Topical application of b-radiation to reduce intimal hyperplasia after

carotid artery balloon injury in rabbit

A possible application for brachytherapy in vascular surgery

David Rosenthala,*, Scott L. Stevensb, C.S. Skillernb, Eric D. Wellonsa, Keith Robinsonc,John H. Matsuuraa, Brian J. Gannond

aDepartment of Vascular Surgery, Atlanta Medical Center, 315 Boulevard NE, Suite 412, Atlanta, GA 30312, USAbSchool of Medicine, Department of Surgery, University of Tennessee, Nashville, TN, USA

cAtlanta Cardiovascular Research Institute, Norcross, GA, USAdPerkinElmer Life Sciences, Boston, MA, USA

Received 01 April 2002; received in revised form 01 June 2002; accepted 01 July 2002

Abstract Purpose: Endovascular brachytherapy for the prevention of intimal hyperplasia (IH) and restenosis

after balloon/stent angioplasty has proven effective both in animal preparations and clinical trials. A

variety of b-emitting isotopes and catheter-based devices have been developed for the delivery of

low-dose radiation in clinical coronary and peripheral trials. No platform, however, has yet been

developed for brachytherapy in concert with vascular surgical operations. The purpose of this study

was to evaluate the vascular histopathologic response following balloon injury to rabbit carotid

arteries with and without topically applied low-dose b-radiation.Methods: The b-emitting isotope strontium-90 (Sr-90) was conjugated onto the matrix of

polypropylene (PLYP) mesh. Rabbit carotid arteries were balloon-injured with a #2 embolectomy

catheter. Six carotid arteries were wrapped with nonradioactive PLYP mesh (controls) and Sr-90

(� 90 mCi) PLYP mesh in order to deliver low-dose radiation to the vessel wall from the external

(adventitial) surface. Tissue was harvested at 6 weeks and processed for histologic examination.

Results: There was consistent blockade of fibrocellular neointima formation with virtually no

neointima present in all treated segments, compared to moderate neointima formation in

controls. Medial thinning and smooth muscle cell (SMC) necrosis were also associated with

topical brachytherapy.

Conclusion: b-Radiation applied by an externally wrapped PLYP mesh labeled with Sr-90

markedly suppressed neointima formation in an animal vascular surgical injury model. Further

studies, however, are necessary to determine a suitable isotope and dosage for clinical

application. D 2002 Elsevier Science Inc. All rights reserved.

Keywords: b-radiation brachytherapy topical application in vascular surgery

1. Introduction

Human and experimental studies have demonstrated that

intimal hyperplasia (IH) and vascular remodeling are the

principal factors leading to restenosis after arterial catheter-

based and endovascular surgical intervention [1–4]. The

roles of the medial smooth muscle cells (SMCs) and

adventitial fibroblasts after vascular injury is well estab-

lished, and the proliferation and migration of the medial

SMCs to the luminal space is critical in the formation of

neointima. Radiation therapy has been shown to reduce

NIH in animal models and arterial restenosis after clinical

catheter-based intervention. The mechanism of this inhib-

itory effect is thought to be one of chromosomal damage

to the SMCs, creating mitotic cell death when these cells

1522-1865/02/$ – see front matter D 2002 Elsevier Science Inc. All rights reserved.

PII: S1522 -1865 (02 )00137 -3

* Corresponding author. Tel.: +1-404-524-0095; fax: +1-404-658-

9558.

E-mail address: docro@mindspring.com (D. Rosenthal).

Cardiovascular Radiation Medicine 3 (2002) 16–19

are stimulated to divide, as a consequence of vessel wall

medial injury.

Brachytherapy, or the placement of a radioactive source

on tissue surfaces, represents an ideal means of delivering a

low dose of radiation directly to the target vessel [5,6].

Means to deliver ionizing radiation from luminally posi-

tioned catheter based systems have been developed and

found effective in both animal preparations and human

clinical trials, and three systems are currently FDA-

approved and commercially marketed for coronary artery

endovascular brachytherapy. However, no system for the

‘‘extravascular’’ topical application of low-dose radiation

has been developed, but such a device would be a useful

adjunct at the time of cardiovascular operations.

We developed a prototype system to deliver b-radiationfrom the adventitial surface of an artery by conjugating

strontium-90 (Sr-90) onto a PLYP mesh substrate in order to

‘‘wrap’’ the vessel in a radiation blanket. The purpose of this

study was to evaluate the vascular histopathologic response

following balloon injury to the rabbit carotid artery, with

and without topical low-dose b-radiation.

2. Materials and methods

The animal work was approved by the Animal Care and

Use Committee of University of Tennessee College of

Medicine. All animal care was in accordance with the

‘‘Guide for the Care and Use of Laboratory Animals’’

(Office of Science and Health Reports CPRR/NIH 1996).

Six New Zealand white rabbits weighing between � 2.5 and

3.5 kg each were subjected to overstretch balloon injury of

one carotid artery by a #2 embolectomy catheter. The

animals were sedated with a combination of ketamine

hydrochloride (35 mg/kg) and xylazine (7 mg/kg). After

intravenous administration of � 1000 units of heparin, the

embolectomy catheter was introduced via the external

carotid artery and advanced retrograde into the common

carotid artery. The balloon was inflated and withdrawn to

the carotid bifurcation three times, after which the external

carotid was ligated.

PLYP mesh (Johnson and Johnson, Summerville, NJ)

was wrapped around a � 1-cm length of the injured artery

(control), while the low-dose Sr-90-labeled mesh (test) was

wrapped around an adjacent � 1-cm segment; mesh wraps

were secured with 5–0 PLYP suture. The proximal and

distal control versus test mesh sites was alternated with each

animal and the wounds closed in a standard fashion. All

animals were allowed to recover and returned to routine

care. Animals were maintained on a normal laboratory diet

and were euthanized 6 weeks after operation.

Postmortem, the arteries were isolated and flushed with

heparinized saline followed by 10% neutral buffered form-

alin. Segments of proximal and distal artery wrapping were

cut and dehydrated in graded ethanol series, exchanged with

xylene and embedded in paraffin. Sections � 5–6-mm-thick

were cut on a rotary microtome and collected on glass

slides. Adjacent sections were stained with hematoxylin and

eosin (H&E) and Verheoff-Masson elastin-trichrome to

evaluate general tissue and cell morphology. An experi-

enced observer unaware of treatment group assignment

performed histopathologic analysis and semiquantitative

scoring (0 = none, 1 =mild, 2 =moderate, 3 = severe) of each

segment for: extent of balloon-induced vascular injury,

neointima formation, thrombus, inflammation and media

necrosis was performed on each segment. Mann–Whitney

rank sum tests were used to compare irradiated to control

segments for these specific histopathologic criteria, and a

critical value of P < .05 was considered to indicate a

significant between-groups difference.

3. Results

Six animals underwent carotid artery injury. One animal

was found dead 1 day after surgery and another died at

2 weeks. The causes of death were not established. The

arteries from these rabbits were harvested and examined, but

were not included in the semiquantitative histopathologic

scoring. Both vessels contained both postmortem thrombus

and varying degrees of platelet-rich, antemortem thrombus;

in the animal found dead at 2 weeks, this showed partial

organization and IH was also present. In the animal found

dead at 1 day, the vessel segment that had been wrapped with

Sr-90-conjugated mesh displayed extensive media necrosis.

The appearances of vessel segments harvested 6 weeks

after injury and wrapping showed a distinctly bimodal

histomorphologic pattern. Artery segments, which had been

Fig. 1. Light microscopy of rabbit carotid arteries fixed 6 weeks after

balloon injury. (a, b) Wrapped with polypropylene mesh only (control);

(c, d) wrapped with polypropylene mesh incorporating Sr-90 � 90 mCi/cm2

(irradiated). (a, c) Verheoff-Masson stain, 40� ; (b, d) H&E stain, 200� .

Control segment shows meointima formation typical of this animal

preparation; irradiated vessel shows virtually no fibrocellular neointima,

but necrosis in deeper layers of the attenuated tunica media.

D. Rosenthal et al. / Cardiovascular Radiation Medicine 3 (2002) 16–19 17

wrapped with radioactive mesh, showed virtually no fibro-

cellular neointima formation and often displayed substantial

media necrosis (Fig. 1a and b). Control-wrapped segments

had no necrosis but had considerable neointima formation

consisting of stellate, round and spindle-shaped cells in an

abundant mixed extracellular matrix (Fig. 1c and d). Occa-

sionally, thrombus was present in the lumens of either

irradiated or control-wrapped segments; in the case of one

control segment, this showed partial organization by fibro-

cellular tissue of composition similar to proliferative neo-

intima. Diffuse granulocytic inflammatory infiltrates were

present in the adventitia and perivascular space of all vessel

segments and did not appear greater in irradiated samples

than controls; occasionally, more focal and intense infiltrates

were seen.

Semiquantitative scoring of vessel segments for histopa-

thologic criteria was performed with the observer unaware

of treatment group assignment (Table 1). The code was then

broken and the scores grouped, and between-groups com-

parisons made using Mann–Whitney rank sum tests. There

were no significant differences between sham-treated and

irradiated segments for vascular injury (P= .69), thrombus

(P= .88) or inflammation (P= .34). However, the irradiated

segments showed significantly reduced neointima formation

(P= .03) and a trend towards increased necrosis (P= .11).

4. Discussion

The problem of NIH, and hence arterial restenosis, is a

major clinical problem for cardiovascular surgeons. Informa-

tion from HCFA demonstrates that over 700,000 cardiovas-

cular operations were performed on Medicare-aged patients

in the United States in 1998 [7] and reported restenosis rates

from these operations approach, at a minimum 15% at 2 years

[8–10]. It is easy to comprehend, therefore, the enormity of

this problem in terms of patient suffering and healthcare

costs. Worldwide, these figures quadruple.

This study was performed to evaluate the use of topically

applied b-radiation to balloon injured rabbit carotid arteries

in an attempt to reduce IH. A b-emitting source was chosen

because it is less penetrating than gamma radiation to

surrounding tissues, requires less exposure to achieve the

desired affect, has an appropriate shelf-life and has minimal

handling risks for operating room personnel; indeed, the

only protection necessary for operating room personnel is

latex gloves.

The histologic analysis in this study demonstrated a

marked reduction in neointima formation in the irradiated

artery segments. There was virtual abrogation of neointima

formation in irradiated vessel segments without signific-

antly increased thrombus or inflammatory response. These

results are promising; however, the trend towards increased

necrosis in the tunica media suggests possible adverse

complications related to aneurysm or perforation with this

dose and isotope.

Despite the beneficial inhibitory effect on neointima

formation, Sr-90 with its half-life of 28 years, probable

associated need for radiation protection (despite relatively

low activity) and its nature as essentially a permanent

implant, could not be recommended as a desirable isotope

for this topical application.

This study demonstrates that topically placed b-emitting

source can decrease IH. Further studies are ongoing which

will aid in defining a more suitable isotope and the dosimetry

necessary to inhibit IH without excessive necrosis. If topical

brachytherapy proves efficacious, this simple and safe means

of preventing IH and, thereby, arterial restenosis will be a

significant adjunct for cardiovascular surgical patients.

5. Study limitations

This study was an initial pilot experiment to assess the

feasibility and histopathologic effects of topical b-radiationin an animal preparation of neointimal hyperplasia, and was

not powered to enable standard histomorphometric evalu-

ation of neointima formation and other vascular changes in

response to injury in the presence and absence of irra-

diation. Therefore, only a semiquantitative evaluation of

vascular morphology was performed, with between-groups

comparisons made using nonparametric tests appropriate for

such a study design and evaluation technique. Additionally,

in this pilot study, no dosimetry was undertaken, so at

present, we have no data to indicate the radiation dose

required to elicit an inhibitory effect on neointima formation

using this approach.

References

[1] Lafont A, Guzman LA, Whitlow PL, Goormastic M, Cornhill JF,

Chisolm GM. Restenosis after experimental angioplasty. Intimal, me-

dial and adventitial changes associated with constrictive remodeling.

Circ Res 1995;76:996–1002.

[2] Labinaz M, Pels K, Hoffert C, Aggarwal S, O’Brien ER. Time course

Table 1

Semiquantitative scoring of specific histopathologic criteria in mesh-

wrapped balloon-injured rabbit carotid arteries at 6 weeks

Rabbit Segment Group VI NF Thromb Inflam Necr

9 prox rad 2 0 1 1 3

9 dist ctrl 2 2 2 1 0

10 prox ctrl 0 1 0 1 0

10 dist rad 0 0 0 0 0

12 prox rad 2 0 0 1 3

12 dist ctrl 2 2 0 2 0

13 prox rad 2 0 1 1 2

13 dist ctrl 1 2 0 1 0

0 = none, 1 = mild, 2 =moderate, 3 = severe, rad = radioactive mesh,

ctrl = control mesh, VI = vascular injury, NF = neointima formation,

thromb= thrombus and hemorrhage throughout vessel wall, inflam= in-

flammation throughout vessel wall, necr =medial necrosis.

D. Rosenthal et al. / Cardiovascular Radiation Medicine 3 (2002) 16–1918

and importance of neoadventitial formation in arterial remodeling

following balloon angioplasty of porcine coronary arteries. Cardio-

vasc Res 1999;41:255–66.

[3] Shi Y, O’Brien JE, Fard A, Mannion JD, Wang D, Zalewski A.

Adventitial myofibroblasts contribute to neointimal formation in in-

jured porcine coronary arteries. Circulation 1996;94: 1655–64.

[4] Glagov S. Intimal hyperplasia, vascular modeling, and the reste-

nosis problem. Circulation 1994;89:2888–91.

[5] Waksman R, Kim WH, Chan RC, Vodovotz Y, Pierre T, Collins S,

Seabron R, Tio FO, Sweet W. Effectiveness of a b-emitting radio-

active stent for the treatment of in-stent restenosis in porcine

coronaries. Circulation 1998;98:I–779.

[6] Ali MN, Kaluza J, Mazur W, Fajardo LF, Schultz D, Bradshaw

K, Berens K, Khan M, Raizner AE. The effect of intracoronary

beta-radiation on neointimal formation and vascular remodeling in

balloon-injured porcine coronary arteries. Effect of dose rate.

J Interv Cardiol 1999;12:271–82.

[7] Health Care Finance Administration: office of information systems,

Medicare Division Support Systems, Office of Strategic Planning,

1998.

[8] Schwarcz TH, Yates GN, Ghobrial M, Baker WH. Pathologic char-

acteristics of recurrent carotid artery stenosis. J Vasc Surg 1987;5:

280–8.

[9] Carballo RE, Towne JB, Seabrook GR, Freischlag JA, Cambria RA.

An outcome analysis of carotid endarterectomy: the incidence and

natural history of recurrent stenosis. J Vasc Surg 1996;23:749–54.

[10] Nobuyoshi M, Kimura T, Nosaka H, Mioka S, Ueno K, Yokoi H,

Hamasaki N, Horiuchi H, Ohishi H. Restenosis after successful

percutaneous transluminal coronary angioplasty: serial angiographic

follow-up of 229 patients. J Am Coll Cardiol 1988;12:616–23.

D. Rosenthal et al. / Cardiovascular Radiation Medicine 3 (2002) 16–19 19