design

serving european medical device and in vitro diagnostic product manufacturers

Simulated Testing in
Medical Device Design
S. Kenny and E. McDermott
Creganna Medical Devices, Galway, Ireland

Image: Eyewire Images

Bench-top testing of a typical catheter is described here to show how early
prototype evaluation can decrease product development time. Trackability
and pushability are assessed and compared using some novel laboratory tests,
which can be applied to a variety of medical devices.

The test sequence for creating a sales message in the field. ■ Phase 4: Clinical application (unlim-
Developers of new medical devices The objective of the testing will ited availability and further trials).
are keen to perform animal testing determine the approach that is taken. Industry usually performs Phases
early in the product-development Unsophisticated test setups are 1 and 2 without clinical input, hence
process. Yet, a properly constructed adequate early in the design cycle to the accuracy of the results from
laboratory-based test sequence will confirm that the device will perform these Phases is essential. These Phases
produce convenient, reproducible and its basic functions. At a later stage in encompass testing of all materials and
cost-effective results. This approach the process, the design can be tested prototypical devices prior to testing or
facilitates a better understanding of against relevant standards or the prod- use in humans. Guidance in designing
how devices will perform in vivo and uct specifications. The evaluation of test plans for prototypical medical
permits product designers to appraise new designs usually follows a general devices may be obtained from a
early prototype designs in the labora- sequence such as number of sources. However, most
tory. Additional benefits include an ■ Phase 1: Bench testing guidance documents are relatively
accelerated product-development ■ Phase 2: Preclinical animal studies nonspecific in nature and require the
timescale and lower overall develop- ■ Phase 3: Clinical evaluation application of sound scientific method
ment costs. (restricted availability) to the specific device in question.
Medical devices are tested against
a range of requirements at various Table I: Established tests used in the evaluation of new designs.
stages in their life cycle. Numerous Balloons Stents Catheters/Delivery systems
groups need to be satisfied by this
Bond (σTS/σC/σP) Bend/σF strength Bond (σTS/σC/σP) strength
device testing at different phases strength
throughout the design process, for
Diameter and profile Radial strength Bend/ σF strength
example:
Pressure testing Diameter and profile Diameter and profile
■ Feasibility testing following com-
pletion of the initial design phase; Inflation and deflation Stent crimping Contrast media flow rate
time
extensive testing is performed on
Fatigue testing Crossing profile
the prototype to ensure the required
product performance can be met. Tip pulling and torquing
■ Statistically valid testing is required Trackability and pushability
for regulatory approval. Kink resistance
■ Performance and comparative test-
ing for marketing groups to illustrate Deployment force
ease of use, performance enhance-
ments and superiority over competi- σTS = Tensile strength, σC = Compressive strength, σP = Peel strength, σF = Flexural strength.
tive products, data that are essential
 design

Devices should be tested in animals of known failure modes (risk assess- Figure 1: (a) One of the peripheral models used, which can
to evaluate performance characteristics ment) is employed by most engineers be adapted to the anatomical relationships of different vascular
only after changes in material and to identify the sequence in which regions. The red line represents the path taken during the test.
design have been completed during tests will be performed. Device (b) Custom-made trackability testing equipment.
product development. Hence, Phase 1 models chosen for testing should (a)
testing must involve a large number encompass the range of ideal charac-
of mechanical and chemical tests. teristics to be achieved and consider,
Some of the most established mech- for example, the most tortuous vessel
anical test methods are listed in Table I. pathways and environmental factors.

Moving to Phase 2 testing Trackability test (b)

Many medical device manufacturers Measurement of a stent delivery sys-
are keen to conduct initial animal tem’s trackability allows the designer
testing prior to the completion of a to obtain measured force values for
full range of bench testing. Although prominent points in the 2D model Moveable
animal models offer an in vivo envi- via the charted curves. The forces crosshead
ronment, their anatomy and physiol- obtained allow an assessment of typi-
ogy can differ significantly from that cal tracking qualities and enable com-
of a human. As well as differences in parison of competitive systems. The
structural anatomy, the size of, and higher the trackability force, the more Clamp
Waterbath holding
accessibility to the site requiring force the physician will have to exert at 37˚c introducer
intervention must also be compatible to track the device to the location
with clinical device function and use requiring treatment. This would result Introducer
2D arterial
in man. For example, the size of the in a higher probability of a perfora- model
access (femoral) artery for percutane- tion in the vessel. Figure 1 (a) is an
ous vascular devices must be similar illustration of a typical 2D model. This Figure 2: (a) Trackability graphs; (b) pushability values obtained
to that of man, which is approxi- model is used in the trackability for sample A and sample B.
mately 8–10 mm outer diameter testing of stent delivery systems.
(o.d.). Few common laboratory The route chosen can be changed to Displacement (mm)
animals have a femoral artery this suit the desired tortuous path being (a)
large.1 Hence, selection of an animal replicated. The test was performed 0 15 30 45 60 75 90 105 120 135 150 165 180 195
0
model for preclinical medical device using custom-made equipment as
–0.5
testing is a complex and often dif- shown in Figure 1 (b). The water
–1
ficult task. It can also be costly, time bath lies in a horizontal position and
Force (N)

consuming and require dedicated the crosshead has mobility along it. –1.5

facilities. Furthermore, it is difficult to After the guidewire is fed into the –2

compare many characteristic qualities model, the catheter is threaded up –2.5
because the measurement conditions the guidewire and advanced through –3
are not standardised.2,3 Consequently, the chosen pathway. Silicone tubing –3.5
the use of bench-top equipment that is normally used inside the 2D model –4
can replicate the anatomy and in vivo to reproduce as closely as possible the
environment in question, facilitates dimensions and texture of the vessel.
quick, cost-effective, repeatable and The red line marked in Figure 1 (a)
comparable results. follows the path taken during a (b)
particular trackability test. The black 70
Simulated 2D Phase 1 testing arrow indicates the point at which the
60
Two-dimensional (2D) tortuous test commenced. The parts underwent
Pushability (%)

vessel pathways have been used in 60° and 90° bends during the track- 50
the past to great effect during the ability test. 40
development phase of a large number 30
of devices. For the purposes of illus- Test results Sample A
20
tration in this discussion, bench-top Various samples were tested using this Sample B
testing will focus on the testing of a path. Comparisons of some sample 10
typical catheter; however, similar test- results are presented in Figure 2. The 0
ing can be applied to practically any test is sensitive enough to recognise a
medical device. The application 7% reduction in o.d. between Sample A ➔
design

Figure 3: (a) Trackability and (b) pushability results obtained ➔ and Sample B. It is clear from Figure when considering the pros and cons
during benchmark testing of stent delivery system A and stent 2 (a) that the trackability force has of a device with a physician. The
delivery system B. decreased with a reduction in o.d. of trackability and pushability results for
the sample. The results of pushability two competitive stent delivery systems
Displacement (mm) tests conducted on the same samples are shown in Figure 3, where the
(a)
0 50 100 150 200 250 300 along the same path are shown in model in Figure 1 (a) was used. The
0
Figure 2 (b). A 6% reduction in push- systems were pushed and released
–0.2
–0.4 ability was recorded when the o.d. every 50 mm for 300 mm during
Force (N)

–0.6 was reduced by 7%. the trackability test to imitate the

–0.8 Benchmarking a new or existing physician’s insertion of the system
–1 product against an accepted market during a procedure, hence the vertical
–1.2
leader is a common practice in the lines present in Figure 3 (a). When
–1.4
–1.6 medical device industry. Trackability comparing the results, it is clear that
–1.8 and pushability are frequently used although delivery system A and delivery

Figure 5: (a) Overview of the 3D vascular track model in which the black line indicates
(b) the path taken during trackability testing; (b) trackability curves obtained for stent deliv-
60 ery system B using 3D model show repeatability of test; (c) comparison of trackability
curves obtained for delivery system B using the 2D and 3D models.
50
Pushability (%)

40
30
System A
20
System B
10
0

Figure 4: (a) Coronary block model; (b) trackability test

(a)
conducted on sample C and sample D.

(a)
Displacement (mm)
0 50 100 150 200 250
0

–0.2

–0.4
Force (N)

–0.6

–0.8

(b) –1

Displacement (mm)
0 50 100 150 200 250
(b) Displacement (mm)
105
120
135
150
165
185
195
15
30
45
60
75
90
0

0 –0.3
Force (N)

–0.5
Force (N)

–0.8
–1

–1 Sample C –1.3 2D model

Sample D 3D model
–1.5
(c) –1.8
 design

system B have similar pushability ated to ascertain whether or not it Figure 6: (a) The black line indicates the path taken during
characteristics, system A requires could distinguish minor variations testing of Tip A and Tip B; (b) trackability curves obtained for Tip
less force to track through the same in stent delivery systems. Two stent A and Tip B; (c) a close-up of the maximum tip force recorded.
pathway. delivery systems with different tips
Similar investigations were per- were tracked through the model and
formed to measure the effect that sur- the forces were recorded. The path
face treatments had on the trackability taken during the execution of this
of a stent delivery system. Figure 4 (a) investigation is shown in Figure 6 (a).
illustrates the 2D coronary model At approximately 195 mm, the device
used in this test. The outer surface of emerges from the guide catheter,
Samples C and D, which were made which results in an increase in track
of the same base catheter, were treated force. The system with Tip A experi-
with similar coatings. Sample C was ences greater track forces (1.42 N)
coated with a lubricious, hydrophilic when it emerges from the guide
coating and Sample D had a slightly catheter, as shown in Figure 6 (c),
different, covalently crosslinked, compared with the system with Tip B
hydrophilic polymer coating. As can (1.10 N). These results suggest that (a)
be seen in Figure 4 (b), Sample C the model set-up is sensitive enough
experienced greater frictional force to differentiate between minor differ- 0 50 100 150 200
during the tracking process compared ences in stent delivery systems. -0.1
with the results obtained with Sample D. -0.3

Force (N)
-0.5
Conclusion -0.7
Simulated 3D Phase 1 Testing Although trackability testing is not -0.9
The next generation of testing has a requirement by the United States -1.1 Tip A
-1.3
evolved with a three-dimensional Food and Drug Administration or Tip B
-1.5
(3D) model of the vascular system. International Organisation for Stand- (b) (c)
A glycerine–H2O composition with ardisation standards, it is becoming
equivalent viscosity to blood4 is an increasingly important perform-
circulated through the model using ance measure. Most medical device 2. J.F. Dyet et al., “Mechanical Properties of
a pulsating pump (Figure 5). A pulse organisations evaluate this feature Metallic Stents: How Do These Properties
rate of 70 beats/min is employed and use it as an input to develop- Influence the Choice of Stent for Specific
Lesions?” Cardiovasc. Intervent. Radiol., 23, pp.
during use. The guidewire is tracked ment. This is so common that the 47–54 (2000).
to the intended location, followed by medical establishment has listed it 3. R. Schmiedl and M. Schaldach, “X-Ray
a guide catheter. The force required to as a parameter it needs to know.5–6 Imaging of Coronary Stents,” Prog. Biomed.
track the stent delivery system under Most consultant cardiologists judge Res., 5, pp. 184–196 (2000).
investigation along the guidewire is the performance of the devices they 4. W.L. Lim et al., “Particle Image Velocimetry
in the Investigation of Flow Past Artificial
subsequently recorded. use on their ability to track to the site Heart Valves,” Ann. Biomed. Eng., 22, pp.
Delivery system B employed in location: 307–318 (1994).
Figure 3 was tracked through the “It provides excellent trackability 5. H.C. Tan and Y.T. Lim, “What You Need To
3D model using the path indicated … is suited for both proximal and Know, Coronary Stenting – What’s new on
in Figure 5 (a). Trackability curves distal lesions … This is among the the Horizon?” Singapore Med. J., 40, 06
(1999).
presented in Figure 5 (b) indicate that most popularly used stents in Europe, 6. R. Balcon, “Recommendations on Stent
the test generates repeatable results. because of its remarkable trackability.”6 Manufacture, Implantation and
The 3D model is a replica of the 2D Therefore, by allowing designers to Utilisation,” Study Group of the Working
model used to test delivery system B evaluate early prototype designs real- Group on Coronary Circulation, Eur Heart J.,
in Figure 3. Both results are compared istically in the laboratory, trackability 18, pp. 1536–1547 (1997). mdt
in Figure 5 (c). It is clear that the testing can help decrease product
force required to track delivery development time, while providing Sinéad Kenny, Ph.D.
system B through the 2D model is functional comparative data. is a Design and Development Engineer and
greater than that recorded using the Eamonn McDermott, BSc.
3D model. From past experience, this References is a Design and Development Engineer at
dissimilarity reflects the variation in 1. T. McCarthy, “Animal Models in Medical Creganna Medical Devices, Parkmore West,
Device Development and Qualification,”
results when comparing 2D model Galway, Ireland, tel. +353 91 757 801,
CRL reference paper, 10, 2 (1997)
results with those obtained from www.criver.com/flex_content_area/ e-mail: [email protected]
animal trials. documents/rm_rm_r_animal_models_ www.creganna.com
The 3D model was further evalu- med_device.pdf.

Reprinted from Medical Device Technology, May 2006 • Copyright © 2006 Canon Communications LLC