Dose-dependent smooth muscle cell proliferation induced by thermal injury with pulsed infrared lasers

Epstein, MB Leon and RF BonnerPC Douek, R Correa, R Neville, EF Unger, M Shou, S Banai, VJ Ferrans, SE

with pulsed infrared lasersDose-dependent smooth muscle cell proliferation induced by thermal injury

ISSN: 1524-4539 Copyright © 1992 American Heart Association. All rights reserved. Print ISSN: 0009-7322. Online

72514Circulation is published by the American Heart Association. 7272 Greenville Avenue, Dallas, TX

1992, 86:1249-1256Circulation

http://circ.ahajournals.org/content/86/4/1249located on the World Wide Web at:

The online version of this article, along with updated information and services, is

http://www.lww.com/reprintsReprints: Information about reprints can be found online at [email protected]: 410-528-8550. E-mail: Kluwer Health, 351 West Camden Street, Baltimore, MD 21202-2436. Phone: 410-528-4050. Permissions: Permissions & Rights Desk, Lippincott Williams & Wilkins, a division of Wolters http://circ.ahajournals.org//subscriptions/Subscriptions: Information about subscribing to Circulation is online at

at VA MED CTR BOISE on July 10, 2011http://circ.ahajournals.org/Downloaded from

http://circ.ahajournals.org/

1249

Laboratory Investigation

Dose-Dependent Smooth Muscle CellProliferation Induced by Thermal Injury With

Pulsed Infrared Lasers

Philippe C. Douek, MD; Rosaly Correa, MD, PhD; Richard Neville, MD; Ellis F. Unger, MD;Matie Shou, MD; Shmuel Banai, MD; Victor J. Ferrans, MD, PhD; Stephen E. Epstein, MD;

Martin B. Leon, MD; and Robert F. Bonner, PhD

Background. Recently, laser-heated and radio frequency-heated balloon angioplasty techniques havebeen proposed as a means to treat or minimize dissection and elastic recoil but have been associated witha high rate of clinical restenosis. Similarly, pulsed laser angioplasty techniques proposed to minimize

thermal injury while ablating obstructing atheroma have failed to reduce clinical restenosis. Because "hotballoon" and pulsed laser angioplasty create both mechanical and thermal injury, it has been difficult todiscern the cause of the smooth muscle cell (SMC) proliferation resulting in restenosis and whether suchmagnitude of proliferation is predictable and dose related. This study was undertaken to explore theseissues.Methods and Results. Localized thermal lesions accompanying efficient ablation were created with a

pulsed Tm:YAG laser in nine rabbit aortas, which consistently led to a focal proliferation of SMC thatfilled the ablated region by 4 weeks. Transcutaneous Ho:YAG pulsed laser irradiation at multipleindependent sites of 24 central rabbit ear arteries without ablation led to brief =30°C thermal transientsand thermal damage to the artery wall resulting in significant neointimal proliferation by 3 weeks and a

mean cross-sectional narrowing of 59±17% at a dose of 390 mJ/mm2. Acute and chronic responses tovarying total energy deposition were studied by histology after the rabbits were killed at 2 hours to 4weeks. Arterial segments midway between laser injuries were unaffected and served as internal controls.Neointimal proliferation at 3 weeks- after laser injury exhibited a clear dose dependence. Meancross-sectional narrowing increased from 34±10%o to 85±15% as laser fluence increased from 240 mJ/cm2to 640 mJ/cm2 (r=0.84). Similarly, cross-sectional narrowing caused by SMC neointimal proliferationincreased from 20+10%o to 77+17% for a fixed surface irradiation as the depth of the most superficialarterial media decreased from 600 ,im to 330 Am (r=0.94).

Conclusions. Thermal injury to the arterial wall is a potent stimulus for SMC proliferation and maynecessitate reduction in laser or thermal energy used for angioplasty. Moreover, a dose-response relationexists between the degree of thermal injury and SMC proliferative response. Hence, this technique couldbe used as a practical model of restenosis suitable for screening therapies for inhibition of SMCproliferation. (Circulation 1992;86:1249-1256)KEY WORDs * laser * injury, thermal * restenosis * cells, smooth muscle * atherosclerosis,

accelerated * neointima

R estenosis at the site of interventional proce-dures for coronary artery disease remains amajor unsolved clinical problem.' Chronic re-

coil and smooth muscle cell (SMC) proliferation haveboth been proposed as contributing mechanisms of

From the Biomedical Engineering and Instrumentation Pro-gram (P.C.D., R.F.B.), National Center for Research Resources,Cardiology Branch (P.C.D., R.N., E.F.U., M.S., S.B., S.E.E.),Ultrastructure Section (R.C., V.J.F.), Pathology Branch, NationalHeart, Lung, and Blood Institute, National Institutes of Health,Bethesda, Md., and the Cardiology Division (M.B.L.), WashingtonHospital Center, Washington, DC.

Presented in part at the American Heart Association ScientificSessions, November 1991, Anaheim, Calif.Address for correspondence and reprints: Robert F. Bonner,

PhD, Building 13, Room 3W13, National Institutes of Health,Bethesda, MD 20892.

Received January 8, 1992; revision accepted June 16, 1992.

restenosis.2 SMC proliferation is a common response toexperimental vascular wall injury,3-8 and specimensretrieved by directional atherectomy and at necropsyfrom patients dying after percutaneous transluminalcoronary angioplasty exhibit SMC proliferation at thesite of the restenosis lesion.9-1" Thus, SMC proliferationappears to be a critical component of the developmentof restenosis after balloon angioplasty as well as newerangioplasty procedures.12

Recently, laser-heated and radio frequency-heatedballoon angioplasty have been proposed as a means oftreating or minimizing dissection and elastic recoil;however, these approaches have been associated with ahigh rate of restenosis.13-19 Because thermal balloonangioplasty and pulsed laser angioplasty can createboth mechanical and thermal injury to the arterial wall,it has been difficult to determine the relative contribu-tion of each to SMC proliferation. This determination



1250 Circulation Vol 86, No 4 October 1992

would be important to better understand and therebyultimately prevent the restenosis process. We thereforesought to elucidate the role of thermal injury in the highrestenosis rates associated with thermal interventionsand to determine the limits of heating the artery wallnecessary to minimize the risk of significant SMCproliferation. Another purpose of this investigation wasto determine whether laser-induced SMC proliferationcould serve as a reliable experimental model of thecellular proliferative response to vascular injury and ofthe effects of potential therapeutic interventions.

MethodsAnimals

Fourteen New Zealand White (NZW) and sevenWatanabe rabbits weighing 2.8-3.5 kg were used in thisstudy with the approval of the National Institutes ofHealth Animal Care and Use Committee, with adher-ence to the guidelines of the Guide for the Care and Useof Laboratory Animals (Department of Health andHuman Services publication No.[NIH]86-23, revised1985). All rabbits underwent intramuscular injection of44 mg/kg ketamine HCl and 4 mg/kg xylazine forgeneral anesthesia.

Tm:YAG Pulsed Laser Thermal InjuryAccompanying Efficient Ablation

Rabbit (seven Watanabe and two NZW) aortas weresurgically exposed, and focal channels were ablatedfrom the adventitia to the lumen using a Tm:YAG laserat 2.01 ,um and 1 pulse/sec (Quantronix) deliveredthrough a 500-,um fiber at 300-600 mJ/mm2 in contactwith the descending aorta. A series of pulses weredelivered until or just before perforation. For eachanimal, 15-20 independent sites (n= 182) of focal abla-tion were created and marked with surgical sutures. Thetemporal history of the injury response was studied byhistology in the nine rabbits that were killed 2 hours to4 weeks after pulsed laser ablation.

Ho:YAG Pulsed Laser Thermal Injury WithoutMechanical Injury to Media

Over the 10-cm length used, the central ear arteriesof the 12 male NZW rabbits were easily visualized, lyingfrom 600 ,um (most proximal) to 330 gm (most distal)below the skin surface. Pulses (300 ,usec long, at 3pulses/sec) of energy from a Ho:YAG laser (SchwartzElectro-Optics) were delivered transcutaneously to a3.4-mm diameter spot on the overlying skin. This led toreproducible heating, decreasing with increasing depthof the underlying central artery. Alternatively, heatingincreased with increasing pulse energy for arteries atfixed depth. In this manner, reproducible thermal injuryto the media at a given depth could be achieved at afixed laser dose. Using a sequence of two to four laserpulses with individual pulse fluences between 120 and160 mJ/mm', the surface heating could be varied butremained below 100°C without ablative vaporization.The total energy of the laser pulses delivered wasdirectly measured. Skin surface temperature elevationinduced by the laser beam was monitored using athermographic camera (Inframetrics). Thermographicimages were recorded on videotape at 30 frames/sec for

surface of the ear (at a depth of 1.5-2.5 mm) induced bythermal diffusion from the heated surface were mea-sured with a thermocouple and recorded. Seven to nineindependent localized thermal injuries were createdalong the length of each rabbit ear central artery.Thermal lesions (n = 192) were created first distally thenproximally to avoid transient spasm at one injury sitefrom occluding flow at other sites subsequently injuredin the same artery. Between adjacent lesions there was=5 mm free of thermal injury that served as an internalcontrol. The skin was marked to indicate the exact sitesof thermal injury.The reproducibility of SMC neointimal proliferation

to a fixed thermal dose was studied in 56 lesions createdby a standard dose of 390 mJ/mm', with the rabbitskilled 21 days after thermal injury. Additionally, 56lesions created at other doses from 240 to 640 mJ/mm2were examined by histology at 21 days in order todetermine the dose response of SMC proliferation toincreasing thermal injury. Before the rabbits werekilled, macroscopic examination by transillumination ofthe ears was performed every other day after thethermal injury. The animals were euthanized with anoverdose of intravenous pentobarbital. In addition, atotal of 64 lesions were examined by histology at 2 hours(n= 16), 2 days (n= 16), 3 days (n=16), or 5 days (n= 16)after injury to study the acute and subacute effects ofthe transient thermal injury on the arterial walls.

Tissue PreparationAortas and whole ears were removed immediately

after death and fixed by immersion for at least 48 hoursin McDowell-Trump fixative.20 All lesions were easilyidentified externally and measured less than 5 mm.Transverse cuts were made through the artery and thesurrounding tissue throughout each lesion and at con-trol sites between thermally injured sites. Gross section-ing of the fixed vessels and surrounding tissue wasperformed at 0.5-mm increments perpendicular to thevessel axis before embedding in paraffin. Serial sectionswere obtained at these increments. Serial sections weremade at 100-,um increments to study the spatial distri-bution of smooth muscle proliferation. Sections werestained with hematoxylin-eosin, Movat, or Masson'strichrome stains. Samples from two additional rabbitswere embedded in epoxy resin, sectioned at 700 A, andstained according to the electron microscopic method ofKajikawa et al.2'

Area MeasurementsSMC proliferation within the internal elastic mem-

brane was quantified, whereas the amount of prolifera-tion within the injured media could not be accuratelymeasured due to varying medial necrosis. Morphomet-ric analysis of Movat-stained sections was performed bydigitizing video microscopic images and subsequentautomatic image processing (Image 1.28 on Mac II).Computerized planimetry of luminal, neointimal, andmedial areas were determined by tracing the luminalperimeter and internal and external elastic lamina witha digitizing tablet. Cross-sectional narrowing was de-fined by the ratio neointimal area to the area within theinternal elastic lamina at the lesion site. Adjacentnormal arterial sections midway (5 mm) between sitesof thermal injury were used to verify the focal nature of

subsequent analysis. Thermal transients on the opposite the biological response and as a paired reference value



Douek et al SMC Proliferation With Pulsed Lasers 1251

L

,t

-1 --r -

j '4. /.%rs.. --

A

4o ¾N%

B

FcGURE 1. Light micrographs of rabbit aorta (Movat stain). Panel A: At 2 days: focal perforation (with thrombus betweenarrows) ablated with a Tm:YAG laser (2.01 jim) from the surgically exposed adventitial surface. The ablated tissue is borderedby 200-300-gm-wide zones of thermal damage (X-X) with loss of cell nuclei in the media. Panel B: At 4 weeks: the healed siteofperforation (between arrows) was filled with new smooth muscle cells (SMCs) and associated collagen and elastin deposition.New SMC nuclei are also seen in zones of early SMC necrosis (between x's). n, Thin neointima, <40 ,um; L, lumen; m, normalmedia; a, adventitia. Magnification, x100 (bar=250 gim).

of media area in the absence of injury. All serialsections within a lesion were examined to determine thesite of maximal SMC proliferation. The reported re-sponses are the maximal observed proliferative re-sponses for each lesion.

Statistical AnalysisAll data are reported as mean±SD. Reproducibility

of response was determined by measuring the meancoefficient of variation between responses measured atequivalent sites on different ears of all animals.

ResultsRabbit Aorta

Response to efficient Tm:YAG laser ablation. Thethreshold for single-pulse ablation of aorta wall by theTm:YAG laser (2.01 jim) was determined to be ':'260mJ/mm2 both in vitro and in vivo when using 500-jmoptical fibers in contact with tissue. Using energies from1.5 to 3 times this threshold, small (500 gm) perforatingholes (n= 163) were ablated inward from the surgicallyexposed adventitial surface of the descending aorta withuninterrupted flow. Acutely (n-45) and at 12 hours(n=33) after ablation, a 200-300-jm-wide zone of mildthermal damage to both the media and adventitiasurrounded the small perforating channel occluded bythrombus (Figure 1). At 1 week (n=33) and 2 weeks(n=18), necrosis of the thermally damaged layer wasaccompanied by initiation of cellular proliferationwithin the injured site. In all 53 sites examined at 4weeks, a proliferative response had filled the ablatedmedia with minimal neointimal growth (Figure 1). SMCproliferation did not significantly reduce luminal areabut frequently extended into the void created by abla-tion of adventitia with significant deposition of newelastin and collagen.

Central Artery of the Rabbit EarThermal transients induced by nonablative Ho:YAG

laserpulses. Two Ho:YAG laser pulses depositing a total

of 390 mJ/mm2 on a 3.4-mm spot induced a brieflocalized thermal elevation peaking at 67°C above am-bient at the irradiated surface followed by a complexcooling time course resulting from thermal diffusion todeeper layers and water evaporation at the heatedsurface (Figure 2). On the opposite side of the ear 1.5mm from the irradiated surface, the measured peaktemperature elevation was only 6±3°C and was delayed~'2 seconds after the laser pulses with a subsequent slowdecay ("'-10 seconds). The estimated peak transienttemperatures at 330-600 jim, the depths of the mostsuperficial portions of the central artery, were only

100

00

S._

0

w1*m

0.E0

100 400 800 1200 1600 2000

time (msec)FIGURE 2. Graph shows temperature elevation after trans-cutaneous irradiation of the central artery of the rabbit ear bytwo Ho:YAG laser pulses (each 190 mJ/mm2) (o) at skinsurface measured by infrared video camera. Estimates ofthermal transients within the tissue at a depth of 330 jim (*)and 600 jm (+) show reductions in peak temperature withincreasing depth.

-11-




BA

t FFIGURE 3. Light micrographs of central artery of rabbit ear after Ho:YAG transcutaneous irradiation. Panel A: Cross section(hematoxylin-eosin stain) of entire rabbit ear 24 hours after standard dose (390 mJ!mm2). Panels B and C: 400-gm-deep artery2 days after low dose (panel B, 240 mJ/mm2) and high dose (panel C, 640 nJ/mm2). Panels D-G: 420-,um-deep artery 21 daysafter low dose (panel D), high dose on paired site (panel E), standard dose on deeper artery (600 ,um) (panel F), standard doseon more superficial artery (330 ,gm) (panel G). Panels B-G, Movat stain. L, lumen; N, neointima; M, media; iem, internal elasticmembrane; x, medial necrosis. Magnification panelA, x50; panels B-G, x200; bar=250 pim. Coagulation ofsuperficial dermallayer (-250 gm) was observed in panelA with inflammatory cells (ic) infiltrating deeper layers andpronounced nuclear dropout(x) in superficial half of media (up to 800 gim deep). Complete medial necrosis (C) when entire media heated to >30°C led tomassive smooth muscle cell proliferation by 3 weeks (panels E and G), whereas more limited medial injury (panels A and B)induced less neointimal proliferation (panels D and F).




G

around 25-35°C above body temperature and lasted foronly a few seconds. When the delivered pulse energywas increased, the thermal transients increased propor-tionally, which allowed study of variation in the focalarterial response to increasing temperature transients.Acute tissue responses to thermal transients. Immedi-

ately after laser irradiations .240 mJ/mm2, focal spasmof the artery was observed at heated sites and sponta-neously disappeared within a few minutes. Desiccationof the epidermis was also observed. One day after thethermal injury (Figure 3A). dermal edema was associ-ated with marked diffuse inflammation in a band 2-3cm wide along the injured central artery and withvasodilation of the collateral vessels. This marked in-flammation resolved in 5-7 days. By 3 weeks, the sites ofdermal injury had healed and appeared nearly normal.Two to 24 hours after the procedure, Movat, hematox-ylin, and eosin-stained sections of fixed tissue demon-strated superficial desiccation of the epidermis withouttissue ablation or carbonization (Figure 3A). A super-ficial zone of "coagulation" was observed within the first250 jxm from the irradiated surface with increasedeosinophilia and pyknotic nuclei in deeper layers. Thecoagulation was most severe and deepest at the centerof the laser spot but did not reach the central artery atthe laser irradiation doses used.One to 2 days after thermal injury, loss of nuclear

staining of the SMCs in the central artery was observedin histological sections of the center of the lesion(Figure 3). Medial necrosis frequently appeared gradedwithin sections, with the most complete loss of SMCnuclei at the most superficial layer and little or no lossof nuclei in the media on the deepest portion of thevessel. SMC damage was evident in the sites of injuryexamined by electron microscopy (Figure 4). At lesions,SMC were enlarged, showing vacuolization of the cyto-plasm and a damaged marginated or clumped chroma-tin. In some cases, it was difficult to delineate thebasement membrane. Internal elastica persisted andwas clearly demonstrated by Kajikawa's stain (Figure 4).For lesions created in the more superficial, distal centralartery, medial necrosis at 1-2 days was frequentlycomplete at the center of the lesion but decreased inserial sections near the periphery of the lesion. Sections5 mm on either side of the thermal lesions appeared

FIGURE 4. Electron micrograph of thermally injured arteryat a depth of 400 gm 2 days after Ho:YAG laser surfaceirradiation (390 m.J/mm2). Smooth muscle cells (smc) andendothelial cells (ec) exhibited vacuolization of the cytoplasmand clumped chromatin. Interstitial edema (ie) was observed,whereas the internal elastic membrane (iem) was intact.Magnification, x 6,600.

normal with no histological sign of thermal injury. Byelectron microscopy, SMCs at these control sites ap-peared normal with well-delineated basement mem-brane, actin filaments, and peripherally located densebodies. Similarly, optical microscopy of serial sectionsshowed a clear transition to viable undamaged media.Thus, mild, short-lived elevation of temperature (to65-75°C for a few seconds) in the artery walls createdan injury to the SMC that led to substantial necrosisobservable after only 2 days. A severe local inflamma-tory infiltrate in the interstitial tissue surrounding theartery (Figure 3) peaked at 2 days and then decreasedrapidly. Inflammatory cells were also seen in the dermisin adjacent normal control sections.

Proliferating or newly migrated SMCs were detectedin the injured media and in the intima 3-5 days afterlaser thermal injury. Neointimal proliferation was morepronounced on the superficial side of arteries when lossof SMC nuclei was much greater superficially and thusappeared to reflect the thermal gradient. On the otherhand, in cases exhibiting complete medial necrosis atthe center of the lesion, early SMC proliferation andmigration were confined to noncentral sections corre-sponding to sites with some viable SMCs remaining.




680

550

C14

EE

0

0CD40

0

U)

420

290

1 60

30 1 7000 20 40 60 80 100

% Cross Sectional Narrowing

FIGURE 5. Graph of mean cross-sectional narrowing(+SD) of the central artery induced by transcutaneousHo:YAG laser irradiation dependent on both depth of arteryand laser dose. Increased surface fluence for constant (420gtm) arterial depth (0) and decreased depth of superficialmedia at the standard dose (o) both correlated with narrow-ing at 3 weeks (R=0.84 and 0.94, respectively). Equivalentpercent cross-sectional narrowing can be obtainedfrompairedleft and right y-axis values (e.g., standard dose on 600-jutm-deep artery read on left y-axis is equivalent to 160 glJ/mm2 atthe average depth of 420 ,um read on right).

Chronic tissue responses to thermal transients. Threeweeks after laser thermal injury at the standardizeddose (390 mJ/mm2), the artery typically showed a pali-sade of proliferating cells within the neointima (Figure3). At this time, the mean cross-sectional area of thelesion sites (selected for maximal neointima) averaged55+± 10% of that at paired, unaffected adjacent sitesbetween two lesions. At sites of medial injury, prolifer-ation did not restore media mass external to the internalelastic lamina but rather was manifested principally bythe thickened neointima (Figure 3). After more severemedial injury in some distal lesions, complete focalmedial necrosis without replacement by the prolifera-tive response led in some cases to complete loss oflayered structure of the vessel at the center of thelesion. At the edges of these distal lesions, medial losswas less severe, and extensive neointimal smooth muscleproliferation was observed. Proliferating or migratingcells within the media and neointima were identified byelectron microscopy as SMCs, differing from normalSMC in the media by their larger size, stellate shape,and haphazard orientation. SMCs were often observedpassing through clefts of internal elastic membrane(Figure 3). An enlarged intracellular space containingnewly formed collagen fibrils and elastin fibrils wasoften observed within the neointima. At 3 weeks, theendothelial layer was regenerated at thermal lesionsites. At the skin surface, exuberant regeneration ofepithelial cells destroyed by the thermal injury resultedin a two- to threefold thicker skin epithelium at lesionsites.

Quantitative morphometry of lesions at 3 weeks aftera standardized dose (390 mJ/cm2) showed a meancross-sectional narrowing of 59+17% resulting from amean neointimal area of 0.040+0.020 mm2. Neointimalproliferation was observed in 90% of the treated sites

(50 of 56). In four of six sites not exhibiting neointimalproliferation, the epidermal/dermal injury as delineatedby epithelial hypertrophy was not centered over theartery, indicating that the laser beam had been mistar-geted. Maximal neointimal proliferation expressed aspercent cross-sectional narrowing was greater in distal(more superficial) arterial lesions than in proximal(deeper-lying) arterial segments (Figures 3 and 5). Themean percent stenosis induced by this standard dosewas correlated (r=0.94) with the mean depth of theartery (i.e., minimal distance of the media) below theirradiated surface. Comparison between ears andamong different animals of responses in arterial seg-ments at the same distance from the bifurcation of themajor proximal branch of the central artery (i.e., thesame approximate size and depth) showed highlyreproducible proliferative responses to a standardlaser dose (390 mJ/cm2) with a mean coefficient ofvariation of 24%.

Correlation of Neointimal ProliferationWith Thermal DoseThe Ho:YAG laser pulses induced localized thermal

injury to media 330-600 gm from the unablated surfaceand triggered SMC neointimal proliferation dependenton the laser dose or energy deposited (Figures 3 and 5),which corresponded to peak temperatures induced inthe media. Mean neointimal area increased from0.02±0.009 mm2 to 0.085±0.01 mm2 and mean cross-sectional narrowing increased from 34±10% to85±15% as fluence increased from 240 mJ/cm2 to 640mJ/cm2. Similarly, cross-sectional narrowing caused bySMC neointimal proliferation increased from 20±10%to 77±17% at standardized dose as the depth of themost superficial arterial media decreased from 600 to330 ,um. There was a strong correlation between thecross-sectional narrowing and both the fluence (r=0.84)and the depth of the artery from the irradiated surface(r=0.94).

DiscussionContinuously heated hot-tip and laser balloon angio-

plasties deposit large amounts of energy that createsubstantial thermal damage within the vessel wall.These forms of thermal angioplasty, when applied clin-ically, have been associated with high restenosis rates.18Pulsed lasers have been proposed to provide preciseablation while minimizing thermal injury,22,23 particu-larly when tissue absorption is high and pulse energy islow. However, in this report, Tm:YAG laser pulsescausing efficient ablation of the artery wall invariablywere associated with an adjacent 200-300-,um-thickzone of thermal injury, leading to loss of medial cellnuclei by 2 days and to focal SMC proliferation within1-4 weeks. Ho:YAG laser pulses are less locally ab-sorbed and created similar thermal injury to the mediaat greater distances (up to 1,000 ,um), which led tosubstantial intraluminal SMC proliferation.Because Ho:YAG laser irradiation was applied exter-

nally without contact or vaporization, injury to themedia of the central artery of the rabbit ear was purelythermal and without mechanical disruption of endothe-lium or vessel wall. Therefore, we conclude that thermalinjury alone is a potent stimulus for neointimal prolif-eration. Furthermore, a clear dose response was ob-

R 08

R=0 .84

R=0.94




served in which percent cross-sectional area narrowingcorrelated with total energy deposited (r=0.84) as wellas with the depth of arterial wall (r=0.94) from theirradiated surface of the rabbit ear. This dose responseappears to be related to the amount of the media in the:3-mm vessel segment underlying the irradiated skinthat was critically injured, exhibiting nonstructural pro-tein denaturation and subsequent dropout of medialcell nuclei.The appearance of significant thermal injury, even

when using precisely absorbed and efficient pulsedlasers, is a predictable consequence of the lower energy( 120 mJ/mg tissue) required to denature nonstructuralproteins24 compared with the 16-fold larger energies(%=2,000 mJ/mg) required to vaporize tissue.23 The heatin vapor products is not reliably evacuated or flushedaway in current forms of laser angioplasty but rather islocally released into surrounding tissues as the vaporcondenses. Therefore, the tissue volume exhibiting crit-ical thermal injury (230°C rises) during pulsed laserangioplasty can be expected to be as large as -'16 timesthe actual volume of tissue vaporized. This may explainthe observed thermal injury reported after experimen-tal excimer laser angioplasty even at low average pow-ers.25 Thermal injury may be reduced by rapidly con-vecting heat away by flushing with cooler fluids (i.e.,blood or saline at body temperature). In both of ourmodels, flow within the artery was maintained duringthe brief deposition of energy by extravascular ratherthan intravascular irradiation. This blood flow providedconvective cooling that decreased both the peak tem-peratures and duration of heating within the mediallayer, thereby reducing the amount of thermal injury.Our observations emphasize the risk of thermal in-

jury of the arterial wall and suggest that the rate ofenergy deposition ought to be limited in new interven-tional procedures. For example, current Ho:YAG cor-onary laser angioplasty catheters operate at 2-3 W(400-600 mJ/pulse at 5 Hz) and advance at an averagerate of ~=0.5 mm/sec.26 In this case, the heat diffusinginto the artery wall from the ablated surface will beroughly equivalent to 400 mJ/mm2, and, based on ourresults, should create thermal injury to media within=700 gm sufficient to cause medial necrosis and neointi-mal proliferation. The stimulus for SMC proliferationmight be confined to somewhat smaller volumes byreducing the average power delivered (e.g., active areaof catheter or pulse rate) or the duration of lasingsequences. However, such modifications run counter tothe concept of using a single catheter that can effectivelytreat (ablate) a large obstructing tissue volume in asingle pass.

Clinical restenosis rates after laser balloon angio-plasty17 increased from 36% to 67% with increasinglaser energy from 250 to 450 J. Similarly, in models ofvascular mechanical injury using oversized balloon in-flations, the degree of endothelial denudation and me-dial injury correlated with the extent of neointimalthickening.8'12'27 In general, reduction of the extent ofmedial cell injury during interventions may result inreduced SMC proliferation. Experimental and patho-logical studies have suggested that arterial vessel heal-ing responses occur regardless of the form of arterialinjury-perhaps by stimulating the expression of avariety of different growth factors.'1228

In the rabbit ear model, the rate and magnitude ofneointimal growth induced by thermal and mechanicalinjury were remarkably similar. For crush injury, thisneointimal growth was shown to correspond to nucleo-side incorporation in locally proliferating SMCs.6 Se-vere medial necrosis and massive infiltration of poly-morphonuclear cells in the surrounding dermis wereobserved shortly after thermal but not mechanical inju-ry.6 Thrombus formation was minimal at sites of ther-mal lesions, whereas platelet deposition and thrombusformation at the site of crush injury were much moreprevalent.29 Although previous studies30 have providedevidence that thermal treatment of the luminal surfaceat 70-90°C reduces the thrombogenicity of exposedmedia, this effect does not appear to prevent neointimalSMC proliferation in either clinical studies or in ouranimal model. Chronic intimal hyperplasia has beenfound to be equivalent after laser-welded or suturedend-to-end microvascular anastomoses of rat femoralarteries3' and equivalent in normal rabbit aorta afterthermal/mechanical injury by electric spark erosion,thermal/mechanical injury by Nd:YAG laser catheter,or mechanical injury by oversized balloon catheters.32'33

Prevention of restenosis after angioplasty and othervascular interventions may require reduction of injury or,alternatively, reduction in SMC proliferation by adjunctivepharmacological therapy. Although many inhibitors ofSMC have been proposed, their efficacy and/or dose-response relation has been difficult to quantify in animalmodels of restenosis. The thermally injured rabbit earartery model may be particularly useful for screening avariety of proposed inhibitors. Many independent sitescan be easily and reliably created in a single animal, andthe proliferative response is highly reproducible (meancoefficient ofvariation is 24%). Because graded injury andproliferative responses are easily induced in our model, itmay be used to study efficacy of drugs on lesions ofincreasing medial injury as well as potential complications(e.g., aneurisms) associated with excessive inhibition ofvascular healing after severe medial injury. This simple,noninvasive, small-animal model is easily accessible tolocal therapies and allows dose response and controlstudies in the same animal; therefore, it may be particu-larly useful as a screening model of therapy for humanrestenosis.

References1. Holmes DR, Vliestra RE, Smith HC, Vetrovec GW, Kent KW,

Cowley MJ, Faxon DP, Gruentzig AR, Kelsey SF, Detre KM,Van-Raeden M, Mock MB: Restenosis after percutaneous trans-luminal angioplasty (PTCA): A report from the PTCA registry ofthe National Heart, Lung, and Blood Institute. Am J Cardiol 1984;53:77-81C

2. Block PC: Restenosis after percutaneous transluminal angioplasty:Anatomiic and pathophysiological mechanism: Strategies for pre-vention. Circulation 1990;81(suppl IV):IV-2-IV-4

3. Schwartz RS, Murphy JG, Edwards WD, Camrud AR, Viestra RE,Holmes DR: Restenosis after balloon angioplasty: A practicalproliferative model in porcine coronary arteries. Circulation 1990;82:2190-2200

4. Steele PM, Chesebro JH, Stanson AW, Holmes D, Dewanjee MK,Badimon L, Fuster V: Balloon angioplasty: Natural history of thepathophysiological response to injury in a pig model. Circ Res1985;57:105-112

5. Faxon PD, Weber VJ, Haudenschild C, Gottsman SB, McGovernWA, Ryan TJ: Acute effects of transluminal angioplasty in threeexperimental models of atherosclerosis. Arteriosclerosis 1982;2:25-133




6. Banai S, Shou M, Correa R, Jaklitch MT, Douek PC, Bonner RF,Epstein SE, Unger EF: Rabbit ear model of injury-induced arterialSMC proliferation: Kinetics, reproducibility, and implications. CircRes 1991;69:748-756

7. Sanborn TA, Faxon DP, Haudenschild C, Gottschalk SB, Ryan TJ:The mechanism of transluminal angioplasty: Evidence for forma-tion of aneurysms in experimental atherosclerosis. Circulation1983;78:654-660

8. Sarembock IJ, LaVeau PJ, Sigal SL, Timms I, Sussman J, Hau-denschild C, Ezekowitz MD: Influence of inflation pressure andballoon size on the development of intimal hyperplasia after bal-loon angioplasty: A study in atherosclerotic rabbit. Circulation1990;80:1029-1040

9. Austin GS, Ratliff NB, Hollman J, Tabei S, Phillips DF: Intimalproliferation of SMCs as an explanation for recurrent coronaryartery stenosis after percutaneous transluminal angioplasty. JAmColl Cardiol 1985;6:369-375

10. Johnson DE, Hinohara T, Selmin MR, Braden LJ, Simpson JB:Primary peripheral arterial stenoses and restenoses excised bytransluminal atherectomy: A histopathologic study. JAm Coil Car-diol 1990;15:419-425

11. von Polnitz A, Backa D, Remberger K, Hofling B: Restenosis afteratherectomy shows increased intimal hyperplasia as compared toprimary lesions. J Vasc Med Biol 1989;1:283-287

12. Ip JH, Fuster V, Badimon L, Badimon J, Taubman MB, ChesebroJH: Syndromes of accelerated atherosclerosis: Role of vascularinjury and smooth muscle proliferation. JAm Coil Cardiol 1990;15:1667-1687

13. Jenkins RD, Spears JR: Laser balloon angioplasty: A newapproach to abrupt coronary occlusion and chronic restenosis.Circulation 1990;81(suppl IV):IV-101-IV-108

14. Mitchel JF, Fram DB, Fisher JR, Sanzobrino BW, Maffucci LM,Dyckman WP, Gillam LD, Abele JE, McKay RG: Low-grade(60°C) heating increases vascular compliance during balloon angio-plasty. (abstract) Circulation 1991;84(suppl II):II-1196

15. Buller CE, Culp SC, Davidson CJ, Tenaglia AN, Phillips HR, StackRS: Thermal VS conventional perfusion balloon angioplasty in anin vivo atherosclerotic model. (abstract) Circulation 1991;84(supplII):II-1 199

16. Lee BI, Becker GJ, Waller BF, Barry KJ, Connolly RJ, Kaplan J,Shapiro AR, Nardella PC: Thermal compression and molding ofatherosclerotic vascular tissue with use of radiofrequency energy:Implications for radiofrequency balloon angioplasty. J Am CoilCardiol 1989;13:1167-1175

17. Spears JR, Reyes VP, Wynne J, Fromm BS, Sinofsky EL, AndrusS, Sinclair IN, Hopkins BE, Schwartz L, Aldridge HE, PlokkerHWT, Maste EG, Rickards A, Knudtson ML, Sigwart U, DearWE, Ferguson JJ, Angelini P, Leatherman LL, Safian RD, JenkinsRD, Douglas JS, King SB: Percutaneous coronary laser balloonangioplasty: Initial results of a multicenter experience. JAm CollCardiol 1990;16:293-303

18. Reis GJ, Pomerantz RM, Jenkins RD, Kuntz RE, Baim DS, DiverDJ, Schnitt SJ, Safian RD: Laser balloon angioplasty: Clinical, angio-graphic and histologic results. JAm Coll Cardiol 1991;18:193-202

19. Rothbaum D, Linnemeier T, Landin R, Ball M, Hodes Z, RiddellR, Morgan S: Excimer laser coronary angioplasty: Angiographicrestenosis rate at six month follow-up. (abstract) JAm Coll Cardiol1991;17:205A

20. McDowell EM, Trump BF: Histologic fixatives suitable for diag-nostic light and electron microscopy. Arch Pathol Lab Med 1976;100:405-414

21. Kajikawa K, Yawmaguchi T, Katsuda S, Miwa A: An improvedelectron stain for elastic fibers using tannic acid. J Electron Microsc(Tokyo) 1975;24:287-288

22. Grundfest WS, Litvack F, Forrester JS, Goldenberg T, Swan HJC,Morgenstern L, Fishbein M, McDermid IS, Rider DM, Pacala TJ,Laudenslager JB: Laser ablation of human atherosclerotic plaquewithout adjacent tissue injury. JAm Coll Cardiol 1985;5:929-933

23. Leon MB, Smith PD, Bonner RF: Design considerations for laserangioplasty, in Moore WS, Ahn SS (eds): Endovascular Surgery.Philadelphia, WB Saunders, 1989, pp 363-376

24. Geoffrey A, Gardiner P Jr, Consigny M: Effects of thermal energyon the arterial wall. Semin Int Radiology 1991;8:94-99

25. Hanke H, Haase KK, Hanke S, Oberhoff M, Hassenstein S, BetzE, Karsh KR: Morphological changes and smooth muscle prolif-eration after experimental excimer laser treatment. Circulation1991;83:1380-1389

26. Geschwind HJ, Dubois-Rande JL, Zelinsky R, Morelle JF, Bous-signac G: Percutaneous coronary mid-infrared laser angioplasty.Am Heart J 1991;122:552-558

27. Webster MW, Chesebro JH, Heras M, Mruk JS, Grill DE, FusterV: Effect of balloon inflation on smooth muscle cell proliferation inthe porcine carotid artery. (abstract) J Am Coll Cardiol 1990;15:165A

28. Waller BF, Pinkerton CA, Rothbaum DA, Linnemeier TJ, OrrCM, VanTassel JW, Slack JD: Restenosis tissue following hot-tiplaser, excimer laser, primary atherectomy, and angioplasty proce-dures: Histologically similar intimal proliferation in 33 atherec-tomy patients. (abstract) Circulation 1990;82(suppl II):II-1238

29. Silver MJ, Ingerman-Wolenski CM, Sedar AW, Smith M: Modelsystem to study interaction of platelets with damaged arterial wall:I. Inhibition of platelets adhesion to subendothelium by aspirinand dipyridamole. Exp Mol Pathol 1984;41:141-152

30. Spears JR, Kundu SK, McMath P: Laser balloon angioplasty:Potential for reduction of the thrombogenicity of the injured arte-rial wall and for local application of bioprotective materials. JAmCoil Cardiol 1991;17:179B-188B

31. Serure A, Withers EH, Thomsen S, Morris J: Comparison ofcarbon dioxide laser assisted anastomosis and conventional micro-vascular sutured anastomosis. Surg Forum 1984;34:634-636

32. Oomen A, Van Erven L, Vandenbroucke W, Verdaasdonk RM,Slager CJ, Thomsen SL, Borst C: Early and late arterial healingresponse to catheter-induced laser, thermal, and mechanical walldamage in the rabbit. Lasers Surg Med 1990;10:363-374

33. Consigny PM, Gardiner GA Jr: Effects of balloon angioplasty andlaser-assisted balloon angioplasty on atherosclerotic rabbit arter-ies. (abstract) Radiology 1990;174:1071



Dose-dependent smooth muscle cell proliferation induced by thermal injury with pulsed infrared lasers

Documents