A Study of Pullout Forces of the Components of Modular Multi-manufacturer Hybrid Endografts Used for Aortic Aneurysm Repair

Article Outline

• Abstract
• Introduction
• Materials and Methods
  • Endografts and experimental setup
  • Test media
  • Test apparatus
  • Statistical analysis
• Results
  • Body with limb extensions
    • Non-hybrid stent-graft systems
    • Hybrid stent-graft systems
  • Limb-to-limb pullout forces
  • Helical IBG
• Discussion
  • Body-to-limb extensions
  • Limb-to-limb extensions
  • Fish mouthing
• Conclusion
• Conflict of Interest/Funding
• References
• Copyright

Abstract 

Objectives

Aorto-iliac angulations may be challenging for modular stent-graft systems (SGSs) from a single manufacturer. This study aims to define the pullout forces (POFs) of SGSs derived from the same (non-hybrid) or different manufacturers (hybrid).

Methods

The POFs were tested in a vertical position in air and 5% albumin. We studied the POFs between legs from Anaconda (Vascutek^®), Excluder (Gore^®), Talent (Medtronic^®) and Zenith (Cook^®) with the contralateral limb of bifurcated aortic bodies from Zenith (12mm), Anaconda and Excluder.

Results

For non-hybrid SGSs, the POFs decreased in the following order: Anaconda (11.2±0.6N), Talent (6.25±0.6N), Zenith (3.5±0.01N) and Excluder (2.5±0.5N). The Zenith body with the Anaconda limb (15mm) registered the greatest POF (13.083±0.821N); the Zenith and Excluder bodies combined with the Excluder limb (16mm) registered the weakest POFs (2.397±0.22N and 2.500±0.479N, respectively). The Zenith body combined with the Excluder limb (16mm) had a POF similar to the Zenith non-hybrid; combined with Talent 14mm and Anaconda limb exhibited POFs greater than the Zenith non-hybrid system. For the limb-to-limb POFs, the greatest was registered for the Anaconda limb, 13mm within a 12-mm extension for 40-mm overlaps (23.06±0.480N); the weakest POFs were recorded for the Excluder limbs at 30-mm overlaps (1.09±0.167N and 1.11±0.250N).

Conclusions

The hybrid SGSs performed as well as or better than the non-hybrid systems, and should be considered for clinical testing in patients whose unique anatomy warrants the flexibility that the use of hybrids provides.

Keywords: Endovascular aneurysm repair, Endografts, Endostents

Back to Article Outline

Introduction 

Endovascular aneurysm repair (EVAR) has become a standard of care, capturing an estimated 40–70% of all elective abdominal aortic aneurysm (AAA) interventions in the USA.¹, ², ³, ⁴, ⁵ The most common endograft design is modular, using a body and one or two limbs. This constitutes a non-hybrid stent-graft system (NHSGS), which is usually supplied by a single manufacturer, and which has undergone individual unit and full-system testing to achieve regulatory certification.⁶ To use any particular NHSGS, however, the patient's anatomy must fall within a certain set of parameters for safe use. Unfortunately, some individuals have anatomical features that prevent the use of EVAR with the currently available NHSGS. Applicability of EVAR is therefore still limited by the incompatibility of anatomical variables and the design characteristics of different types of endografts.⁷, ⁸ In particular, aorto-iliac angulations and iliac artery tortuosities may be challenging for some available NHSGS. To overcome these difficulties and take advantage of different product characteristics, a recent trend has been for clinical practitioners to assemble endografts from different manufactures to create an anatomically compatible hybrid stent-graft system (HSGS) (Figure 1, Figure 2).⁷ Literature suggests that combinations of this type are used in about 7.9% of cases,⁷, ⁹ but this is likely an underestimate.

• 
  • View Large Image
  • Download to PowerPoint

• Figure 1 

  3-D rendering of computerized tomography showing hostile iliac anatomy.

• 
  • View Large Image
  • Download to PowerPoint

• Figure 2 

  Intraoperative angiogram of the same patient in Fig. 1, in whom the aneurysm was treated with a hybrid stent-graft system made of a Zenith Cook body and an Anaconda Vascutek limb.

Manufacturers perform physical testing of NHSGS, following a protocol laid out by the International Organization for Standardization (ISO),¹⁰ which includes pullout forces (POFs), to submit for regulatory approval. There are limited studies addressing the stability of HSGS and, in particular, the mechanical properties of the overlapping section (‘docking zone’). This lack of information is particularly troubling in the light of concerns raised in the past few years by the US Food and Drug Administration (FDA) regarding problems that have surfaced from the routine use of drug-eluting and biliary stents for off-label indications.¹¹, ¹² The increase in the use and promotion of off-label drugs and mechanical devices is an issue that the FDA is becoming more and more concerned about. Thus the need to perform appropriate studies to evaluate off-label device safety and behaviour has become even more critical.

The most common long-term endoleaks requiring urgent re-intervention with stent grafts used for EVAR are graft migration (type 1) and limb separation (type 3).¹³, ¹⁴ This is thought to be caused by the repeated stress applied to the stent graft by the pulsatile blood flow,¹⁵ although with large sac aneurysms, pressure changes resulting from postoperative sac shrinkage can also cause stent-graft movement and separation of the modular components.¹⁶

Because endografts are poorly incorporated into the patient's vessels, they resist dislodgement primarily through mechanical forces at the landing zones and radial and frictional forces at the docking zones. We hypothesised that great variations exist among the NHSGSs from different manufacturers in their ability to withstand POFs at the docking zones and that some HSGS combinations can withstand greater forces.

Since no data are available regarding the physical properties of these HSGSs, the objective of this study is to report a standardized test protocol to study the POFs and to compare the results obtained for modular NHSGSs with those of HSGSs.

Back to Article Outline

Materials and Methods 

Endografts and experimental setup 

Based on commercial availability, stents from four companies were used in this study: Anaconda (Vascutek, Inchinnan, Renfrewshire, Scotland); Zenith and helical iliac branched graft (Cook Medical Inc., Bloomington, IN, USA); Excluder (W. L. Gore and Associates, Newark, DE, USA) and Talent (Medtronic Inc., Minneapolis, MN, USA).

We studied the POFs using limb extensions from Anaconda (Vascutek^®) (12, 13 and 15mm in diameter), Excluder (Gore^®) (16mm), Talent (Medtronic^®) (14mm) and Zenith (Cook^®) (12mm); a helical iliac branched graft (IBG) (Cook^®) and bifurcated aortic bodies from Excluder (Gore^®), Talent (Medtronic^®) and Zenith (Cook^®) (contralateral limb diameter 13, 14 and 12mm, respectively). We also used two different body diameters (28 and 30mm) for the Anaconda (Vascutek^®) system with contralateral limb diameters of 11.5 and 12.5mm, respectively.

For the bifurcated bodies, we studied the POFs between the contralateral limb and a limb extension using an overlap of 30mm. Based on availability, on clinical rationale and suggested combinations from manufacturers, we tested the non-hybrid and hybrid configurations as described in Table 1.

Table 1. Bifurcated body and limb extensions for non-hybrid and hybrid configurations tested.

Contralateral limb of bifurcated body (mm)	Limb (mm)
	Anaconda	Excluder	Talent	Zenith
Anaconda (11.5)	12	16	14	12
	13
	15
Anaconda (12.5)	13		14	12
	15
Excluder (13)	13	16
	15
Zenith (12)	12	16	14	12
	13
	15
Talent (14mm)	12	16	14	12
	13
	15

The POFs of limb extension to limb extension were studied with 30-mm and 40-mm overlaps for the combinations described in Table 2.

Table 2. Limb extension to limb extension tests.

We also tested the POFs between the 17-mm long (12mm in diameter) proximal stent of the helical IBG and the Anaconda limbs (13mm and 15mm) and the Zenith limb extension of 12mm. For these tests, the overlap was limited to 17mm.

In each case, the precise degree of overlap was established by transparency, using a back-lit table and a metal ruler and callipers (Fig. 3). All the POF measurements were conducted with the elements in a vertical position and repeated thrice in an environmental chamber at room temperature.

• 
  • View Large Image
  • Download to PowerPoint

• Figure 3 

  Technique of measurement of the overlap between the body and the extension of a stent-graft system.

Test media 

In this study, we hypothesised that the POFs are greater in liquid solution than in a dry environment. The ISO standards used a water bath, but we postulated that the POFs in water would not be representative of what occurs in vivo since whole blood has a relative viscosity three times greater than water. The coefficient of viscosity of water is 0.001Ns⁻¹m⁻² at 20°C, and of whole blood is 0.0027Ns⁻¹m⁻² at 37°C.¹⁷, ¹⁸ The viscosity of human albumin (5%) has been reported to be between 0.0035 and 0.0040Ns⁻¹m⁻².¹⁹ In view of the greater similarity between the viscosity of whole blood and a 5% albumin solution with an approximate sodium content of 145mEql⁻¹, we chose this medium for testing (Plasbumin-5, Talecris Biotherapeutics, Inc., Research Triangle Park, NC, USA), and also used air as the reference.

Test apparatus 

A custom apparatus comprising of a software and hardware was designed and constructed to ensure uniformity of testing and efficiency.

The frame, made from an acrylic polymer, supported a cylinder containing the albumin solution, into which the stents were submerged (Fig. 4). An anchor point at the bottom of the test vessel fixed the distal stent, and a wire attached to an electronic FT24 tension load cell (MSI sensors, Measurement Specialties, Inc., Hampton, VA, USA) held the proximal stent. The tension load cell, which recorded the measurements of force with a resolution of 0.04N, was connected to the spool of an electric motor used to apply variable forces to the stents.

• 
  • View Large Image
  • Download to PowerPoint

• Figure 4 

  Test apparatus: the frame supports a cylinder containing the albumin solution, into which the stent is submerged. An anchor point at the bottom of the cylinder fixes the distal stent, and the proximal stent is held by a wire attached to the electronic tension load cell. The latter is connected to the spool of the electric motor applying variable forces to the stents.

An electronic control and data acquisition system based on an Atmega168 micro-controller (Atmel Corporation, San Jose, CA, USA) regulated the force applied by the electric motor, while continuously recording the force values from the load cell. The electronic unit detected pullouts through digital signal processing and transmitted the recorded forces back to a computer via a USB connection. Custom-designed software recorded the maximum force experienced during a test into a database as the POF.

The proximal docking zone of the distal stent graft was deployed 30mm (or 40mm, for limb-to-limb tests; 17mm for helical IBG) into the distal docking zone of the proximal stent. Steel wires were affixed to the non-docking zone ends of each endograft in a manner that evenly distributed the force applied by the electric motor. The wire also provided an attachment point for anchoring.

Once the stents were loaded in the cylinder, the test fixture applied a constantly increasing force to the proximal stent, while the distal stent graft remained fixed to the bottom of the apparatus. The motor continued to increase the applied force until the proximal and distal stent grafts were disconnected. The software recorded the maximum force measured over the course of the test as the POF.

Statistical analysis 

Data were expressed as mean and standard deviation (unless specified). Continuous variables were compared by Student's t-test or analysis of variance (ANOVA) and post-hoc analyses with Tukey's correction using SPSS (SPSS v. 16 Inc., 2007, Chicago, IL, USA). P-values ≤0.05 were considered statistically significant. Graphical representation of data was done using SPSS and Microsoft Excel 2008 (Microsoft Corporation, 2007, Santa Rosa, CA, USA).

Back to Article Outline

Results 

When all the tests were considered, the POFs in albumin solutions were greater than in the air environment (8.494±5.485 vs. 7.906±4.236N, P=0.006).

Body with limb extensions 

The POFs in albumin solution and in the dry environment for NHSGS and HSGS combinations of body with limb extensions are described in Table 3, Table 4.

Table 3. Pullout forces in albumin solution for non-hybrid and hybrid combinations of body with limb extensions.

Body (contralateral limb diameter, mm)
Limb extensions (diameter, mm)	Talent (14)	Zenith (12)	Excluder (13)	Anaconda 28 (11.5)	Anaconda 30 (12.5)
Talent (14)	6.250±0.573a	6.360±0.139a	–	5.643±0.289	5.287±0.300
Zenith (12)	3.613±0.269b	3.493±0.098b	–	2.540±0.242cg	2.467±0.150c
Excluder (16)	2.033±0.277d	2.397±0.222d	2.500±0.479d	3.157±0.134g	–
Anaconda (12)	8.047±0.981	5.187±0.501e	–	5.077±0.647e	–
Anaconda (13)	9.63±0.352	8.060±0.250	1.490±0.066	11.19±0.606	4.720±0.383
Anaconda (15)	11.397±0.733	13.083±0.821f	4.910±0.259	12.957±0.443f	12.503±0.413f

Comparisons of all groups across columns and rows by ANOVA were all statistically significant (P<0.001). Tukey's adjusted post-hoc t-test was used for comparison between groups. The same symbol between two comparisons denotes absence of a statistical significant difference. All other comparisons were statistically significantly different (P≤0.05).

Table 4. Dry pullout forces for non-hybrid and hybrid combinations of body with limb extensions.

Body (contralateral limb diameter, mm)
Limb extensions (diameter, mm)	Talent (14)	Zenith (12)	Excluder (13)	Anaconda 28 (11.5)	Anaconda 30 (12.5)
Talent (14)	6.650±0.500	7.603±0.432g	–	6.609±0.166ag	6.029±0.149g
Zenith (12)	4.884±0.06	4.166±0.313	–	3.228±0.063b	2.386±0.041
Excluder (16)	2.281±0.123hi	2.393±0.105h	1.866±0.285ci	3.154±0.124b	–
Anaconda (12)	8.182±0.149	4.953±0.395	–	6.209±0.521a	–
Anaconda (13)	10.073±0.063	8.541±0.191	2.400±0.048c	11.894±0.686	4.815±0.212
Anaconda (15)	11.784±0.299d	13.095±0.133ef	5.063±0.417	13.7298±0.041e	12.322±0.432df

Comparisons of all groups across columns and rows by ANOVA were all statistically significant (P<0.001). Tukey's adjusted post-hoc t-test was used for comparison between groups. The same symbol between two comparisons denotes absence of a statistical difference. All other comparisons were statistically significantly different (P≤0.05).

Among the NHSGSs, the POFs were significantly greater in the Anaconda system compared with all other SGSs (P<0.001); greater for the Talent system compared with the Zenith and the Excluder systems (P<0.001) and greater for the Zenith system compared with the Excluder system, but the latter difference did not reach statistical significance (P=0.07) (Fig. 5).

• 
  • View Large Image
  • Download to PowerPoint

• Figure 5 

  Wet pullout force for non-hybrid body-to-limb extension modular endografts. Each of the Anaconda bodies (28 and 30) was tested with the recommended limb diameters of 13mm and 15mm, respectively.

The Zenith, Anaconda 28 and Anaconda 30 bodies combined with the Anaconda limb extension of 15mm registered the greatest POFs: 13.083±0.821, 12.957±0.443 and 12.503±0.413N, respectively, without statistical difference among them. The weakest POFs were recorded with Talent, Zenith and Excluder bodies combined with the Excluder limb extension 16 limb, 2.03±0.3, 2.397±0.22 and 2.500±0.479N, respectively, without statistical difference among them. When the Zenith limb extension (12mm) was combined with the Talent or Zenith body, there was no difference in the POFs (P=0.468), nor when the Zenith limb was combined with the Anaconda 28 and 30 bodies (P=0.683). The latter two combinations, however, exhibited greater POFs than the hybrid with the Talent and Zenith bodies.

The HSGS using an Anaconda 28 body with Excluder (16mm), Talent (14mm) or Zenith (12mm) exhibited POFs which were weaker than the Anaconda 28 non-hybrid system with a 13-mm diameter limb. The HSGS using an Excluder body with an Anaconda limb (15mm) had a POF more than twice that of the non-hybrid Excluder system. The Zenith body combined with an Excluder limb extension (16mm) had a POF similar to the Zenith non-hybrid; combined with the Talent (14mm) and Anaconda limb extensions (12, 13 and 15mm) exhibited POFs greater than the Zenith non-hybrid system (Fig. 6).

• 
  • View Large Image
  • Download to PowerPoint

• Figure 6 

  Wet pullout forces for hybrid body-to-limb extension modular endografts. The Anaconda body 28mm was tested with the recommended limb diameters of 13mm and 15mm, respectively.

Limb-to-limb pullout forces 

Limb-to-limb combinations were only tested for NHSGS devices and the results are summarised in Table 5, Table 6. The cumulative POFs of limb-to-limb extensions were greater for 40- than for 30-mm overlaps (P<0.01). When looking at the individual devices, the POFs of limb-to-limb extensions were also greater for 40- than for 30-mm overlaps, with the exception of the Excluder limb extension (P=0.9). The greatest POF was registered for the Anaconda limb extension 13mm within a 12-mm extension and for an overlap of 40mm (23.06±0.480N). The weakest were recorded for the Excluder limb extension at 30- and 40-mm overlaps (1.09±0.167N and 1.11±0.250N, respectively). When an Anaconda limb extension was inserted within another of smaller diameter (10 to 9mm; 11 to 10mm; 12 to 11mm and 12 to 12mm), the POFs increased predictably up to a certain diameter of oversizing, beyond which the POF decreased (13 to 12mm and 15 to 12mm) (Fig. 7, Table 5, Table 6).

Table 5. Pullout forces in albumin solution for limb extension to limb extension combinations for overlaps of 30mm and 40mm.

Limb extension (mm)	Overlap
	30mm	40mm
Anaconda (10)/Anaconda (9)	12.83±0.52	16.53±0.024
Anaconda (11)/Anaconda (10)	12.13±0.393	13.28±0.587c
Anaconda (12)/Anaconda (11)	13.98±0.271	17.52±0.606
Anaconda (12)/Anaconda (12)	13.58±0.643a	19.19±0.313
Anaconda (13)/Anaconda (12)	14.01±0.408a	23.06±0.480
Anaconda (15)/Anaconda (12)	9.19±0.345	13.18±0.235c
Excluder (16)/Excluder (16)	1.09±0.167§	1.11±0.250§
Talent (14)/Talent (14)	4.08±0.149b	5.52±0.172
Zenith (12)/Zenith (12)	4.30±0.443b	4.61±0.402

Comparisons of all groups across columns and rows by ANOVA were all statistically significant (P<0.001). Tukey's adjusted post-hoc t-test was used for comparison between groups. The same symbol between two comparisons denotes absence of a statistical significant difference. All other comparisons were statistically significantly different (P≤0.05).

Table 6. Dry pullout forces for limb extension to limb extension combinations for overlaps of 30mm and 40mm.

Limb extension (mm)	Overlap
	30mm	40mm
Anaconda (10)/Anaconda (9)	13.29±0.415	16.0344±0.908e
Anaconda (11)/Anaconda (10)	8.32±0.599	10.8318±0.313e
Anaconda (12)/Anaconda (11)	10.21±0.190b	13.5504±0.235gf
Anaconda (12)/Anaconda (12)	11.38±0.443cd	14.5716±0.373f
Anaconda (13)/Anaconda (12)	10.82±0.145bc	12.3084±0.701g
Anaconda (15)/Anaconda (12)	10.56±0.276bd	12.9432±0.361g
Excluder (16)/Excluder (16)	0.43±0.024	0.5646±0.072
Talent (14)/Talent (14)	3.66±0.313a	4.6356±0.291
Zenith (12)/Zenith (12)	3.81±0.425a	4.7598±0.157

Comparisons of all groups across columns and rows by ANOVA were all statistically significant (P<0.01). Tukey's adjusted post-hoc t-test was used for comparison between groups. The same symbol between two comparisons denotes absence of a statistically significant difference. All other comparisons were statistically significantly different (P≤0.05).

• 
  • View Large Image
  • Download to PowerPoint

• Figure 7 

  Pullout forces in albumin solution for the 30-mm and 40-mm overlap of each limb system.

Helical IBG 

The POFs between the helical IBG and the limb extensions were 4.78±0.24N for the Anaconda (13mm), 3.95±0.17N for the Anaconda (15mm) and 4.58±0.28N for the Zenith limb extension. The POFs were greater using Anaconda (13mm) vs. Anaconda (15mm) (P=0.005) and Anaconda (15mm) vs. Zenith (12mm) (P=0.017) (Fig. 8).

• 
  • View Large Image
  • Download to PowerPoint

• Figure 8 

  Pullout forces of the helical iliac branched graft with Anaconda limb extension 13mm, 15mm and Zenith 12mm limb extensions using a 17-mm overlap. Pullout forces were greater using Anaconda 13mm vs. Anaconda 15mm (P=0.005); and Anaconda 15mm vs. Zenith 12mm (P=0.017).

Back to Article Outline

Discussion 

We have shown that the POFs are greater in albumin solution than in a dry environment. In selecting a fluid as the media for POFs, we excluded water because blood has a much higher viscosity. We considered human blood, equine blood, equine albumin, porcine blood and porcine albumin, as well as an artificial blood solution.²⁰, ²¹, ²², ²³ Human or porcine blood would have been the ideal medium due to their similarities.²⁴, ²⁵ Availability and cost, however, were a concern. In addition, whole blood has several drawbacks: it coagulates once exposed to air, affecting the test environment and potentially damaging the stents; the use of anticoagulants may alter the physical properties of blood in unpredictable ways and finally the physical characteristics of human blood can vary from one sample to the next.²⁶ Because of these considerations and the fact that the primary protein of serum is albumin, we chose a 5% albumin solution.

Body-to-limb extensions 

There is a great variation in the POFs among different manufacturers: for body-to-limb (2–13N) and for limb-to-limb extensions (1–5N). For NHSGS, the POFs decrease progressively in the following order: Anaconda, Talent, Zenith and Excluder systems (Fig. 5). High POFs (greater than 6N) are recorded with hybrids of a Talent body with Excluder and Anaconda limbs or Zenith body with Anaconda limbs. Hybrids using an Anaconda body with Excluder or Talent limbs and an Excluder body with appropriately sized Anaconda limbs (15mm) have POFs lower than the former groups but in the range of or greater than the non-hybrid Zenith and Excluder. All hybrids of a Zenith body have high POFs except the one using an Excluder limb (Fig. 6). The greatest POFs were found for Zenith, Anaconda 28 and 30 bodies with Anaconda limbs (15mm) (12–13N). The weakest POFs were identified with Excluder and Zenith bodies combined with the Excluder limb (16mm) (1.8–2.5N).

Limb-to-limb extensions 

When considering limb-to-limb POFs, these were predictably greater for 40-mm overlaps compared with 30mm. An exception was the Excluder limbs which exhibited POFs which were the lowest of the group and not different for 30- and 40-mm overlaps. The Zenith and Talent limbs performed very similarly, and the Anaconda limbs displayed the greatest POFs. For Excluder, Zenith and Talent, an overlap of 30–40mm is probably advisable, while for Anaconda limbs where POFs are very high, shorter overlaps may be adequate, although this assumption was not tested. In addition, the overall effect of limb separation is the result of the direct POFs at the level of the ‘docking zones’ and of the distribution of the forces derived from the pulsatile blood flow over the entire length of the stent. Since the Anaconda limbs do not have longitudinal support, it is the docking zone that may bear the brunt of the pulsatile force, while stents with greater longitudinal strength (e.g., Talent and Zenith) need to resist less force at the docking zone, because more of that force might be absorbed by the distal elements of the structure.

Fish mouthing 

When an Anaconda limb was inserted within another of smaller diameter, the POF increased predictably up to a certain diameter of oversizing, beyond which the POF decreased. That the most extreme oversizing does not necessarily result in the highest POFs has been recognised in vivo.¹⁶, ²⁷ When the Anaconda limbs are greatly oversized, as with the 15mm in the 12mm, the rings do not fully return to their circular shape, acquiring a fish-mouth shape. We hypothesise that, in this setting, the rings do not interlock and the POF is reduced.

In only one reported instance,²⁸ the POFs were tested in a manner similar to ours, however, using water as a test medium. Two of the stents tested are not available for use any longer (the AneuRx and the Fortron); for the Zenith device they used a BiFab aorto-uni configuration with a limb overlap much greater that 30mm and predictably recorded average POFs greater than ours (7N vs. 5N); for the Talent stent graft they used a bifurcate configuration and the recommended overlap of 30mm. They found average POFs of 5.5N, somewhat lower than ours (7N) likely due to the different test medium (water vs. albumin). They did not however test any hybrid configuration. When manufacturers perform similar testing to submit for regulatory approval, they follow a protocol laid out by the ISO,¹⁰ but the results are not easily available for review.

We established a protocol for testing the POFs that includes four significant changes from earlier studies by others and from the ISO standards: (1) non-realistic test media (water, air, etc.) were replaced with a 5% human albumin solution, which more accurately replicates in vivo demands upon a stent graft once inserted; (2) all models were adjusted, using callipers and a back-lit light table, to a standardized docking zone overlap within the manufacturer's recommended range (generally 30mm, also 40mm for limb-to-limb tests; 17mm for Cook's helical IBG) prior to testing; (3) vertically oriented testing, with the SGS suspended in the test medium, simulates a sitting or standing position; (4) testing of HSGSs in addition to non-hybrid, single-manufacturer SGS systems. We were able to obtain results for ISO standard testing from Vascutek with respect to the body-to-limb POFs: for the body size 28 matched with a limb 13mm in diameter and a body size 30 matched with a limb 15mm in diameter (recommended manufacturer combinations) the POFs were 10.0±2.9 and 10.1±2.6N, respectively. For the same combinations in our testing protocol, the results were 11.2±0.6 and 12.5±0.4N, respectively. The difference was significantly different statistically (P<0.01). This difference is in keeping with the theoretical construct that the POFs in a more viscous liquid environment are greater than in water.

We identify two primary results of this research that should be taken into account when designing human clinical trials to evaluate the compatibility of stent-graft components in a hybrid system. First, care should be taken when considering the use of any HSGSs that have particularly low POFs in preliminary physical testing such as ours. Second, since HSGS in our tests performed as well as or better than the homogeneous systems, hybrid systems are definitely worth studying, especially with regard to choosing the optimal mix of stable HGS components to match unique patient anatomies.

The limitations of this study are that it is a physical study of POF with no clinical correlation with type 3 endoleaks; the peak POF was used instead of the area under the curve which is a measure of the overall energy of displacement; a limited number of possible combinations was investigated; and we did not conduct a study of the diameter of luminal encroachment. Type 3 endoleaks are the result of complex interactions between many different factors, some not yet identified, and many not well understood. On the basis of theoretical constructs, however, low POFs are one significant contributing factor.

Physical studies of endovascular stent-graft systems are used according to the ISO before regulatory approval for any device. The results of this study should not be considered an endorsement of or warning against any particular combination (hybrid or otherwise). However, since there are virtually no studies addressing the hybrid systems, we feel that our work is relevant for the researcher, clinician and regulatory bodies; the researcher who embarks on endovascular graft development and testing may want to consider our technique of physical testing, and consider the medium in which these tests are done, since currently the ISO uses water and the horizontal position for POFs, while perhaps a fluid more compatible with blood and a vertical position (as we have done) may be more appropriate; the clinician, who would like to use a hybrid system to better fit his patient's anatomy, may find himself asking the questions, ‘If I were to use a body supplied by one manufacturer and a limb of another, what is the overlap I should use?’ ‘What is the oversizing? How does my choice impact on the POFs compared with the non-hybrid system?’ The regulatory agencies may also consider asking manufacturers for different ways of testing the physical properties of modular SGSs.

Back to Article Outline

Conclusion 

Hybrid and non-hybrid SGSs exhibited POFs which were similar when appropriate oversizing was used. HSGS should be considered for clinical testing in patients whose unique anatomy warrants the flexibility that their use provides. When considering the use of HSGS in such trials, combinations with very low POFs should be balanced against the potentially increased risk of type 3 endoleaks.

Back to Article Outline

Conflict of Interest/Funding 

None.

Back to Article Outline

References 

1. Lee WA, Carter JW, Upchurch G, Seeger JM, Huber TS. Perioperative outcomes after open and endovascular repair of intact abdominal aortic aneurysms in the United States during 2001. J Vasc Surg. 2004;39:491–496
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (116 KB)
   • CrossRef
2. Anderson PL, Arons RR, Moskowitz AJ, Gelijns A, Magnell C, Faries PL, et al. A statewide experience with endovascular abdominal aortic aneurysm repair: rapid diffusion with excellent early results. J Vasc Surg. 2004;39:10–19
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (301 KB)
   • CrossRef
3. Dillavou ED, Muluk SC, Makaroun MS. Improving aneurysm-related outcomes: nationwide benefits of endovascular repair. J Vasc Surg. 2006;43:446–451
   • View In Article
4. Dillavou ED, Muluk SC, Makaroun MS. A decade of change in abdominal aortic aneurysm repair in the United States: have we improved outcomes equally between men and women?. J Vasc Surg. 2006;43:230–238
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (244 KB)
   • CrossRef
5. Brewster DC, Jones JE, Chung TK, LaMuraglia GM, Kwolek CJ, Watkins MT, et al. Long-term outcomes after endovascular abdominal aortic aneurysm repair: the first decade. Ann Surg. 2006;244:426–438
   • View In Article
   • MEDLINE
6. International Organization for Standardization. Endovascular devices—part 1: endovascular prostheses. Standards. ISO 25539-1; 2005.
   • View In Article
7. Jordan WD, Naslund TC, Adelman MA, Simoni G, Wirthlin DJ. Endovascular abdominal aortic aneurysm repair in patients with challenging anatomy: utility of the hybrid endograft. Vascular. 2004;12:106–113
   • View In Article
   • MEDLINE
   • CrossRef
8. Moise MA, Woo EY, Velazquez OC, Fairman RM, Golden MA, Mitchell ME, et al. Barriers to endovascular aortic aneurysm repair: past experience and implications for future device development. Vasc Endovascular Surg. 2006;40:197–203
   • View In Article
   • MEDLINE
   • CrossRef
9. Wirthlin DJ, Alcocer F, Whitley D, Jordan WD. Use of hybrid aortic stent grafts for endovascular repair of abdominal aortic aneurysms: indications and outcomes. J Surg Res. 2002;108:14–19
   • View In Article
   • Abstract
   • Full-Text PDF (215 KB)
   • CrossRef
10. ISO. Cardiovascular implants – endovascular devices – part 1: endovascular prostheses. Testing methods. 25539-1:2003/Amd 1; 2005.
    • View In Article
11. US Food and Drug Administration . Update to FDA statement on coronary drug-eluting stents. Center for Devices and Radiological Health; 2008;
    • View In Article
12. Bridges J, Maisel WH. Malfunctions and adverse events associated with off-label use of biliary stents in the peripheral vasculature. Am J Ther. 2008;15:12–18
    • View In Article
    • CrossRef
13. Verhoeven EL, Tielliu IF, Prins TR, Zeebregts CJ, van Andringa de Kempenaer MG, Cina CS, et al. Frequency and outcome of re-interventions after endovascular repair for abdominal aortic aneurysm: a prospective cohort study. Eur J Vasc Endovasc Surg. 2004;28:357–364
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (164 KB)
    • CrossRef
14. Buth J, Harris PL, Van Marrewijk C, Fransen G. The significance and management of different types of endoleaks. Semin Vasc Surg. 2003;16:95–102
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (122 KB)
    • CrossRef
15. Mohan IV, Laheij RJ, Harris PL. Risk factors for endoleak and the evidence for stent-graft oversizing in patients undergoing endovascular aneurysm repair. Eur J Vasc Endovasc Surg. 2001;21:344–349
    • View In Article
    • Abstract
    • Full-Text PDF (92 KB)
    • CrossRef
16. Howell BA, Kim T, Cheer A, Dwyer H, Saloner D, Chuter TA. Computational fluid dynamics within bifurcated abdominal aortic stent-grafts. J Endovasc Ther. 2007;14:138–143
    • View In Article
    • MEDLINE
    • CrossRef
17. Levy M, Koeppen B, Stanton B. Cardiovascular system. In: Berne and Levy principles of physiology. Elsevier Science; 2005;p. 235–241
    • View In Article
18. Guyton AC, Hall JE. Blood groups; transfusion; tissue and organ transplantation. In: Textbook of medical physiology. Philadelphia: Saunders Company; 2000;p. 355–361
    • View In Article
19. Jirgensons B. The intrinsic viscosity of serum albumin. Die Makromolekulare Chemie. 2003;16:192–197
    • View In Article
20. Wonder A. Blood dynamics. San Diego: Academic Press; 2001;
    • View In Article
21. Squires JE. Artificial blood. Science. 2002;295:1002–1005
    • View In Article
    • CrossRef
22. Stokstad E. Not blood simple. Science. 2002;295:1003
    • View In Article
    • CrossRef
23. Knock C, Davison M. Predicting the position of the source of blood stains for angled impacts. J Forensic Sci. 2007;52:1044–1049
    • View In Article
    • CrossRef
24. Woodcock JP. Physical properties of blood and their influence on blood-flow measurement. Rep Prog Phys. 1976;39:127–132
    • View In Article
25. James SH, Kish PE, Sutton TP. Biological and physical properties of human blood. Principles of bloodstain pattern analysis. CRC Press; 2005;
    • View In Article
26. Kanai H, Fujii M, Niwa M, Sakamoto K. Physical properties of flowing blood. In: Proceedings of the annual international conference of the IEEE. Vol. 14. Engineering in Medicine and Biology Society; 1992. p. 804–805.
    • View In Article
27. Sternbergh WC, Money SR, Greenberg RK, Chuter TA. Influence of endograft oversizing on device migration, endoleak, aneurysm shrinkage, and aortic neck dilation: results from the Zenith multicenter trial. J Vasc Surg. 2004;39:20–26
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (147 KB)
    • CrossRef
28. Hinchliffe RJ, Natarajan S, Hopkinson BR. In vitro analysis of modular aortic stent-graft failure. J Endovasc Ther. 2006;13:77–84
    • View In Article
    • MEDLINE
    • CrossRef