transducer (Gaeltec Ltd., Dunvegan, Scotland) was calibrated before each experiment to construct a pressure/voltage (P/V) curve.

Transmission of pressure through aneurysm thrombus was examined as soon as possible after harvest and always within 6 hours (requiring some specimens to be refrigerated up to a maximum of 4 hours at 48C). Twelve thrombus specimens from 6 patients were each cut into a 25-mm-diameter disk of 10-mm thickness. With the test disk inserted into the model, pressure was recorded as DC voltage at 1-second intervals with an analogdigital acquisition system on to a desktop personal computer using HP-VEE 5.0 software (Hewlett Packard, Palo Alto, CA, USA). The data were transferred to an EXCEL spreadsheet, and pressure values were computed using the P/V curve. The pressure recorded in cell B was expressed as a percentage of the pressure applied to cell A (120 mmHg).

Histological Examination and Image Analysis After ex vivo pressure testing, thrombus specimens from 6 patients were removed from the pressure cell, cut into 5 to 7-mm cubes, and fixed for histological examination in a freshly prepared paraformaldehyde-based fixative for 24 hours at 48C over rollers. The specimens were then washed with phosphate-buffered saline for an additional 24 hours, after which they were embedded in



glycol methacrylate resin. To prevent polymerization of the resin before tissue infiltration was complete, the specimens were immediately transferred to 2608C for 4 days then taken up to 2208C to initiate polymerization in 2 days. Four-mm-thick sections were cut in 2 planes perpendicular to each other and stained with hematoxylin-eosin (12 sections from 6 thrombus samples). The histological sections were examined by light microscopy at 253 magnification using a Jeneval light microscope (Carl Zeiss, Oberkochen, Germany) and a JVC CCD camera using KS400 image analysis software (Imaging Associates, London, UK). In all, 60 random fields from each specimen were analyzed using an automated routine to measure voids in the amorphous stained material (matrix). The ratio of matrix to total field area (matrix 1 void), expressed as the matrix density, was averaged for each specimen.

527

always $110% of systemic pressure (group B). In one patient, the pressure measured at different sites was equal to or less than systemic pressure (group C), but in 5 patients, measurements taken at different sites in the same aneurysm were sometimes .110% and sometimes ,90% of systemic pressure (group D). Mean pressure within the flow lumen of the aneurysm was always identical to mean radial artery pressure.

Ex Vivo Tests

Descriptive statistics were applied to analyze data from the in vivo study; results are expressed and the mean 6 standard deviation. The relationship between matrix density and pressure transmission in the ex vivo experiments was tested by regression analysis.

The 12 specimens of thrombus harvested from 6 patients and tested ex vivo showed marked variation in transmission of pressure both within and between specimens (Fig. 4). Microscopically, thrombus contained an abundance of stained amorphous material (matrix) interspersed with red blood cells and inflammatory cells (Fig. 5). Voids within each section contained neither cellular nor matrix material and were mostly likely filled with fluid that was either lost or remained transparent at histological examination. The matrix density of the same thrombus specimens (Fig. 6) ranged between 52% and 97%. There was a significant correlation between matrix density and pressure transmission across thrombus (R250.747, p50.001).

RESULTS

DISCUSSION

Statistical Analysis

In Vivo Pressure Distribution The mean intrasac pressure measured in the 6 controls without aneurysm thrombus was always identical to the mean systemic pressure, whereas the mean pressure within aneurysm thrombus (PAT) ranged between 37% and 207% (mean 106% 6 17.4%) of the mean systemic pressure in the 26 test patients (121 sites; Fig. 3). PAT was within 10% of simultaneously recorded systemic pressure at 77 of 121 sites. PAT was ,90% of systemic pressure at 8 sites; at a further 36 sites, it was .110% of the systemic pressure. The distribution of pressure also varied between patients. In 9 patients, PAT was always within 10% of systemic pressure (group A). In 11 patients, however, pressure measured at different sites within the same aneurysm was

Endovascular repair relies on the deployment of an intraluminal stent-graft to isolate the aneurysm from the circulation, thereby preventing rupture. The aneurysm remains in situ, however, and may again be at risk of rupture if it is no longer adequately isolated from aortic pressure at any time after operation or during follow-up.6,7 Persistent or recurrent pressurization of the aneurysm sac (endotension) can occur if the stent-graft is maldeployed, if it migrates, or if structural failure results in perforation or tear of the graft fabric.8 Patients must therefore undergo lifelong monitoring both to ensure the continued integrity of the stent-graft itself and to detect recurrent pressurization of the aneurysm before rupture occurs. Currently, surveillance relies on regular clinical examination, abdominal radiography,

528



Figure 3 l Mean pressure within aneurysm thrombus expressed as a percentage of mean systemic pressure. Each column of data points represents measurements obtained at different sites in each of the 26 patients (circled values in column 26 represent outliers). Group A: pressure measurements always within 10% of systemic pressure; group B: pressure measurements always $110% of systemic pressure; group C: pressure measurements equal to or less than systemic pressure; group D: measurements variable, sometimes .110% and sometimes ,90% of systemic pressure.

and duplex or CT scanning. None of these examinations, however, reliably reveal if the aneurysm is again pressurized. Pressurization may be inferred if there is evidence of a graftrelated endoleak or if the aneurysm continues to expand after treatment.9 However, pressurization may occur in the absence of endoleak, and there is now evidence that expansion can occur without significant pressurization.10 The need to monitor intrasac pressure after endovascular repair is widely accepted, and there is currently great interest in the possibility of intraprocedurally implanting sac pressure sensors that can subsequently be interrogated remotely either intermittently or continuously.4,5 Such sensors would measure the pressure within aneurysm thrombus around the sensor. The risk of aneurysm rupture, however, depends on the pressure adjacent to the aneurysm wall. The predictive value of measurements obtained by means of an implantable sensor would therefore rely on the assumption that pressure is uniformly

distributed throughout the aneurysm thrombus and that the pressure measured at any location within aneurysm thrombus is the same as the pressure adjacent to the aneu-

Figure 4 l Scatter plot of pressure transmission through thrombus; each data point represents a specimen of thrombus. Marked differences can be seen between and within patients.



529

Figures 5 l Histological sections through 3 different specimens of thrombus; arrows mark voids in the matrix background (hematoxylin-eosin, original magnification 325).

rysm wall. We tested this assumption in the clinical setting initially. For the in vivo measurement of pressure transmission through sac thrombus, we chose not to employ a catheter tip transducer system but measured pressure simply through a needle and pressure line. We realized that false readings could arise if the needle were blocked by aneurysm thrombus; we overcame this problem by incorporating a continuous microflush system in the circuit. Continuous delivery of minute volumes of saline would not normally affect the pressure recorded but would result in a steady climb in pressure if the needle became blocked. We were surprised to find significant vari-

Figure 6 l Correlation of pressure transmission through aneurysm thrombus with its matrix density (R250.747, p50.001).

ation in the pressure recorded within aneurysm thrombus both in different patients and within the same aneurysm. Since pressure appeared not to be uniformly distributed within aneurysm thrombus, further studies were performed to examine transmission of pressure across aneurysm thrombus ex vivo and to correlate the pressure transmission properties of thrombus with its microstructure. It is interesting to note that in a number of cases, the mean in vivo pressure within aneurysm thrombus was higher than simultaneously measured mean systemic pressure. When this occurred, we noted that the systolic pressure within aneurysm thrombus was usually close to systemic systolic pressure. The difference occurred because of a much higher diastolic pressure within thrombus. This phenomenon has been reported by others11 and may be explained by the semi-fluid consistency of thrombus, which could in effect trap pressure so that the diastolic pressure cannot fall to systemic diastolic levels. Further experiments are in progress to elucidate this intriguing phenomenon. Ex vivo analysis of pressure transmission through sac thrombus confirmed the variation noted in the in vivo studies. The pressure cell had been validated by examining pressure transmission through substances of known porosity, and it seemed very unlikely

530


that the pressure recorded in cell B was in any way influenced by leakage around the thrombus disk itself. Image analysis of the thrombus revealed a significant variation in matrix density that correlated with the variation in pressure transmission. In conclusion, we have demonstrated in clinical and ex vivo experiments that significant variation exists in the transmission of pressure through aneurysm thrombus. This observation raises the possibility that pressure measured at a single location by an intrasac pressure monitor may not accurately reflect the pressure applied to the aneurysm wall and may not therefore reliably predict the risk of AAA rupture. Acknowledgments: The authors would like to thank Deborah Hunt and Lyndsey Wood for their assistance in the histology laboratory and Gill Rycroft-Holmes for producing Figures 1 and 2.

REFERENCES 1. Harris PL, Dimitri S. Predicting failure of endovascular aneurysm repair [Editorial]. Eur J Vasc Endovasc Surg. 1999;17:1–2. 2. Gilling-Smith G, Brennan J, Harris P, et al. Endotension after endovascular aneurysm repair: definition, classification, and strategies for surveillance and intervention. J Endovasc Surg. 1999;6:305–307.


3. Baum RA, Carpenter JP, Cope C, et al. Aneurysm sac pressure measurements after endovascular repair of abdominal aortic aneurysms. J Vasc Surg. 2001;33:32–41. 4. Yadav JS. Is an implantable pressure-sensing device for monitoring intrasac pressures feasible? Paper presented at: 29th Annual VEITH Symposium; November 21–24, 2002; New York City. 5. Berger E. Microimplants may save lives one day. Available at: http://www.cnn.com/2002/ HEALTH/01/22/microchip.heart/ 6. Harris PL, Vallabhaneni SR, Desgranges P, et al. Incidence and risk factors of late rupture, conversion, and death after endovascular repair of infrarenal aortic aneurysms: the EUROSTAR experience. J Vasc Surg. 2000;32:739–749. 7. Zarins CK, White RA, Moll FL, et al. The AneuRx stent graft: four-year results and worldwide experience 2000. J Vasc Surg. 2001; 33:S135–145. 8. Veith FJ, Baum RA, Ohki T, et al. Nature and significance of endoleaks and endotension: summary of opinions expressed at an international conference. J Vasc Surg. 2002:35:1029– 1038. 9. White GH. What are the causes of endotension? J Endovasc Ther. 2001;8:454–456. 10. Risberg B, Delle M, Eriksson E, et al. Aneurysm sac hygroma: a cause of endotension. J Endovasc Ther. 2001;8:447–453. 11. Parodi JC, Berguer R, Ferreira LM, et al. Intraaneurysmal pressure after incomplete endovascular exclusion. J Vasc Surg. 2001;34:909– 914.