Selected Technique

1Biomedicasity, 2-1 Seiryo

2InstituteAoba, Sendai

CorrespondResearch Orga980-8575, Jap

Ann Vasc SurgDOI: 10.1016/� Annals of VPublished onli

Application of Shape Memory AlloyPressure-Controlled Vascular Clampfor Atraumatic Vessel Occlusion

Ye Zhang,1 Yun Luo,1 Shingo Kodaira,2 and Toshiyuki Takagi,2 Sendai, Japan

Background: To realize atraumatic vessel occlusion, a new hemostatic clamp using superelas-tic shape memory alloy (SMA) as a pressure control limiter has been proposed. It was designedtaking advantage of a unique mechanical property of SMA. The ability to control pressure withthe newly designed SMA clamp was investigated. The traumatic effect on vessel walls was eval-uated in order to confirm the SMA clamp’s biological effect of protecting vessels from damage.Methods: Twenty-four pig thoracic aortas were divided into four groups: SMA group 1 (0.3 mmdiameter SMA wire used), SMA group 2 (0.4 mm diameter SMA wire used), base model group(hemostatic clamp of Mimura type), and a control group. The biomechanical interaction betweenthe pressure-controlled clamp and animal aortas was evaluated with a micropressure analyzingsystem. The atraumatic effect of the SMA clamp was examined on pig thoracic aortas in vitro andcompared with that of its base model. The morphological injury of each vessel was evaluatedafter being clamped for 15 min.Results: Pressure saturation was shown in the displacementepressure curve of the SMAclamp. In both SMA group 1 and SMA group 2, except for a slight imprint of compression inthe intima, no obvious injury was observed, while in the base model group the endothelial lacer-ation was observed when the clamp was closed to notch I. More serious injuries in the endothe-lial intima and media were observed when the clamp was closed to notches II and III.Conclusion: The effect of pressure-controlling SMA clamps on the in vitro vessel model wasevaluated. Histological observation and the traumatic score proved that the safety of the clampswas improved with the new design of pressure control. This technique supplies an effective andapplicable way of realizing atraumatic clamping.

INTRODUCTION

The clamp technique is essential in any surgical

operation.1 Clamps are used to arrest bleeding and

to control blood flow from the arteries. They are

also used to ensure a clear surgical view through

temporary local vascular control in order to favor-

ably accomplish operations. This is achieved by

l Engineering Research Organization, Tohoku Univer-, Aoba, Sendai 980-8575, Japan.

of Fluid Science, Tohoku University, 2-2-1 Katahira,980-8577, Japan.

ence to: Ye Zhang, MD, Biomedical Engineeringnization, Tohoku University, 2-1 Seiryo, Aoba, Sendaian, E-mail: zhang@tubero.tohoku.ac.jp

2009; 23: 813-820j.avsg.2009.06.001ascular Surgery Inc.ne: September 11, 2009

either cross- or lateral clamping.2 However, clamps

may also cause moderate to severe damage to the

vessel wall. Injuries caused by incorrect use of

clamps have been reported in several articles.3,4

Coelh et al.,4 using arteriographic and ultrasonic

evaluation, showed that the proportion of intimal

tears and flaps will reach 26% in some cases.

Atraumatic technique is an important issue in

surgery. Since Guthrie suggested its importance in

1903, many kinds of atraumatic clamps, such as Sat-

insky and De Bakey, have been invented.5 Although

these clamps have been widely used, they could not

fully meet the demand of the atraumatic technique.

Dujovny et al.6 demonstrated that these ‘‘atrau-

matic clamps’’ may also cause considerable endo-

thelial damage, which is proportional to the

‘‘closure force’’ applied. Injuries of vessels could be

minimized when force was controlled by closing to

813

Fig. 1. The newly designed pressure-controlled clamp

and SMA wires. The SMA wire was embedded in one

handling branch of the clamp with an additional fulcrum.

The wire and fulcrum formed a pressure-control

structure.

814 Zhang et al. Annals of Vascular Surgery

the minimal occlusive force (MOF),6,7 which refers

to the force applied that barely occludes the vessel.

Many improvements on pressure controlling have

proved that effective control of clamping pressure can

improve safety. For example, Darcin et al.2 invented

an atraumatic occluding clamp using a percutaneous

transluminal coronary angioplasty balloon combined

with a real-time pressure monitor. Other methods,

such as the use of silastic rubber,5,8 have also been

proposed; however, because of complication or

imprecision, those methods could not be practically

used or generally accepted.

Lack of an effective and consistent method for

pressure control impedes the improvement of

clamps. To create an atraumatic clamp that reduces

vessel injuries, the design can usually be improved

by two aspects of the clamp: the structure and the

material. Improvement on the structure has been

widely adopted, such as changes in the architecture

of the jaw face and the clamp geometry.9

On the other hand, the basic structure of the

clamp, a lever, allows the displacement of the

handle and the pressure of the jaw to show an expo-

nential relation when using general materials.10

With this basic structure, controlling pressure

becomes quite difficult. Because of the restrictions

of the material, it is not easy using only improve-

ments in the structure to realize an ‘‘atraumatic’’

technique. Selecting a new material which facili-

tates pressure controlling may be helpful to improve

the atraumatic ability of clamps.

Shape memory alloy (SMA) is an excellent bioma-

terial thathasbeenwidely used inclinical settings.11,12

SMA has the advantage of stress saturation.13 In order

to improve clamping safety, Luo et al13 proposed

taking advantage of this property to control the

closure pressure of clamps. The proposal is to embed

superelastic SMA wire into conventional hemostatic

clamps. Such a design, based on the unique mechan-

ical property of SMA, will limit the pressure between

the clamping jaws during its stress-induced transfor-

mation. The stress reaction of SMA wire has been

analyzed in our previous study.13 In this article, the

ability to control pressure with the newly designed

SMA clamp was investigated using an in vitro vessel

model. The traumatic effect on vessel walls was eval-

uated in order to confirm the SMA clamp’s biological

effect of protecting vessels from damage.

MATERIALS AND METHODS

SMA Pressure-Controlled Clamp

The basic concept of the new hemostatic clamps is to

ensure that the clamping pressure remains within

a safe range through an additional limiter made of

the superelastic wire of SMA. In addition to supere-

lastic and shape memory, some kinds of SMA mate-

rial have another phenomenon: With increasing

force, those SMAs reveal stress-induced transforma-

tion, during which the material shows a constant

stress for further extension.13,14 This is called stress

saturation.13

In the new design (Fig. 1), the SMA wire serves as

a stress limiter. It is embedded in one handling branch

of the clamp with an additional fulcrum. Thus, when

the clamp is quickly closed, the SMA wire is elon-

gated. With the transformation, the SMA wire is

elongated to cross a critical point. After reaching the

critical point, stress saturation will be present and

the pressure remains constant. Further displacement

of the handling rings does not increase the pressure.

The improvement was realized on a conventional

base model of the clamp. The base model (Mimura

type; Mizuho Ikoh, Tokyo, Japan) has the following

dimensions: 50 mm in length of jaws and 180 mm in

total length. The additional fulcrum was set to

95 mm in distance from the tip of the jaw.

Biomechanical Characteristic of SMA

Wires and Clamps

To construct the pressure-controlled vessel clamp,

TiNi SMA with the Ni component of 55.98-

56.04% was used.13 The diameters of wires were

0.3 and 0.4 mm. The length of the SMA wire was

40 mm. The biomechanical characteristic of SMA

wires was tested using a universal tensile testing

machine (Instron 4505 Universal Machine; Instron,

Norwood, MA) with the procedure described previ-

ously.13 The tensile test was performed with rates of

Fig. 2. A schematic diagram of the experimental setup of

biomechanics for the testing of clamping pressure and the

handling displacement of vessel clamps.

Vol. 23, No. 6, November 2009 Shape memory alloy pressure-controlled vascular clamp 815

60 mm/min.13 For each kind of wire the tensile test

was repeated three times. The modulus of elasticity

is calculated from the stressestrain curve with the

method described previously:6,13 s ¼ E e, where sis stress, e is strain, and E is modulus of elasticity.

Only the slope of the initial straight-line portion

(elastic deformation) was taken.

The schematic diagram of the micropressure

analyzing system used for testing the biomechanical

characteristic of clamps is shown in Fig. 2. The

experimental setup consists of a holder for fixing

one branch of the clamps and a uniaxal moving

stage. The SMA clamp was fixed on the operation

stage. An electronic micropressure sensor (PSM-

2KA, 200 KPa; Kyowa, Tokyo, Japan) was attached

to the jaws of the clamp, 1 cm from the tips. Its

upper handle was fixed to the load cell (CB17-3K-

11, 29.6 N; Minebea, Fujisawa, Japan). The load

cell was moved by a uniaxal moving stage controller

(QT-CD1; Chou Precision Industrial, Tokyo, Japan).

The pressure sensor and the controller were con-

nected to a computer. After the vessel model was

settled between the jaws, 1 cm from the tips, the

load cell was moved vertically downward. Data of

the displacement of load cell and the pressure of

the clamp tip were simultaneously collected and

analyzed by QT-EDIT software (Kyowa). The

sampling frequency for the synchronous data acqui-

sition of load cell force, displacement sensing, and

the jaw pressure was set at 50 Hz.

Two kinds of clamps with SMA wires of different

diameter and the base model clamp were tested. For

each kind of clamp the loadingereleasing experi-

ment was repeated three times. The velocity was

60 mm/min.13

Vessel Model Preparation

All procedures were approved by the Animal Ethics

Screening Committee of the University of Tohoku.

The animals used in this study received humane

care in compliance with the Guide for the Care and

Use of Laboratory Animals, published by the Univer-

sity of Tohoku.

Twenty-four healthy male Landrace pigs with an

average weight of 48.3 ± 3.6 kg were used. After

intramuscular administration of ketamine chloride

(10 mg/kg) and atropine sulfate (0.02 mg/kg), anes-

thesia was induced via an endotracheal tube and

maintained with isofluorane, 5% and 1-2%, respec-

tively. A total of 24 thoracic aortas were taken out

and prepared as in vitro vessel experimental samples.

The aortas were taken out within 30 min of anes-

thesia being administered. The region of the aorta

used began 5 cm from the apex of the aortic arch.

Aortic diameter was measured (16.4 ± 4.1 mm,

n¼ 24). Branches and surrounding tissues were

removed. The samples were perfused and preserved

in 4�C phosphate buffer.

Clamping Traumatic Experiment

The 24 thoracic aortas were randomly divided into

four groups: group 1 of SMA clamp, group 2 of

SMA clamp, a group of the base model clamp, and

a control group. In group 1, a clamp embedded

with SMA wire of 0.3 mm diameter was used. In

group 2, SMA wire of 0.4 mm diameter was used.

The clamping experiment was conducted according

to the protocol of Margovsky et al.7 and Neville

et al.15 Briefly, clamps were placed simultaneously

on the middle section of each artery at 1 cm inter-

vals, one next to the other, avoiding any branches

and surrounding tissues. The clamps were closed

to notches I, II, and III and held in place for

15 min at each notch. In the control group, a dummy

clamping operation was done with an adjustable

clamp, the pressure of which was controlled at

50 mm Hg.

The clamps were then removed and the arteries

were cut into 1 cm sections, each corresponding to

the site of application of the clamps. The procedure

was repeated on 24 arteries, and the specimens

were evaluated with histological stain.

Histological Stain and Traumatic

Evaluation

The histological stain was used to evaluate damage

on the vessel walls. Sagittal sections were cut off

from the blood vessel wall, preserved in 10% form-

aldehyde, embedded in paraffin, sliced (6 mm), and

Fig. 3. The stressestrain result of SMA wires. The diam-

eters of SMA wires used were 0.3 and 0.4 mm.

Fig. 4. The displacementepressure reaction of pressure-

controlled SMA clamps and the base model clamp. The

diameters of SMA wires used were 0.3 and 0.4 mm.

816 Zhang et al. Annals of Vascular Surgery

stained with hematoxylin and eosin. Histological

sections were observed by microscopy (DM 5000B;

Leica, Wetzlar, Germany), then evaluated by deter-

mining the grade of damage according to the

method described by Moore et al.:16 grade 0, no

injury; grade 1, endothelial deformation, intact

endothelium; grade 2, endothelium compressed

obviously; grade 3, endothelial tear <3 mm; grade

4, endothelial denudation, flap �3 mm; grade 5,

endothelial and minor medial injury; grade 6, major

medial injury.

All evaluations were done by the same observer

in a blinded fashion. Three values were made for

each specimen. All values were expressed as the

mean ± standard error (n¼ 18). Values for vessel

trauma caused by SMA clamps were compared

with the base model and control clamps using anal-

ysis of variance. Values were considered signifi-

cantly different at p< 0.05.

RESULTS

Biomechanical Character of SMA Wires

and Pressure-Controlled Clamps

The results of tensile tests with SMA wires (0.3 and

0.4 mm) are shown in Fig. 3. From the stressestrain

curves, it can be seen that the remarkable stress-

induced transformation starts at 1.5-2% strain and

finishes at 6.5-7%. The modulus of elasticity of the

0.3 mm SMA wire was 21.77 ± 0.03 GPa and that

of the 0.4 mm SMA wire was 26.80 ± 0.06 GPa.

The results of biomechanical tests of SMA pres-

sure-controlled clamps are shown in Fig. 4. The

base model clamp was taken as a control. When the

handle was displaced along the vertical line, a very

small displacement was accompanied by a dramatic

increase of pressure (Table I). The displacementepressure curve shows an exponential type (Fig. 4).

When using the clamp with inserted SMA wire of

0.3 or 0.4 mm in diameter as a limiter, it could be

clearly seen that displacementepressure curves

were altered. At first, the pressure on the clamp tip

gradually rose with the handling displacement

(Fig. 4). However, when the prolonged deformation

of the SMA wire arrived at the critical point, the

direction of the displacementepressure curve was

changed (Table I). It displayed pressure saturation

(Fig. 4). In other words, pressure remained constant

after the critical point. The saturated pressure of the

clamp in the 0.4 mm wire group was larger than the

pressure in the 0.3 mm wire group.

Histological Evaluation

After 15 min of clamping, it can be seen from

Figure 5 that in the base model group there are rela-

tively apparent histological changes on the vessels.

After receiving compression of clamps closed at

notch I, the intimal layer was disrupted. Under it,

very small cracks in the media formed. At notch II,

the vessel wall showed an obviously compressed

deformation. The media was broken, replicated,

and overlapped. When the handle closed forward

of notch III, the pressure was so strong that the

endothelial intima and media were crushed and

shreds formed. An interesting phenomenon is the

adventitia; due to its relative softness, except flat-

tening, its integrity showed no apparent change. If

this occurred during surgery, the intimal trauma

would be neglected when judged only by the

appearance of the outer surface of the vessel.

Compared with the base model group, at notch I

no obvious imprint of clamp compression could be

observed in either SMA group. The intima remained

intact. When the clamp was closed at notch II, slight

imprints could be seen on the intima. We observed

Tab

leI.

Cla

mpin

gpre

ssu

rean

dtr

au

mati

csc

ore

son

the

vess

el

wall

pro

du

ced

by

clam

ps

(n¼

18)

Pre

ssu

re(m

mH

g)

Tra

um

ati

csc

ore

Gro

ups

Notc

hI

Notc

hII

Notc

hII

Ip

aN

otc

hI

Notc

hII

Notc

hII

Ip

SM

A(0

.3m

m)

409.1

4±

69.6

7490.5

4±

69.6

7495.9

0±

74.2

20.9

50.5

0±

0.2

60.6

7±

0.2

30.7

2±

0.2

10.6

0

SM

A(0

.4m

m)

640.9

1±

84.7

7811.7

2±

140.4

5846.2

5±

159.2

10.7

90.8

3±

0.1

51.6

6±

0.2

31.8

3±

0.1

50.5

5

Base

model

1,0

16.4

3±

159.2

12,7

20.5

8±

192.3

74,6

00.3

2±

420.4

50.0

32.8

3±

0.1

54.6

7±

0.4

15.6

7±

0.5

20.0

07

Con

trol

50

50

50

0.2

2±

0.1

80.1

7±

0.1

50.2

2±

0.1

80.6

8

p0.0

15

b<

0.0

01

b<

0.0

01

b<

0.0

01

<0.0

01

<0.0

01

aC

om

pari

son

betw

een

notc

hII

an

dn

otc

hII

I.bC

om

pari

son

of

SM

Agro

up

1(0

.3m

m)

an

dSM

Agro

up

2(0

.4m

m)

wit

hbase

model

gro

up.

Vol. 23, No. 6, November 2009 Shape memory alloy pressure-controlled vascular clamp 817

that the elastic lamina structure on the media was

slightly compacted. When the clamps closed at the

notch III, the clamp imprint on the vessel wall

became more obvious. Because the force remained

in the safety range, the intima was preserved in

both groups. The continuity of the intima was not

disturbed or broken. Only mild compressions could

be seen on the elastic structure of the media. The

compressions found in group 1 (0.3 mm diameter

SMA) were less obvious than those in group 2

(0.4 mm diameter SMA). However, no other

distinct signs could be observed.

In the control group, the intima, media, and

adventitia remained intact (Fig. 5).

Traumatic Evaluation

The traumatic score and its relation to clamping

pressure are shown in Table I. It could be seen that

vessel trauma correlated with clamp pressure. The

scores of the SMA groups were lower than the score

of the base model group. On the other hand, in the

base model group, as the pressure increased, trau-

matic scores increased from 4.67 ± 0.41 at notch II

to 5.67 ± 0.52 at notch III. There was a significant

difference in scores. However, in the two SMA

groups, since the pressure at notch II and notch III

remained constant, the traumatic score did not

increase (Table I).

DISCUSSION

As a kind of shape-memory, superelastic, and

fatigue-resistant biomaterial, SMA has been used

in many minimally invasive surgery and diagnostic

applications, such as vessel stents11 and ortho-

dontic appliances.12 On the other hand, with the

component changed, some SMA materials have

another unique mechanical property, stress satura-

tion.14 Thus, it may be a useful material in those

circumstances in which controlling pressure is

important.

In our study, a pressure-controlled clamp using

SMA material was designed. First, it should be

understood whether this design could effectively

control the pressure put on vessels. An in vitro

vessel model of pig thoracic aorta was used for its

comparable size to that of humans. From the results

of the loadingedisplacement experiment (Fig. 4), it

could be seen that the pressure put on vessels by

base model clamps increased with the handle’s

displacement. The displacementepressure curve is

an exponential shape. In the SMA groups the rela-

tion was altered. There is an obvious ‘‘pressure

constant’’ on the displacementepressure curve,

Fig. 5. Representative histological outcome of blood

vessels after clamping by pressure-controlled clamps,

the base model clamp, and the control clamp at notches

I, II, and III, respectively. First column, pressure-controlled

clamp using SMA wire diameter of 0.3 mm; second

column, pressure-controlled clamp using SMA wire

diameter of 0.4 mm; third column, base model clamp;

fourth column, control group. Upper row, vessel traverse

observation after clamping at notch I; middle row, vessel

traverse observation after clamping at notch II; bottom

row, vessel traverse observation after clamping at notch

III. Hematoxylin and eosin stain, �200.

818 Zhang et al. Annals of Vascular Surgery

which represents the pressure saturation. It proved

that the new design of an additional fulcrum

combined with SMA wire successfully displays the

SMA material’s advantages. The ‘‘pressure

constant’’ supplies a long protective range of pres-

sure. This is the basis of the new design, which

improves safety. In this way the damage caused by

overpressure could be prevented.

To keep the pressure within the safety range, the

pressure of saturation of the SMA clamp could be

precisely regulated by two factors. One is the

diameter of wire. From Figure 3 it can be seen

that the saturation pressure of clamps using

0.4 mm SMA wire is higher than that in clamps

using 0.3 mm SMA wire. Another regulating factor

is the length of wires. This relates to the critical

point. The critical point, as calculated by our

group,13 started at 1.5-2% of the SMA wire stress-

induced transformation and finished at 6.5-7%

(Fig. 3). In our experiment the length of SMA

wire used was 40 mm, so the pressure was strictly

controlled within the safety range measured by

other researchers.7

In practical usage, clamps will be closed at suitable

notches. The change in the relation of loading and

strain will be reflected in the clamping pressure at

each notch. Thus, the clamping pressure at each

notch was measured as well. In this study, the clamp-

ing pressure at each notch clearly displayed the

change in the relation of stress and strain of SMA

(Table I). It can be see that with a small displacement

moving from notch II to notch III, clamping pressure

in the base model group rose dramatically from

2,720.58 ± 192.37 to 4,600.32 ± 420.45 mm Hg. In

contrast, in the pressure-controlled group, the force

rose only from 490.54 ± 69.67 to 495.90 ± 74.22 mm

Hg with clamp closing from notch II to notch

III (0.3 mm SMA wire group) and from

811.72 ± 140.45 to 846.25 ± 159.21 mm Hg (0.4 mm

SMA wire group). The pressure increase was restric-

ted. The pressure-controlling effect of the clamp was

revealed. This has important implications for the

practical usage of clamps because excessive pressure

higher than MOF is the direct reason leading to

trauma of vessels.6,7

To further check whether the safety of clamps

could be improved through this new design, the

vessel model was also evaluated histologically

following clamping. Results showed that in the

base model group intimal damage occurred even at

notch I, due to high pressure (Fig. 5). At notch II,

with pressure dramatically increased, the trauma

Vol. 23, No. 6, November 2009 Shape memory alloy pressure-controlled vascular clamp 819

penetrated into the media. Shreds formed and

media folds could be observed. In contrast, in both

of the SMA groups, since the pressure was effec-

tively controlled, even at notch III only slight

compression could be seen in the sample of vessels.

The intima and media maintained integrity.

The innermost layer of blood vessels, the tunica

intima, is composed of a monolayer of endothelial

cells, the endothelium.17 Pabst et al.,5 using scan-

ning electron microscopy, proved that marked

endothelial fractures and intimal injury occurred

when vessels were clamped at a pressure higher

then MOF. The intimal lesion will be followed by

a pathological reaction such as fibrin deposition,

pseudoaneurysm formation,5 and embolization,

which could accelerate vessel injury.7,18 For this

reason, an intact intima is very important.

The relation between the score of trauma and the

force at each notch proved that it is due to the effec-

tive control of pressure that the difference in the

pathology was produced. This also shows the impor-

tance of controlling pressure for the safety of blood

vessels (Table I). Clamping pressure at notch I led

the base model clamp to cause trauma in vessel

walls. However, the trauma was within the intima.

As the handle moved to notch II and notch III, the

dramatically increased pressure caused the vessel

wall more serious trauma. The traumatic score

increased from 4.67 ± 0.41 at notch II to

5.67 ± 0.52 at notch III. This would bring serious

complications in clinical application. In the SMA

group, pressure increased moving from notch I to

notch II, leading the imprint in the vessel wall to

become more obvious. However, when the handle

moved from notch II to notch III, the pressure

remained constant. Thus, the score for damage on

vessel walls did not increase. This means that the

trauma caused by increased pressure in the control

group was avoided. This confirmed that the safety

of the clamp was improved.

The experiment was not in vivo, and some trau-

matic phenomena such as thrombin formation,

flow-induced shear stress, and reperfusion lesion

could not be analyzed. However, on the basis of

intimal lesions, this kind of pathological lesion would

certainly further accelerate vessel injury, which has

been proved by many other researchers.18,19

This experiment confirmed that the safety of the

base model improved with design improvement.

SMA plays a role by controlling the pressure of the

clamp. Certainly, the clamp is only a rough model.

The final design should take into account other

factors, such as jaw geometry, vessel size, systematic

blood pressure, and arterial wall elasticity. Other

special conditions, such as atherosclerotic arteries,

need to be considered as well. This technique should

be studied further.

Above all, this unique mechanical property was

effective at controlling clamp pressure. The new

design successfully improved the clamp’s safety.

This pressure-controlling method can be also used

in many related research areas, such as in the use

of endoscopes, computer simulation, and surgical

navigation.

CONCLUSIONS

The effect of pressure-controlling SMA clamps on

the in vitro vessel model was confirmed. Pressure

saturation was shown in the displacementepressure

curve. Occlusive pressure could be controlled in

a safe range.

Histological observation and the traumatic score

proved that the safety of clamps was improved

with the new design of pressure control.

This work was partially supported by the special coordination

funds for Promoting Science and Technology from the Ministry

of Education, Culture, Sports, Science, and Technology, Japan.

The authors thank Mizuho Ikoh Co. for assistance in prototyping.

REFERENCES

1. Gersak B, Trobec R, Krisch I, et al. Loss of endothelium-

mediated vascular relaxation as a response to various

clamping pressures. Eur J Cardiothorac Surg 1996;10:

684-689.

2. Darcin OT, Cengiz M, Ozardali I, et al. Pressure-controlled

vascular clamp: a novel device for atraumatic vessel occlu-

sion. Ann Vasc Surg 2004;18:254-256.

3. Margovsky AI, Lord RS, Meek AC, et al. Artery wall damage

and platelet uptake from so-called atraumatic arterial clamps:

an experimental study. Cardiovasc Surg 1997;5:42-47.

4. Coelho JC, Sigel B, Flanigan DP, et al. Arteriographic and ultra-

sonic evaluation of vascular clamp injuries using an in vitro

human experimental model. Surg Gynecol Obstet 1982;155:

506-512.

5. Pabst TS, 3rd, Flanigan DP, Buchbinder D. Reduced intimal

injury to canine arteries with controlled application of vessel

loops. J Surg Res 1989;47:235-241.

6. Dujovny M, Wakenhut N, Kossovsky N, et al. Minimum

vascular occlusive force. J Neurosurg 1979;51:662-668.

7. Margovsky AI, Chambers AJ, Lord RS. The effect of increasing

clamping forces on endothelial and arterial wall damage: an

experimental study in the sheep. Cardiovasc Surg 1999;7:

457-463.

8. Moore WM, Manship LL, Bunt TJ. Differential endothelial

injury caused by vascular clamps and vessel loops. I. Normal

vessels. Am Surg 1985;51:392-400.

9. Harvey JG, Gough MH. A comparison of the traumatic

effects of vascular clamps. Br J Surg 1981;68:267-272.

10. Dogliotti AM, Guglielmini G. A new two-lever compression

clamp for vascular surgery. Surgery 1955;38:717-719.

820 Zhang et al. Annals of Vascular Surgery

11. Vogel PM, Parise C. Comparison of SMART stent placement

for arteriovenous graft salvage versus successful graft PTA.

J Vasc Interv Radiol 2005;16:1619-1626.

12. Es-Souni M, Es-Souni M, Fischer-Brandies H. Assessing the

biocompatibility of NiTi shape memory alloys used for medical

applications. Anal Bioanal Chem 2005;381:557-567.

13. Luo Y, Kodaira S, Zhang Y, et al. Application of superelastic

SMAs in less invasive hemostatic forceps. Smart Mater.

Struct 2007;16:1061-1065.

14. Gall K, Tyber J, Brice V, et al. Tensile deformation of NiTi

wires. J Biomed Mater Res A 2005;75:810-823.

15. Neville RF, Padberg FT, DeFouw D, et al. The arterial wall

response to intimal injury in an experimental model. Ann

Vasc Surg 1992;6:50-54.

16. Moore WM, Jr, Bunt TJ, Hermann GD, et al. Assessment of

transmural force during application of vascular occlusive

devices. J Vasc Surg 1988;8:422-427.

17. Gasser TC, Schulze-Bauer CA, Holzapfel GA. A three-

dimensional finite element model for arterial clamping.

J Biomech Eng 2002;124:355-363.

18. Manship LL, Moore WM, Bynoe R, et al. Differential

endothelial injury caused by vascular clamps and vessel

loops. II. Atherosclerotic vessels. Am Surg 1985;51:

401-406.

19. Arnljots B, Dougan P, Wieslander JB, et al. Platelet accumu-

lation and thrombus formation after microarterial injury. An

experimental study in rabbits. Scand J Plast Reconstr Surg

Hand Surg 1994;28:167-175.