Effect of blood upon the selective ablation of atherosclerotic plaque witha pulsed dye laser

Lasers in Surgery and Medicine 10533-543 (1990)

Effect of Blood Upon the Selective Ablation of Atherosclerotic Plaque With a

Pulsed Dye Laser Kenton W. Gregory, MD, Martin R. Prince, MD, Glenn M. LaMuraglia, MD,

Thomas J. Flotte, MD, Lisa Buckley, BS, Jill M. Tobin, BS, Andrew A. Ziskind, MD, John Caplin, MD, and R. Rox Anderson, MD

Wellman Laboratory of Photomedicine, Massachusetts General Hospital, Boston, Massachusetts 021 14

Laser angioplasty systems with laser energy preferentially absorbed by atherosclerotic plaque may offer a safe method of plaque removal. This study evaluated the effect of blood upon selective energy absorption using a pulsed dye laser at 480 nm. Intra-arterial laser irradiation of normal rabbit femoral arteries demonstrated a perforation threshold energy with blood perfusion of 13.1 m J per pulse compared to 87.9 m J with saline (P < .0001), indicating a deleterious effect in the presence of blood. An adverse effect upon arterial healing at 3 days was noted in sheep following intra-arterial irradiation during blood but not saline perfusion. Normal and atherosclerotic human aorta ablation thresholds differed significantly (P < .0002) under saline (plaque: 20 mJ and normal: 120 mJ) but the difference under blood (plaque: 5 m J and normal: 20 mJ) was not significant. We con- clude that absorption of laser energy by blood can reduce the effect of differential absorption by endogenous chromophores in normal and pathologic vascular tissues and, therefore, removal of blood may be a prerequisite for selective ablation of atherosclerotic plaques.

Key

INTRODUCTION

words: laser angioplasty, laser atherectomy, pulsed dye laser ablation, hemoglobin absorption

The clinical use of lasers for removal of atherosclerotic obstructions from arterial lumens continues to be limited by the concern of inadver- tent destruction of normal tissue. One method of potentially minimizing laser damage to normal tissue is to utilize laser radiation that is absorbed preferentially by plaque. A wave band from 450 to 500nm has been identified where absorption by atherosclerotic plaque is substantially higher than that of normal artery due to the presence of yellow carotenoid pigments in plaqueJ21 In this waveband, short pulses of laser radiation of suffi- cient intensity to ablate plaque under saline do not result in ablation of normal arterial wall [3,41. Preliminary results in clinical trials have suggested, however, that the selective advantage demonstrated in vitro may not always extend to

the in vivo setting. Damage to normal portions of vessel walls including dissection and perforation have been noted angiographically [5-91. While these findings may be due to mechanical effects of the delivery device, laser-tissue interactions not predicted by in vitro models may also be a plau- sible explanation. Studies of microvessel laser irradiation in hamster cheek pouches have revealed significant damage to normal vessel

Accepted for publication July 10, 1990. Address reprint requests to Kenton W. Gregory, M.D., Well- man Laboratory of Photomedicine, Wellman-2, 50 Blossom Street, Massachusetts General Hospital, Boston, Massachu- setts, 02114. This work was supported by the National Institutes of Health contract SBIR 2 R44 HL 37180-02 and by the SDIO MEFL program under contract N00014-K0117.

0 1990 Wiley-Liss, Inc.

534 Gregory et al. resulting from 577 nm light absorption by hemoglobin [13-151. From the absorption spectra of hemoglobin [1,16], it is known that blood also has a high absorption coefficient at 480 nm and that absorption of continuous wave argon laser radiation at 488 nm by blood can result in thermal damage to adjacent vessel wall 117-191. While very short pulse-duration lasers, such as the 1 ps pulse duration dye laser, result in minimal thermal injury to surrounding structures, it is possible that blood absorption of 480 nm radiation may result in secondary non-thermal damage to normal artery despite negligible light absorption by vessel wall at this wavelength.

The purpose of this study was to compare in vitro and in vivo ablation thresholds of arterial tissues in blood and saline fields to determine the feasibility of selective intra-vascular plaque ablation of laser energy in a blood field. In vitro mea- surements of the ablation thresholds of normal and atherosclerotic aorta were compared when exposed to 480 nm 1 ps radiation immersed in saline or blood to determine the effect of blood upon selective ablation of pathologic tissues. In order to validate this work in an in vivo setting, perforation thresholds of rabbit femoral arteries were determined when these vessels were exposed to 480 nm 1 ps radiation while perfused with ei- ther saline or whole blood. Additionally, sheep carotid vessels were exposed to 480 nm 1 ps radiation during saline or blood perfusion and the subacute vascular inflamatory responses were compared.

MATERIALS AND METHODS Laser

Radiation at 480 nm was obtained from a flashlamp excited dye laser (Candela Laser Corp, Wayland, MA) The pulse width was 1 p s (full width, half maximum) at a pulse repetition rate of 2 Hz. A 320 pm core diameter step-index quartz optical fiber was coupled to the laser with a 37 mm focal length quartz lens. For in vivo experiments, the distal end of the fiber was fused into a sphere approximately 450 pm in diameter to reduce mechanical injury or perforation. The full divergence angle of the beam from the fiber was approximately 1 lo, which increased the beam diameter of the sphere-tipped fiber 9 pm and produced a 329 pm diameter spot size. For in vitro experiments, radiation was delivered with a sim- ple cleaved fiber. Distal optical fiber average en-

ergy output was measured with a power meter at 2 Hz (Scientec model 395).

In Vivo Perforation Threshold To determine the in vivo effect of 480 nm

radiation to normal arteries, 25 New Zealand white rabbits (4-5 kg) were anesthetized with in- tramuscular ketamine and xylazine and the femoral arteries were exposed by dissection. To minimize perturbations in blood flow and undue mechanical damage, the optical fiber was intro- duced into the femoral artery via the hypogastric branch and placed against the internal wall, perpendicular to the long axis of the femoral artery. A perpendicular orientation of the fiber relative to the vessel wall ensured a uniform and repro- ducible radiant exposure of the vessel wall. Hep- arin, 100 units/kg, was administered intrave- nously. The placement was confirmed using a dissecting microscope (Bausch and Lomb). A 37°C saline bath was formed around the femoral artery to facilitate ultrasonic imaging and to maintain a normal thermal environment. Ultrasonic imaging was obtained with a 12 MHz transducer (Bio- sound Surgiscan, Indianapolis, IN) placed above the artery within the saline bath, and images were recorded on videotape. Systemic arterial pressure was monitored via a brachial artery can- nulation and femoral artery flow monitored via a 2 mm doppler flow probe (Transonics TlOl Flow meter, Transonic Systems, Ithaca, NY) placed 2 cm distal to the optical fiber. A more proximal femoral artery side branch was cannulated with a lmm diameter silastic tube and connected to a microinfusion pump for saline infusion. The proximal femoral artery was occluded with vascular clamps when saline was perfused at a rate comparable to the blood flow rate measured prior to clamping the vessel. Femoral arteries were exposed to 20 pulses of laser radiation at each energy level beginning at 2.5 mJ per pulse, 5 mJ per pulse, and then increasing at increments of 10 mJ, until perforation was observed. The preparation was viewed with the dissecting microscope positioned above the apparatus after each exposure. Perforation resulted in the echogenic presence of extravascular bubbles and blood in the saline bath and was confirmed by observing extravascular protrusion of the fiber following laser irradiation. (Fig. 1) Following irradiation, the vessels were removed for histologic analysis. Two arteries were exposed to 20 pulses at 15 mJ with saline infusions and then removed prior to perforation for histologic comparison.

Effect of Blood On Ablation of Plaque 535 In Vivo Ablation Study

Six female sheep (35 kg) were anesthetized with 2% halothane prior to surgical exposure of the distal superficial femoral arteries or proximal common carotid arteries through 3 cm incisions. After systemic heparinization (1,000 unitdkg), proximal and distal control of the artery was obtained for a 3 mm longitudinal arteriotomy to in- troduce a 400 pm diameter fiber optic device with a n outward-tapering spherical tip which allowed beam divergence to provided a 2.0 mm diameter laser spot size at the surface of the ball tip. Co- axial irradiation of the artery was achieved after re-establishment of blood flow for a blood field or with irrigation of normal saline for a saline field. The energy per pulse was 300 m J (Fluence 9 J/ cm2). Total energies of 15 J, 3 J, and 0.6 J were delivered to the vessel at 1 cm intervals at 3 Hz. The arteriotomies were transversely repaired with prolene sutures and the wounds surgically closed. Six animals were sacrificed acutely or at 3 days, whereupon the arteries were harvested and examined under light microscopy with hematoxylin-eosin and Voerhoff elastin stains.

In Vitro Ablation Threshold

To measure ablation thresholds for human aorta, 613 spots were irradiated on specimens of aorta obtained prior to fixation from 18 cadavers within 24 hours of death. The tissue was graded as normal, soft plaque, or calcified plaque based upon appearance. A determination of normal tissue was made if the area of exposure was com- pletely free of atheroma visually and there was no evidence of intimal thickening. Atheromatous aortic specimens were characterized by grossly elevated intimal thickening, presence of yellow pig- mentation, and/or presence of calcification. These specimens were divided into 4 x 4 cm portions and pinned to paraffin slabs. Marker pins were placed at 1 cm intervals before submerging the specimens under a 2 cm layer of heparinized blood, leaving the marker pins exposed. The marker pins were used to direct the laser fiber to an es- tablished position under blood utilizing a process of triangulation between pins. A 320 pm diameter cleaved fiber was positioned flush against the surface of the aortic specimen. A row of laser exposures were made at energy increments between 2.5 and 80 m J per pulse (fluences 3-100 J/cm2) with 40 pulses delivered at each location. Tactile fiber recoil or an acoustic report produced during the energy delivery was noted when it occurred.

Following laser irradiation, the specimen was removed and carefully washed with normal saline while leaving the marker pins in place. A stereo dissecting microscope was then used to scan the specimen using the same process of triangulation to locate the site of optical fiber placement. An average threshold for ablation was defined, con- sidering the heterogeneity of tissue pigments, to be the energy level at which tissue removal (crater formation) was observed in 50% of the laser exposure sites. To compare the mechanical effect of fiber placement, the fiber was placed adjacent to the exposed area for 20 seconds at similar pressure and observed for effect. The tissue effect was observed and noted to have no effect or evidence of tissue removal by a surface defect or crater formation. This experiment was performed on 14 specimens under blood and 8 specimens under saline. Whenever possible, the same specimen was studied under both blood and saline. Tissues were then fixed in formalin and examined under light microscopy with hematoxylin-eosin or glu- teraldehyde with buffer and studied with scanning electron microscopy.

An additional 4 samples were studied with a cleaved 320 pm diameter optical fiber positioned 200 pm above the aortic tissue. The fiber was brought down to make contact with the tissue and then elevated 200 microns above the tissue using a micromanipulator. (The spot size diameter had been determined to be 339 pm [SE = 3.311 by positioning the fiber 200 pm above an exposed Kodak photographic paper submerged in a saline bath and ten laser exposures were made at each energy level and spot diameters measured by oc- ular micrometry.) Blood or saline was poured over the tissue and laser energy was delivered at 20 and 50 mJ per pulse (fluences 25 or 62 J/cm2) for 40 pulses. The blood was then decanted off and the tissue washed with saline prior to gross, light microscopy, and scanning electron microscopic study.

STATISTICAL ANALYSIS

Mean perforation thresholds for in vivo rabbit experiments were compared using Fisher’s test for unpaired data. Perforation values are expressed as the mean rt 1 SE. Human aorta laser ablation data in proportions were transformed using an arcsin square root transformation, and energy level was transformed using a log transformation. Analysis of variance procedures in the SAS (Statistical Analysis System Program) were

536 Gregory et al.

Fig. 1. 2D ultrasonic video image of the long axis of a rabbit femoral artery demonstrating a perforation. S: Saline bath. A: Artery lumen. M: Muscle. *Extravascular cavitations. Arrow: optical fiber tip.

used to assess the effect of immersion medium, type of tissue, and pulse energy. An initial analysis of a representative subset of the data consid- ered whether the results of the procedure were different in different cadavers. This model included medium (saline or blood), tissue (normal or plaque), pulse energy and cadaver, and showed no independent intra-cadaver effect. All further analyses therefore included only the medium, log- energy level, and tissue. Thresholds for the ablation of aortic tissues were defined as the threshold energy within 95% confidence limits, at least 50% of the tissues demonstrate evidence of tissue ablation. All in vitro values are expressed as the mean 2 1 SE.

RESULTS In Vivo Perforation Threshold Study

Perforation did not occur as a result of optical fiber placement within the vessel. At all energy exposures above 2.5 mJ per pulse (3 J/cm2), irradiation in blood was associated with ultrasonic appearance of bubbles at the site of irradiation which was associated with a sharp audible report. The mean threshold for perforation with irradiation during blood flow was 13 ? 5.4 mJ per

pulse (16k6.7 J/cm2) for 20 pulses at 2 Hz. Under saline, the ultrasonic appearance of bubbles was only evident at the highest energy levels. The ap- proximate mean perforation threshold during irradiation with saline perfusing the vessel was found to be 88 * 20 mJ per pulse (109?25 J/cm2) for 20 pulses. Almost one-half of the specimens irradiated under saline did not perforate at 100 mJ per pulse (124 J/cm2), the maximal laser output for this experiment. For the purpose of comparing data, these arteries were assigned a perforation energy of 100 mJ, an effect which would minimize the significance and decrease the mean difference between the ablation thresholds. There was no overlap of perforation thresholds between vessels treated under saline or blood and the difference was highly significant (Fisher’s test for unpaired data, P< .OOOl) (see Fig. 2).

Histologic analysis of specimens ablated in a blood field revealed prominent damage to all vascular layers, demonstrating intimal tears, dissection, and a disrupted appearance suggesting a blast injury (Fig. 3). Thermal injury was not observed. At comparable energies below 40 mJ per pulse (50 J/cm2), tissues exposed under a saline field revealed negligible damage, and in most cases, tissue injury could not be detected by gross or microscopic evaluation.

Effect of Blood On Ablation of Plaque 537 120

100 E? E E

!?!

Y

80 0 r u)

r I- C 0

60

.- CI

40

20

I A Threshold I mwA

I" I A Mean 1 A A

A

0

Blood Saline

Fig. 2. Laser energy thresholds for perforation of rabbit femoral arteries perfused with blood (mean 13.1 mJ) or saline (mean 87.9 mJ). Bars indicate standard error of the mean.

In Vivo Ablation Study Mechanical or laser-induced perforation did

not occur. In a saline field, co-axial laser irradiation of sheep arteries was not accompanied by audible (popping) sounds nor recoil of the fiber. His- tologic examination of these vessels immediately after exposure and 3 days later demonstrated no evidence of endothelial or smooth muscle injury, nor disruption of the internal elastic lamina or other supporting structures. Laser irradiation performed in a blood field resulted in fiber recoil and a popping sound, suggesting that vaporiza- tion of blood elements occurred. Histologic examination immediately after exposure demonstrated hyperchromatic and pyknotic changes of some smooth muscle cells, intermittent loss of endothelial cells, rare disruption of the internal elastic laminae, and occasional dissection. By 3 days, the structural changes were similar to the acute findings; however, an almost transluminal or circum- ferential loss of smooth muscle cells with karyol- ysis was noted. There was loss of medial cells, but no inflammatory cells were observed. While elastic fibers were occasionally disrupted, loss of medial cells was present even without disruption of elastic fibers (Fig. 4).

In Vitro Ablation Threshold Laser irradiation of human aortic tissues

was observed to have an energy-dependent abla- tive effect. Below threshold energy, there was no detectable effect. When the energy per pulse was

increased above the threshold energy, an audible signal was produced, a recoil force could be felt while holding the fiber, and evidence for tissue removal was detected. The amount of tissue removed generally increased as the energy per pulse was increased above the threshold for ablation. The average energy threshold for ablation was defined as the pulse energy level at which tissue removal (crater formation) was observed in 50% of the laser exposure sites (within 95% confidence limits).

In saline, the average ablation threshold was 20 mJ per pulse (25 J/cm2) for atherosclerotic tissue and 120 mJ per pulse (149 J/cm2) for normal aorta. This difference was highly significant (P < .0002) (Fig. 5). All types of aortic tissues studied were ablated at a lower energy per pulse under a blood field as compared to a saline field (P < .OOOl) (Fig. 6). The threshold for ablating atherosclerotic tissue decreased from 20 mJ per pulse to 5 mJ per pulse (6.2 J/cm2). The threshold for ablating normal tissue was decreased from 120 mJ per pulse to 20 mJ per pulse. Therefore, ablation of normal aorta decreased much more than diseased aorta in a blood field, resulting in no practical difference in ablation thresholds for the two tissues. The average ablation thresholds were not significantly different between normal and atherosclerotic tissues under blood.

The tissues studied with the optical fiber suspended 200 pm above the aortic sample in saline and blood fields were irradiated for 40 pulses at 20 mJ (25 J/cm2) and 50 mJ (62 J/cm2) per pulse. Scanning electron micrographs of soft yellow plaque ablation comparing tissue contact under saline, tissue contact under blood and non- contact under blood are shown in Figure 7a-c. Atheromas ablated in contact with the optical fiber under saline generally resulted in round, smooth craters, removing a cylindrical volume of tissue approximately the dimensions of the 320 pm diameter fiber. Atheromas ablated in contact with the fiber under blood resulted in craters which were notable for rough, irregular borders with tissue fracturing. When ablation occurred with an intervening 200 pm blood film, a blast- type injury resulted with tissue fracturing radi- ally from a central crater with rough edges and a considerably larger amount of tissue removal. The craters produced with a thin intervening blood layer were greater in diameter (1.0-1.5 mm) and depth than those seen with direct fiber contact in blood at comparable energies. Grossly, the crater borders were frequently stained with

538 Gregory et al.

Fig. 3. Light micrograph of rabbit femoral arteries following intravascular contact radiation with 20 pulses at 15 mJ per pulse. a: With blood perfusion, demonstrating near full-thickness disruption of the vessel wall. b With saline perfusion, demonstrating no alterations. ( x 100, Trichrome stain)

Fig. 4. Light micrographs of sheep carotid arteries. a: 3 days after irradiation with blood perfusion showing loss of cells in the media. b: Normal (control) artery. x 200. Hematoxilyn and eosin.

blood or blood remnants which occurred with DISCUSSION blood field exposures and was particularly strik- ing when an intervening layer of blood was exposed, suggesting that the force of detonation drove remains of red blood cells into the tissues.

Acceptance of lasers as a useful clinical tool in the treatment of intravascular atherosclerotic obstructions awaits the development of a safe and effective means of delivery of laser energy. Fluo-

C 0 Q .- C

li a

Effect of Blood On

Normal

100

40

20

Plaque

T

n = 18 21 26 22 24 1 1 1 1 7 14 28 40 21

5 10 20 30 40 90 1 0 0 1 2 0 1 3 0 1 4 0 1 5 0 1 6 0

Energy (mJ per pulse)

Fig. 5. Normal human aorta and plaque specimens irradiated under saline in percentage of tissues with evidence of tissue ablation compared at varying laser energies. n = num- ber of laser exposures.

100

20

0

I I normal plaque i ,T

T ,

Ablation of Plaque 539 sions. In order to achieve laser energy delivery selectively to atherosclerotic lesions, adjunctive imaging with intravascular ultrasound and f luo- rescence spectroscopy are being investigated. A unique potential of lasers may be realized if wave- lengths are chosen which are selectively absorbed by chromophores present preferentially in atherosclerotic tissues. Definitive in vitro evidence to support this concept has been demonstrated [2- 41; however, subsequent evaluation in human clinical trials have not confirmed this concept [5- 91. Despite the use of a short pulsed laser at a wavelength preferentially absorbed by atheroma where normal tissues, in vitro, appeared nearly impervious to this radiation, some clinical studies have been complicated by arterial dissection and perforation. Since all in vitro studies were performed in air or saline and all clinical studies have been performed in blood, a potent absorber of radiation at this wavelength, this study was performed to re-examine the hypothesis of selective photobiologic effects of laser irradiation of normal and atherosclerotic tissues in the presence of blood.

n = 29 75 86 26 33 37 21 10

2.5 5 10 20 30 4 0 50 80

Energy (rnJ per pulse)

Fig. 6. Normal human aorta and plaque specimens irradiated under blood in percentage of tissues with evidence of tissue ablation compared at varying laser energies. n = num- ber of laser exposures.

roscopy, even with orthogonal biplane projections, is not ideal for resolving laser delivery strategies in coronary arteries, especially with complex le-

In Vivo Studies

In vivo delivery of 480 nm radiation to rabbit femoral arteries allowed the establishment of energy dosimetry related to damage of normal arterial tissue. Perpendicular placement of the fiber, while less representative of a clinical situation, was chosen to ensure a more accurate delivery of the prescribed amount of radiation. Perforation was chosen as a reliable, well-defined, and relevant marker of laser mediated vascular damage. The absorption coefficient of normal arterial wall at 480 nm has been measured to be approximately 18 cm-' (K.W.G., unpublished observation). The depth of penetration of this wavelength in arterial tissue (the inverse of the absorption coefficient), should therefore be approximately 560 pm. Since the femoral artery wall thickness in 3.5 kg rabbits in this study was 300-400 pm, it was felt that this substrate would represent a sensitive measure of light-induced vascular injury. Tissues were exposed to 480 nm radiation in saline, and no demonstrable effect was observed by gross or histologic evaluation until energy intensity reached 88 2 20 mJ per pulse (109 J/cm2). In contrast, blood flow in the vessel during energy exposure resulted in extensive injury and perforation at an average of 13 * 0.4 mJ per pulse (16 J/cm2). This latter finding occurred despite the fact that the fiber-optic was placed firmly against

Figure 7

Effect of Blood On the vessel wall where interposed blood would be expected to be minimal. The severe blast-type in- juries to arteries exposed to radiation under blood occurred at energy levels well below the predicted threshold for ablation of normal tissues in saline. Energy delivery in a blood field appears to be limited to a very narrow energy range (<16 J/cm2) before significant damage to normal vessel oc- curs, whereas delivery of energy in saline allows the operator to considerably expand the range of safely deliverable laser energies. This study resulted in concern as to the clinical feasibility of selective laser ablation of atheroma at radiant exposures currently utilized in human trials and prompted us to re-evaluate the previous threshold determinations for atherosclerotic tissues in the presence of blood.

In vivo studies of the phenomenon of blood field related vascular damage were performed in sheep under more relevant clinical circumstances for artery size and energy delivery with co-axial laser irradiation. In saline perfused conditions, no explosion or blast effect was evident nor was histologic evidence of light or blast injury present in cellular or supporting structures at up to 9 J/cm2. Following irradiation with blood perfusion, cellular injury and cell loss was evident immediately after exposure as well as at 3 days. In some cases smooth muscle cell loss was present in a circum- ferential and transmural distribution suggesting an acoustic tissue effect rather than a conse- quence of the expected distribution of light within the vessel wall. This effect in blood may result in increased acute and chronic vessel injury and, ul- timately, intimal hyperplasia; however, long- term studies will need to be performed.

In Vitro Studies The threshold energies for the ablation for

human aorta have been previously characterized for normal and atherosclerotic plaques of various compositions when performed in air or saline [2,3]. In our study, ablation threshold was defined as evidence of tissue removal or crater formation

Fig. 7. Scanning electron micrographs of in vitro laser ablation sites of human atheroma at 50 mJ per pulse for 20 pulses. a: With optical fiber in contact with the atheroma under saline showing a round, well-defined crater approximately the same size as the fiber. ~ 1 3 9 . b With fiber contact under blood with a basically round crater with some uplifted and laterally disrupted tissue. x 140. c: With the optical fiber 200 pm from the atheroma showing a blast-type injury with extensive radial fracturing of the tissue and tissue removal surrounding a central crater. x 140. Bars represent 200 pm.

Ablation of Plaque 541 after 40 pulses of laser energy observed under 40 power dissecting microscopy whereas Prince et al. examined tissues after a single laser pulse. The use of 40 pulses in this study may amplify single pulse findings in the previous study and a different spot size may also explain the difference in observed ablation thresholds. In both studies, a significantly higher energy threshold for ablation is noted for normal tissues as compared to atherosclerotic tissues. These findings support the con- tention that endogenous chromophores, identified as carotenoids, may provide the basis for selective laser ablation of pathologic atherosclerotic obstructions. However, when these data were examined in the blood field setting, where they are being applied in human clinical trials, this study demonstrated that the margin of safety, provided by a large difference between ablation thresholds, was not realized.

Ablation of normal and atherosclerotic human aorta in blood demonstrated that normal tissue is damaged at much lower thresholds than predicted by optical methods and studies con- ducted in saline fields. While this study confirmed the earlier results demonstrating a significant difference in threshold of ablation between normal and pathologic tissue in saline, the presence of blood reduced this energy difference to a point where the concept of selective ablation is not clin- ically relevant. These results indicate that the presence of even a thin film of a highly absorbing chromophore such as blood appears to reduce the effect of the endogenous chromophore difference between plaque and normal vascular tissues. Moreover, in the presence of blood, significant damage to normal vascular tissue can occur at energy intensities lower than the threshold for ablating pathologic tissues predicted under saline. For example, with 5 mJ pulses, only 10% of atherosclerotic samples were ablated when exposed under saline but in a blood field, 5 mJ pulses result in ablation in 70% of atherosclerotic tissues and 40% of normal tissues.

These results suggested that the interaction between light and blood created a powerful detonation even while the fiber was in intimate contact with the tissue. We were concerned that c h - ical circumstances would result in occasional laser activation without direct tissue contact with the delivery device and, therefore, present a much more explosive potential. The results of experiments performed with a 200 pm layer of blood interposed between fiber and tissue implied that the presence of more blood will result in more

542 Gregor tissue damage. In this series of experiments, it appeared that the ablation threshold may be decreased and greater tissue removal accomplished when a thin layer of blood is interposed between optical fiber and tissue. Compared to ablation under saline seen in Figure 7A where a well-defined cylinder of tissue was removed, ablation under blood without tissue contact resulted in a wide area of marked tissue disruption which may be undesirable in terms of acute and chronic compli- cations. With the qualitative observation that chromophores (hemoglobin staining) can be driven into the tissues, the potential for an induced increased absorption for subsequent laser pulses may exist in tissues naturally bereft of highly absorbing chromophores at 480 nm.

Pulsed dye laser energy delivery was studied previously [13-151 in hamster cheek pouch vessels as a prelude to the development of selective photothermolysis for treatment of vascular dis- eases of the skin such as port wine stains. The objective in those experiments, in contrast to laser angioplasty, was light induced destruction of normal vessels. Those studies demonstrated that exposure of microvessels to 577 nm radiation at 1-4 J/cm2 caused oxyhemoglobin mediated rup- ture and other damage to vessels followed by an intense necrotizing vasculitis, much as what we have seen in our study with laser energy at 480 nm. At 480 nm, the absorption coefficient of blood, while less than that at 577 nm, is approximately 300 cmpl 111 and the corresponding depth of light penetration is 33 pm. The absorption coefficients and depths of light penetration of normal aorta are 18 cm-', 560 pm and atherosclerotic aorta are 60 cm-' and 172 pm [21. These values indicate that, assuming the same irradiance, more energy can be absorbed by blood in a much smaller volume which could create a much larger explosive potential than absorption in vascular wall mate- rial.

IMPLICATIONS

This study demonstrated that the presence of blood, even with contact-type delivery devices, can reduce the potential for selective delivery of laser energy to atherosclerotic plaque. Removal of blood from the area of tissue irradiation may im- prove the likelihood of selective ablation of atherosclerotic plaque based upon differences in endogenous chromophores between normal and pathologic tissues.

*y et al. ACKNOWLEDGMENTS

The authors would like to give special thanks to J.A. Parrish, MD, and Reginald Birn- gruber, PhD, for their continued support and col- laboration. We appreciate the statistical analysis by Diane Finkle, PhD, and John Newell, PhD. We also thank N. Nishioka, MD; Boris Bergus, MD, for helpful discussions; and Norman Michaud, MS, and Margo Goetschkes for preparation of samples for LM and SEM examination.

REFERENCES

1. Eaton WW, Hanson LK, Stephens PJ, Dunn JBL: Optical spectra of oxy and deoxyhemoglobin. J Am Chem SOC 1978; 100:4991-5003.

2. Prince MR, Deutsch TF, Mathews-Ross M, Margolis R, Parrish JA, Oseroff A R Preferential light absorption in atheromas in vitro. J Clin Invest 1986; 78:295-302.

3. Prince MR, Deutsch TF, Shapiro AH, Margolis RJ, Ose- roff OR, Fallon JT, Parrish JA, Anderson RR: Selective ablation of atheromas using a flashlamp-excited dye laser a t 465nm. Proc Natl Acad Sci 1986; 84:7064-7068.

4. Murray A, Crocker PR, Wood R FM: The pulsed dye laser and atherosclerotic vascular disease. Br J Surg 1988; 75:

5. Leon MB, Prevosti LG, Smith PD, Bonner R, Lu LY, Mur- phy-Chutorian D, Mok W: Probe and Fire laser Angio- plasty: Fluorescence detection and selective laser atheroma ablation. Circulation 1987; Part 11: 76 409.

6. Leon MB, Almagor Y, Bartonelli LG, Teirstein PS, Chang R, Miller DL, Smith PD Bonner RF: Fluorescence-guided laser angioplasty in patients with femoropopliteal occlu- sions. Circulation 1988; 78:ll-294.

7. Geschwind HJ, Dubois-Rande JL, Bonner FR, Boussig- nac G, Prevosti LG, Leon MB: Percutaneous pulsed laser angioplasty with atheroma detection in humans. JACC 1988; ll:107A.

8. Geschwind HJ, Dubois-Rande JL, Shafton EP, Boussig- nac G, Wexman MP: Spectroscopic guidance of pulsed laser assisted balloon angioplasty. Circulation Oct 1988; 78:ll-504.

9. Geschwind HJ, Dubois-Rande JL, Poirot G, Boussignac G: Guided percutaneous pulsed laser angioplasty: Results and Follow-up. JACC Feb 1989; 13:13A.

10. Isner JM, DeJesus SR, Clarke RH, Gal D, Ronsione AJ, Donaldson RJ: Mechanism of laser Ablation in an Ab- sorbing Fluid Field. Lasers Surg Med 1988; 8543-554.

11. Nishioka NS, Teng PT, Deutsch TF, Anderson RR: Mech- anism of laser-induced fragmention of urinary and biliary calculi. Lasers in Life Sci 1987; 1:231-245.

12. Teng PT, Nishioka NS, Farinelli WA, Anderson RR, Deutsch TF: Microsecond-long flash photography of laser induced ablation of biliary and urinary calculi. Lasers Surg Med 1987; 7:394-397.

13. Anderson RR, Jaenke KF, Parrish JA: Mechanisms of selective vascular changes caused by dye lasers. Lasers Surg Med 1983; 3:211-215.

14. Anderson RR, Parrish JA: Microvasculature can be selectively damaged using dye lasers: A basic theory and

349-351.

Effect of Blood On Ablation of Plaque 543 experimental evidence in human skin. Lasers Surg Med

15. Anderson RR, Parrish JA: Selective photothermolysis: Precise microsurgery by selective absorption of pulsed radiation. Science 1983; 220524-527.

16. LaMuraglia GM, Anderson RR, Parrish JA, Zhang Z, Prince MR Selective laser ablation of venous thrombus: Implications for a new approach in the treatment of pul- monary embolus. Lasers Surg Med 1988; 8:486-493.

1981; 1~263-276. 17. Kaplan MD, Case RB, Choy DSJ: Vascular recanalization

with the argon laser: The role of blood in the transmis- sion of laser energy. Lasers Surg Med 1985; 5275-279.

18. Abela GS, Crea F, Smith W, Pepine CJ, Conti CR: In vitro effects of argon laser radiation on blood: Quantitative and morphologic analysis. JACC 1985; 5:231-237.

19. Crea F, Fenech A, Smith W, Conti CR, Abela GS: Laser recanalization of acutely thrombosed coronary arteries in live dogs. JACC 1985; 6:1052-56

Documents

Effect of blood upon the selective ablation of atherosclerotic plaque witha pulsed dye laser