A method to determine the kink resistance of stents and stent delivery systems according to international standards

1 Introduction and background

Vascular stents must resist not only the radial load caused by the vessel wall and the blood pressure, but also additional mechanical loads from the surrounding depending on the position in the body. Coronary stents, e.g. implanted into the right coronary artery (RCA), are stressed because of the excessive movement of the RCA and repetitive kinking during the cardiac cycle [1]. In peripheral arteries like the carotid artery or the superficial femoral artery high stress of the implanted stents result from large deformations of the vessel during body motion as well as high tortuosities of the vessels [1], [2].

To reduce the risk of in-stent restenosis caused by a stent indicated lumen loss of the vessel the kink behavior of stents is of particular interest for clinicians, stent manufacturers and regulatory.

This study presents methods to determine the kink resistance of balloon expandable as well as self-expandable stents and of stent delivery systems according to international standards.

2 Requirements of international standards

The FDA guidance document no. 1545 recommends determining the kink resistance of stents for peripheral indications, where bending of the implant during normal body motion is expected. The smallest radius of curvature that the stent can withstand without kinking should be determined and the stent recovery of its original size and shape should be demonstrated. Secondly it is recommended to demonstrate that the stent delivery system will not kink at an appropriate bending radius [3].

The ISO 25539-2 standard claims for the smallest radius of curvature that the stent can adapt without kinking or a lumen loss of more than 50%. Furthermore the recovery of the stent after bending has to be recognized. In the appendix of the standard test methods are described in more detail, applicable for both balloon expandable and self-expandable stents, respectively. The two test methods described differ in the used test equipment: The stent within a silicone tube is bent around different mandrels (test method A) and the stent is released in a rigid model with bend (test method B). For both methods the minimum bending radius, as well as a kinking or a significant lumen reduction should be documented. The bending radius is reduced until kinking or a significant lumen loss could be determined. After the bent state is documented, the stent has to be examined without external load to determine the stent recovery [4].

In addition the ISO standard claims to test the bendability of the stent delivery system with respect to a clinical relevant curvature [4].

3 Material and methods

For all investigations each a commercially available balloon expandable stent 2.25 × 30 mm, a self-expandable stent 8 × 40 mm and stent delivery system were used.

3.1 Kink resistance of expanded stents

For determination of the kink resistance of the stents a special test setup inside a 37°C water bath was used (see Figure 1). The test setup consists of a mandrel with diameters from 5 to 15 mm in 2.5 mm steps and from 15 to 65 mm in 5 mm steps. As described for test method A in the ISO 25539-2, D.5.3.6.5, a silicone tube was used as implantation vessel [4].

Figure 1

Test setup for determination of kink resistance of stents.

Figure 2

Silicone tube with implanted stent: In original configuration (left) and in bent configuration (right).

The balloon expandable stent as well as the self-expanding stent were implanted into a silicone tube of appropriate dimensions (vessel diameter as intended for each stent). The silicone tube with implanted stent can be clamped inside the test setup and a force can be applied at both ends by a spring to guarantee a 180° adaption of the tubing to the mandrel. Outside the water bath a digital camera (CANON EOS 350D) was mounted on a tripod to generate the images for visual investigation and measurements of the kink behavior of the stents. The camera position was not changed during the whole test. The maximum magnification was used for every photograph.

3.1.1 Reference measurement

To determine the diameter of the stent inside the silicone tube in original state, serving as reference diameter, the silicone tube with implanted stent was clamped at the upper clamp. The silicone tube with stent was positioned central on the largest diameter of the mandrel. The tube has to be free from air bubbles and remained for 1 min inside the water bath to allow warming up. This configuration was documented with the digital camera.

The reference diameter of the examined stent d_stent was determined with an image editor (GIMP, GNU Image Manipulation Program) by counting the number of pixels in the image of the stent diameter (d_stentPx) as well as of the largest mandrel diameter to be used for the test (d_mandrelPx) and then using eq. 1.

(1)dstent=dstentPx×dmandreldmandrelPx

For further calculations this reference diameter is herein after referred to as d₀.

3.1.2 Determination of lumen loss/kinking during bending

To determine the lumen loss of the stent during bending the silicone tube with stent was clamped at the upper and lower clamps. The stent hast to be aligned symmetrical with respect to the clamps and the mandrel. The spring force was adjusted in a way that the silicone tube was adapted to the mandrel at least 180°, but did not flatten out. This configuration was then documented with the digital camera. The stent diameter in bend state was determined according to eq. 1 by using the known diameter of the mandrel as absolute dimension.

The lumen loss Δd_s of the examined stent while bent around the respective radius (=radius of the mandrel + wall thickness of the used silicone tube) was calculated according eq. 2, with d₀ as reference stent diameter in original state and d_S as stent diameter in bend state.

(2)Δ⁢dS=d0−dSd0×100%

Measurements of elastic recovery were conducted after every bending step.

3.1.3 Determination of the residual diameter change after bending

After each bending step of this test method the residual stent diameter was measured. Therefore the silicone tube was released from the lower clamp to straighten up the stent. The upper clamp was kept close. This configuration of the released stent within the silicone tube was documented with the digital camera.

The stent diameter in released state was determined according to eq. 1 as described above. The residual diameter change Δd_res was calculated as difference between the reference stent diameter d₀ and the diameter of the stent after bending d₁ and then related to the reference stent diameter. The stent recovery is given in percent (eq. 3).

(3)Δ⁢dres=d0−d1d0×100%

3.2 Kink resistance of stent delivery systems

The test to determine the kink resistance of the stent delivery system concentrates on the distal tube of the stent system since this is the part of the catheter which is intended to be exposed to the most challenging curvatures during stent application.

Before the measurement of the kink radius the stent system was stored for 1 min in 37°C heated water. The test was also performed in the 37°C heated water environment. No guide wire was used to realize worst case conditions. The measurement was conducted by bending the distal tube of the test specimen around a mandrel with a given radius (Figure 3).

Figure 3

Distal part of the stent delivery system bent around mandrel for determination of kink resistance.

The photographic documentation was done with the help of an incident light microscope (Olympus SZX16) and the integrated digital camera (Olympus UC30). The test was repeated with different mandrel radii. The smallest radius without kinking was documented.

4 Results

The results of determining the kink resistance of the balloon expandable stent and the self-expandable stent are displayed in Figures 4 and 5.

Figure 4

Lumen loss and residual diameter change of a balloon expandable stent dependent on the bending radius.

Figure 5

Lumen loss and residual diameter change of a self-expanding stent dependent on the bending radius.

Kinking or more than 50% lumen loss occurred at a bending radius of 3.5 mm and 21.0 mm for the balloon expandable stent or self-expanding stent, respectively. The residual diameter change after diameter reduction of more than 50% was about 42% for the balloon expandable stent and about 2% for the self-expandable stent.

The stent delivery system showed no kinking down to a bending radius of 2 mm.

5 Conclusion

In this study two methods were presented to determine the kinking resistance of stents and stent delivery systems according to international standards.

With the described methods it is possible to determine the lumen loss and the residual diameter change for every bending radius, so that the minimum curvature for each stent can be derived where the lumen loss is smaller than 50%. Methods are applicable for balloon expandable stents as well as for self-expanding stents.

Both, the lumen loss as well as the residual diameter change depend on the stent diameter, the stent design and of course the stent material. The residual diameter change of the balloon expandable stent increases with decreasing bending radius. At the smallest radius where the lumen loss was smaller than 50% (31% at 21 mm bending radius) the residual diameter change was already 22%. Plastic deformation is expected to occur at balloon expandable stents because it is mandatory for stent expansion.

Even the super elastic self-expandable nitinol stent showed small residual diameter changes, which refer to the interaction or restrictions of the stent and the silicone tube.

The presented method is limited in a way that it is not possible to create an exact 50% lumen loss of the tested stent to determine the exact minimum bending radius, because of the used stepped mandrel. The use of further steps to gain accuracy of the minimum bending radius would be possible but would complicate the test setup and reduce the universal usability for small and large stents.

Measurement accuracy depends on the camera setup used. It was in the range of ±0.07 mm (±2 Px), and thus well acceptable compared to the bending radii of inspected stents.

Transparency of the silicone tube is mandatory. The use of refractive index adjusted fluids could potentially gain accuracy [5].

The use of a rigid and curved vessel model (ISO method B) might also be appropriate. However, the presented method can use one stent for measurement of kink radius by bending it multiple times to decreasing radii. It appears to be an economic approach providing all necessary information.