Lasers in Surgery and Medicine 12520-527 (1992)
Ultrafast Imaging of Tissue Ablation by a XeCl Excimer Laser in Saline -
Melitta 6. Preisack, MD, Walter Neu, PhD, Ralf Nyga, Manfred Wehrmann, MD, Karl K. Haase, MD,and Karl R. Karsch, MD Division of Cardiology, Tubingen University (M.5.P., M. W., K.K.H.,K.R.K.), and Laser-Laboratorium (W.N., R.N.), Gottingen, Germany
To determine the temporal evolution of laser induced tissue ablation, arterial wall specimens with either hard calcified or fatty plaques and normal tissue were irradiated in a 0.9%saline solution using a XeCl excimer laser (wavelength 308 nm, energy fluence 7 J/cm2, pulse width 30 ns) through a 600 pm fused silica fiber pointing perpendicular either at a 0.5 mm distance or in direct contact to the vascular surface. Radiation of a pulsed dye laser (wavelength 580 nm) was used to illuminate the tissue surface. The ablation process and the arising bubble above the tissue surface were recorded with a CCD camera attached to a computer based image-processing system. Spherical cavitation bubbles and small tissue particles emerging from the irradiated area have been recorded. The volume of this bubble increased faster for calcified plaques than for normal tissue. 0 1992 Wiley-Liss, Inc.
Key words: atherosclerosis, excimer laser, pulsed laser angioplasty, transmission
Using the technique of ultrafast imaging, the temporal development and spatial distribution of The possibility for effective ablation and prethe laser ablation has been visualized by several cise etching of vascular tissue with only minimal groups [5,8-101. Recent studies [7,9]have reported thermal side effects 111 had proved XeCl excimer on the results of noncontact tissue ablation by a laser angioplasty as a feasible method for treatmid-infrared holmium laser system performed in ment of coronary artery disease. Beside its ability both saline and blood. Several investigators have t o ablate tissue efficiently, the excimer laser is visualized excimer laser ablation using a fiber deable to reduce coagulation and necrosis of the inlivery system [5,101. However, all experiments timal layer by minimizing thermal injury during dealing with excimer laser ablation were deterablation. During the past two years excimer laser mined under atmospheric conditions. Thus, the angioplasty has gained increasing interest as an current investigation was performed irradiating alternative to balloon angioplasty 12-41. tissue samples in a 0.9% saline solution in order t o However, the mechanism associated with absimulate comparable conditions of coronary laser lation of vascular tissue has been subject to quite angioplasty as in the clinical setting. To evaluate a number of investigations [5-71. Since ablation the effects of excimer laser light on various tissue has been observed even with short pulsed infrared calcified and fatty plaques as well as normal aortic lasers [71 where the photon energy is not suffi- wall have been investigated. Tissue craters and cient to break chemical bonds, it is probable that adjacent damage were examined histologically. two processes are responsible for the basic mechanism of ablation: bond breakage and water superheating [81. Despite the phenomenon leading to tissue ablation, it is well known that a rapid Accepted for publication May 6, 1992. expansion of gaseous ablation products occurs Address reprint requests t o Dr. Melitta B. Preisack, Univerduring ablation, and results in an explosive re- sity Medical Center, Tiibingen, Otfried-Muller Str. 10, 7400 Tubingen, Germany. moval of the irradiated material [6-81. INTRODUCTION
0 1992 Wiley-Liss, Inc.
--
Tissue Ablation Imaging
.Fiber
-
XeCl Laser, 308nm
atrenuator Trigger
CCD Camera
b Sample
PC
+
Frame grabber
Fig. 1. Experimental setup for ultrafast imaging of vascular tissue ablation by a XeCl excimer laser. The ablation process and the arising bubble above the tissue surface were recorded with a CCD camera attached to a computer-based image-processing system.
METHODS Tissue Specimens
Thirty samples of fresh human aortic wall specimens were investigated using ultrafast imaging. Beside healthy tissue the arterial segments showed different types of atherosclerotic plaque, from lipid rich to hard calcified plaques. The fresh vessels were cut into a size of at least 2 cm x 2 cm and placed in a quartz cuvette containing a 0.9% saline solution. Experimental Setup
In Figure 1 the experimental setup used for ultrafast imaging of vascular tissue ablation is shown schematically. A XeCl excimer laser (Lambda-PhysikLPX 210 ICC) with a wavelength of 308 nm and a pulse width of 30 ns (full width at half-maximum [FWHM]) was used for tissue ablation. The pulses were transmitted through a fixed 600 pm fused silica fiber with the fiber tip pointing perpendicular t o the vascular surface. The energy at the distal end of the fiber was adjusted with a variable attenuator. The pulse energy was set to 20 mJ, which corresponds t o fluences of about 7 J/cm2at the distal end of the fiber. The ablation threshold is about 1-4 J/cm2 i l l ] .
521
Visible radiation of a pulsed dye laser operated at a wavelength of 580 nm with a pulse width of about 10 ns FWHM was used t o illuminate the tissue surface for visualization of the ablation process. This system served as a flash light source for the imaging process. A fast photodiode in combination with a digital storing oscilloscope was used t o monitor the delay time between the ablation and the dye laser pulse. The ablation process was recorded by a CCD camera for each excimer laser pulse. Triggering and delaying of both laser systems were controlled by a PC via an electronic pulse generator (Hewlett-Packard)which allowed reproducible adjustment of delay times in the range of nanoseconds up to hundreds of microseconds. The CCD camera was attached to a PC-based imageprocessing system to record the photographed ablation process. The magnification factor of the microscope imaging system was 16. Study Design
Firstly, a probing pulse of the dye laser was used to image the selected sample segment before ablation. Care was taken t o photograph the ablation site without any shadowing of neighbouring tissue segments. Series of pictures were recorded from calcified and fatty plaques as well as from normal arterial wall at increasing delay times with a step width of 5 ps. Since the repetition rate of the excimer laser pumped dye laser is much too low for high speed cinematography, only single pictures at certain delay times have been recorded for each ablating pulse. By means of a micromanipulator fitted with a vernier micrometer the fiber tip was adjusted at a 0.5 mm distance t o the sample. In a second series the fiber was positioned in direct contact t o the vascular surface. Each laser pulse was applied to different positions of the same plaque specimen. This was achieved by moving the samples via the micromanipulator in steps of 800 pm. Histological Examination
The arterial segments irradiated were examined histologically t o specify the type of plaque which had been ablated. After fixation in a 10% formalin solution, the excised samples were embedded in paraffin and prepared for histological examination. The segments were cut into series of 4 pm thick slices and stained with Haematoxylin and Eosin or Elastica van Gieson. The plaques were classified in calcified and fatty plaques, depending on the major component of the plaque. Additionally, the arterial samples were examined
Preisack et al.
522
TABLE 1. Time Course of the Calculated Bubble Volume for Thirty Samples With Calcified and Fatty Plaques and for Normal Aortic Wall Calcified 0.04 f 0.07 40-50 1.31 f 0.43 130-160 30.2 t 9.2
Accuracy (mm3) Delay time at the maximal volume (ps) Maximal volume (mm3) Lifetime (ks) Volume increase after rise (cm3/s)
for sharpness of crater edges, carbonization of the surrounding tissue and debris within the crater lumen. Statistical Analysis
Fatty 0.06 t_ 0.08 30-55 0.34 f 0.11 100-120 11.4 2 4.9
ume V of the observed bubble can be described according to:
V
The volume of the observed bubble was calculated assuming rotational symmetry and using the model of a spherical segment. All values, unless otherwise indicated are expressed as mean 1 standard deviation. For these evaluated data a systematic error due t o the deviation from a spherical segment was not taken into account. To determine the accuracy of this determination, the deviation from a sphere was estimated by averaging the difference between the volume of a spherical segment completely enclosing a bubble and the volume of a spherical segment that is completely included in the same bubble (Table 1). Bubbles with an estimated error > 10% of the calculated volume were excluded from this analysis. An unpaired t-test was used to compare the maximal bubble volume of normal and atherosclerotic specimens. P values < .05 were regarded as significant.
Normal 0.04 f 0.05 50-70 0.32 t 0.25 100-140 5.4 t 1.9
rh
= 7(3r2+h2)
6
where r is the radius and h the height of the spherical segment. The volume-time relation for calcified, fatty, and normal tissue samples at a 0.5 mm distance is provided in Figure 3. The expansion of the calculated volume was different for normal tissue and calcified plaques resulting in different slopes of the figured curves. For the latter type of tissue a considerably steeper increase of the bubble volume was noted (P=O.OOl). The bubbles showed a maximum after 40-50 ps in case of calcified plaques and after 50-70 p s for irradiation of normal arterial wall (Table 1). The maximal extension of the bubble was considerably higher for calcified samples than for normal aortic wall (P= 0.0001). The life time of the bubble was up t o 160 ps for calcified and up to 140 p s for normal tissue. The evaluated data for irradiation of fatty plaques reveal a quite similar slope compared t o that of normal tissue. RESULTS In Figure 4 the ablation effect of direct conUltrafast Imaging tact between the fiber tip and the vascular wall is Thirty aortic specimen were subjected to ex- figured. After about 10 p s a bubble rises from the cimer laser ablation. Figure 2 shows a represen- tissue. This bubble has an irregular shape and, in tative series of photographs of an ablation process comparison to noncontact tissue ablation shown of normal arterial wall. The distance between the in Figure 2, appears somewhat suppressed. A fiber tip and the tissue surface was 0.5 mm, and maximum is already reached at a delay time of 40 the energy fluence was 7 J/cm2. Each photograph to 45 ps, the collapse is followed by an ejection of was taken at different delay times ranging up to small particles emerging from the irradiated several hundreds of microseconds after the ablat- area. ing laser pulse. The first photograph of this series shows the tissue surface before ablation. At about Histological Examination The consistence of the plaque irradiated was two microseconds after ablation is initiated a spherical bubble starts developing from the tissue confirmed by histological examination. In Figure surface. Growing steadily, a maximal extension is 5 a crater after excimer laser ablation (energy reached after 50 to 70 ps, then it begins to col- fluence 7 J/cm2, distance 0.5 mm) is shown. Only small disruption of crater edges and of the surlapse. Assuming rotational symmetry and using rounding tissue was apparent. Minimal thermal the simple model of a spherical segment, the vol- effects were found on the crater borders.
*
Tissue Ablation Imaging
523
Fig. 2. Series of photographs taken from normal tissue at different delay times with the experimental setup shown in Fig. 1(distance between fiber tip and tissue surface: 0.5 mm). The distal end of the fiber appears as a transparent bar pointing from the top in each photograph. The irregular horizontal
contour in front of the illuminated background is the lateral view of the tissue surface. The ablation process is figured as an increasing bubble appearing as a shadow. Delay time: a)0 ps, b) 2 ps, c ) 20 ps, d) 40 ks, e ) 60 ps, D 100 ps.
DISCUSSION
effectively treat patients with coronary artery disease [2-41. A variety of catheter delivery systems and different transmission devices has been used, yielding 60-70% acute success rates E21.
The coronary excimer laser angioplasty has, in the past two years, demonstrated its ability to
524
Preisack et al.
T
"f *
// I
py
\
{
1.0
I
E E
I
i
/i I I
c
I
>
f
,I
i
,
.o 2
/
10
I
I
I
1
I
I
....._......,
100
50
,
I
150
Delay time/ps Fig. 3. Mean bubble volume for the ablation of healthy tissue (n),calcified plaques (A)and fatty plaques (0)at a 0.5 mm distance to the fiber tip (energy fluence 7 J/cm2)for delay times up to 160 ps. The error bars refer to the mean error due to the deviation from a spherical segment.
Complications, such as severe dissections and vessel perforation have been reduced using pulsed laser energy [ll. However, the problem of restenosis still exists. Since the intimal smooth muscle cell proliferation was found to be the predominant mechanism of this process [12], several investigators attempt to prevent the proliferative response of the vessel wall by reducing vascular injury during ablation. The development of novel guiding systems t o discriminate atherosclerotic plaque from normal tissue during laser ablation has become a key issue of recent investigations [13,141. However, besides the study of prevention sources, the investigation of the ablation process and its potential of damage to the surrounding tissue is important for a better understanding of the underlying ablation mechanisms. The effects of laser light on tissue can be divided into differ-
ent ablation mechanisms: photochemical, thermal, photoablative and photomechanical effects [XI. This latter mechanism is governed by nonlinear interactions and depends on the formation of a plasma of free electrons, which propagates a shock wave through the target. A recent investigation has reported on excimer laser ablation under atmospheric conditions using the ultrafast imaging technique [ 5 ] .According t o this study, we found that excimer laser ablation results in an explosive ejection of ablated tissue material in form of small particles. This was thought to be the result of electronic excitations induced by laser photons producing extreme local heating. As a consequence rapid volume expansion with explosive ablation of vascular tissue takes place [16]. In the present study cavitation bubbles aris-
Tissue Ablation Imaging
525
Fig. 4. Ablation of a soft plaque at different delay times with the fiber being in direct contact with the tissue surface. Delay times: a) 0 ks, b) 10 ps, c ) 40 ps, d) 80 k s , e ) 120 k s , f, 200 I*s.
ing above the tissue surface have been visualized at about two microseconds after the ablating laser pulse. Cavitation bubbles often occur in liquids which are subject to a strong expansion or flow
[171. In this case, microbubbles resulting from gaseous ablation products, could be the nuclei for the cavitations. More or larger nuclei, i.e. more gaseous products, advance the appearance of cav-
526
Preisack et al.
Fig. 5. Histological examination of a sample after excimer laser irradiation of normal arterial wall shows smooth crater border, with only minimal disruption of the surrounding tissue resulting from ablation.
itation bubbles. The generation mechanism of cavitation bubbles after an excimer laser pulse is largely unknown. One hypothesis is that formation of cavitation is caused by evaporation of irradiated tissue due to surface heating. Moreover, ablation of vascular material mediated by a microexplosion of tissue particles and formation of a plasma of free electrons propagates a shock wave, which has been detected acoustically 118,191. Fujimoto et al. I201 have shown that this pressure wave induced by optical breakdown of water travels at supersonic velocity for several hundred micrometers before slowing t o normal acoustic transients. This strong expansion of the pressure wave within the saline solution may generate cavitation bubbles which were detected about 2 ps after the laser pulse in the present study. The potential of damage t o near solid surfaces induced
by cavitation bubbles during collapse has been examined in the past 1211. Previous investigations reported on a lack of ablation following laser irradiation in air compared to the tissue damage created under water. Initially, this was attributed to a poor acoustical coupling in air. Yashima et al. 1221 interpreted this phenomenon as an inertial confinement of material by a column of water, providing better acoustical matching. Nishioka et al. [231 also noted an enhancement of fragmentation of biliary calculi with tunable dye lasers in saline and concluded that a shock wave would be more efficiently coupled into a stone when it was immersed in saline. However, our observations suggest that for excimer laser ablation formation of cavitation bubbles above the tissue surface in saline may be responsible for a more efficient removal of material when laser irradiation is performed under water. Particle displacement occurring after the cavitation bubble had collapsed supports this suggestion. The fact that the volume of the bubble increases much faster for calcified plaques than for healthy tissue or fatty plaque is possibly based on the different amount of generated gaseous ablation products, which must strongly depend on the absorption of the irradiated material. Additionally, differences in viscoelastic properties of calcified and normal tissue could explain the different expansion modalities of the observed cavitations. Crazzolara et al. [19] have found that excimer laser irradiation of calcified plaque results in a shorter rise time and a higher rate of pressure increase of detected acoustic signals than normal arterial wall. They suggested that the different mechanical properties of hard calcified plaques and normal tissue result in different acoustic impedance. This leads to differences in acoustic coupling of the vessel to the water, providing a nearly total reflection of the acoustic signal in the case of hard plaque, whereas normal tissue is acoustically better matched to water, allowing a propagation of the pressure wave into the tissue. Therefore, it can be suspected that a corresponding relation exists for the increase in cavitation expansion described in our experiments. For ablation in a contact mode the spatial distribution was not analyzed due to the strong deviation of the bubble from the model of a sphere used in this study. However, maximal volume expansion was sooner reached for this ablation modus in direct contact to the vascular surface than for noncontact ablation, which may be due t o different energy fluences at the tissue surface.
Tissue Ablation Imaging
CONCLUSIONS
In conclusion. XeCl excimer laser irradiation may induce an explosive ejection of ablated tissue material. Spherical cavitation bubbles arising from the irradiated area have been recorded. The volume increase of this bubble was different for calcified plaque and normal arterial wall, suggesting that different specific tissue properties must be taken into account. Beside the known photomechanical mechanisms, the generation and collapse of cavitation bubbles in liquids might play an additional role for excimer laser induced ablation of vascular tissue.
REFERENCES 1. Haase KK, Wehrmann M, Duda S, Karsch KR: Experimental intracoronary excimer laser angioplasty. Z Kardiol 1990; 79(3):183-188. 2. Karsch Kr, Haase KK, Voelker W, Baumbach A, Mauser M, Seipel L: Percutaneous coronary excimer laser angioplasty in patients with stable and unstable angina pectoris. Acute results and incidence of restenosis during 6-month follow-up. Circulation 1990; 81:1849-1859. 3. Sanborn TA, Bittle JA, Hershman RA, Siege1 RM: Percutaneous coronary excimer laser-assisted angioplasty: Initial multicenter experience in 141 patients. JACC 1991; 17:169B-l73B. 4. Margolis JR, Litvack F, Grundfest W, Eigler N, Goldenberg T, Laudenslager J, Tsoi D, Wong S, Segalowitz J, Hestrin L, Rothbaum D, Linnemeier T, Helfant R, Forrester J : Excimer laser angioplasty: Results of a multicenter study. Circulation 1989; 80 (4):11-477. 5. Srinivasan R, Casey KG, Haller JD: Sub-nanosecond probing of the ablation of soft plaque from arterial wall by a 308 nm laser pulse delivered through a fiber. IEEE J Quantum Electr 1990; 262279-2283. 6. Izatt JA, Albagli D, Itzan I, Feld M: Pulsed laser ablation of calcified tissue: Physical mechanisms and fundamental parameters. In Jacques SL (ed): “Laser-Tissue Interaction 1990.” Proc. SPIE 1202, 133-140. 7. Vorwerk D, Zolotas G, Hessel S, Adam G, Guenther RG: Vascular tissue ablation by erbium-YAG laser: A fibertransmittable pulsed laser in the infrared range. Invest Radio1 1990; 25235-239. 8. Srinivasan R: Ablation of polymers and biological tissue by ultraviolet lasers. Science 1986; 234:559-565. 9. Leeuwen TG, van der Veen MJ, Verdaasdonk RM, Borst
527
C: Non-contact Tissue Ablation By Holmium: YSGG Laser Pulses in Blood. Lasers Surg Med 1991; 11:26-34. 10. Puliafito CA. Stern D, Krueger RR, Mandel ER: Highspeed photokaphy of excimei laser ablation of the cornea. Arch Ophthalmol 1987; 1051255-1259. 11. Singleton DL, Paraskevopoulos G, Jolly GS, Irwin RS, McKenney, Nip WS, Farrell EM, Higginson LAJ: Excimer laser in cardiovascular surgery: Ablation products and photoacoustic spectrum of arterial wall. Appl Phys Lett 1986; 482378-880. 12. Hanke H, Haase KK, Hanke S, Oberhoff M, Hassenstein S, Betz E, Karsch KR: Morphological changes and smooth muscle cell proliferation after experimental excimer laser treatment. Circulation 1991, 83:1380-1389. 13. Clarke RH, Isner JM, Gauthier T, Nakagawa K, Cerio F, Hanlon E, Gaffney E, Rouse E, Dejesus S: Spectroscopic characterization of cardiovascular tissue. Lasers Surg Med 1988; 8:45-59. 14. Cutruzzola FR, Stetz ML, OBrien KM, Gindi GR, Laifer LI, Garrand TJ, Deckelbaum LI: Change in laser-induced arterial fluorescence during ablation of atherosclerotic plaque. Lasers Surg Med 1989; 9:109-116. 15. Ischinger T, Coppenrath K, Weber H, Enders S, Unsold E, Hessel: Use of thermal laser effect of laser irradiation for cardiovascular applications exemplified by the Nd: YAG laser. Z Kardiol 1989; 78(11):689-700. 16. Husinsky W, Mitterer S, Grabner G, Baumgartner I: Photoablation by UV and visible laser radiation of native and doped biological tissue. Appl Phys 1989; B 49:463-467. 17. Hentschel W: On the transient response of gas bubbles in water. Acustica 1986; 6O:l-20. 18. Cross FW, Al-Dhahir RK, Dyer PE, MacRobert AJ: Time resolved photoacoustic studies of vascular tissue ablation at three laser wavelength. Appl Phys Lett 1987; 50: 1019-1021. 19. Crazzolara H, von Muench W, Rose C, Thiemann U, Haase KK, Ritter M, Karsch KR: Analysis of the acoustic response of vascular tissue irradiated by an ultraviolet laser pulse. J Appl Phys 1991; 70(3):1847-1849. 20. Fujimoto JG, Lin WZ, Ippen EP, Puliafito CA, Steinert RF: Time resolved studies of Nd:YAG laser-induced breakdown: Plasma formation, acoustic wave generation, and cavitation. Invest Ophthalmol Vis Sci 1985; 26:17711777. 21. Lauterborn W, Bolle H: Experimental investigations of cavitation bubble collapse in the neighborhood of a solid boundary. J Fluid Mech 1975; 72:391-399. 22. Yashima Y, McAuliffe DJ, Lacques SL, Flotte TJ: Laserinduced photoacoustic injury of skin: Effect of inertial confinement. Lasers Surg Med 1991; 11:62-68. 23. Nishioka DA, Lewins PC, Murray SC, Parrish JA, Anderson RR: Fragmentation of biliary calculi with tunable dye lasers. Gastroenterology 1987; 93250-255.
Ultrafast Imaging of Tissue Ablation by a XeCl Excimer Laser in Saline -
Melitta 6. Preisack, MD, Walter Neu, PhD, Ralf Nyga, Manfred Wehrmann, MD, Karl K. Haase, MD,and Karl R. Karsch, MD Division of Cardiology, Tubingen University (M.5.P., M. W., K.K.H.,K.R.K.), and Laser-Laboratorium (W.N., R.N.), Gottingen, Germany
To determine the temporal evolution of laser induced tissue ablation, arterial wall specimens with either hard calcified or fatty plaques and normal tissue were irradiated in a 0.9%saline solution using a XeCl excimer laser (wavelength 308 nm, energy fluence 7 J/cm2, pulse width 30 ns) through a 600 pm fused silica fiber pointing perpendicular either at a 0.5 mm distance or in direct contact to the vascular surface. Radiation of a pulsed dye laser (wavelength 580 nm) was used to illuminate the tissue surface. The ablation process and the arising bubble above the tissue surface were recorded with a CCD camera attached to a computer based image-processing system. Spherical cavitation bubbles and small tissue particles emerging from the irradiated area have been recorded. The volume of this bubble increased faster for calcified plaques than for normal tissue. 0 1992 Wiley-Liss, Inc.
Key words: atherosclerosis, excimer laser, pulsed laser angioplasty, transmission
Using the technique of ultrafast imaging, the temporal development and spatial distribution of The possibility for effective ablation and prethe laser ablation has been visualized by several cise etching of vascular tissue with only minimal groups [5,8-101. Recent studies [7,9]have reported thermal side effects 111 had proved XeCl excimer on the results of noncontact tissue ablation by a laser angioplasty as a feasible method for treatmid-infrared holmium laser system performed in ment of coronary artery disease. Beside its ability both saline and blood. Several investigators have t o ablate tissue efficiently, the excimer laser is visualized excimer laser ablation using a fiber deable to reduce coagulation and necrosis of the inlivery system [5,101. However, all experiments timal layer by minimizing thermal injury during dealing with excimer laser ablation were deterablation. During the past two years excimer laser mined under atmospheric conditions. Thus, the angioplasty has gained increasing interest as an current investigation was performed irradiating alternative to balloon angioplasty 12-41. tissue samples in a 0.9% saline solution in order t o However, the mechanism associated with absimulate comparable conditions of coronary laser lation of vascular tissue has been subject to quite angioplasty as in the clinical setting. To evaluate a number of investigations [5-71. Since ablation the effects of excimer laser light on various tissue has been observed even with short pulsed infrared calcified and fatty plaques as well as normal aortic lasers [71 where the photon energy is not suffi- wall have been investigated. Tissue craters and cient to break chemical bonds, it is probable that adjacent damage were examined histologically. two processes are responsible for the basic mechanism of ablation: bond breakage and water superheating [81. Despite the phenomenon leading to tissue ablation, it is well known that a rapid Accepted for publication May 6, 1992. expansion of gaseous ablation products occurs Address reprint requests t o Dr. Melitta B. Preisack, Univerduring ablation, and results in an explosive re- sity Medical Center, Tiibingen, Otfried-Muller Str. 10, 7400 Tubingen, Germany. moval of the irradiated material [6-81. INTRODUCTION
0 1992 Wiley-Liss, Inc.
--
Tissue Ablation Imaging
.Fiber
-
XeCl Laser, 308nm
atrenuator Trigger
CCD Camera
b Sample
PC
+
Frame grabber
Fig. 1. Experimental setup for ultrafast imaging of vascular tissue ablation by a XeCl excimer laser. The ablation process and the arising bubble above the tissue surface were recorded with a CCD camera attached to a computer-based image-processing system.
METHODS Tissue Specimens
Thirty samples of fresh human aortic wall specimens were investigated using ultrafast imaging. Beside healthy tissue the arterial segments showed different types of atherosclerotic plaque, from lipid rich to hard calcified plaques. The fresh vessels were cut into a size of at least 2 cm x 2 cm and placed in a quartz cuvette containing a 0.9% saline solution. Experimental Setup
In Figure 1 the experimental setup used for ultrafast imaging of vascular tissue ablation is shown schematically. A XeCl excimer laser (Lambda-PhysikLPX 210 ICC) with a wavelength of 308 nm and a pulse width of 30 ns (full width at half-maximum [FWHM]) was used for tissue ablation. The pulses were transmitted through a fixed 600 pm fused silica fiber with the fiber tip pointing perpendicular t o the vascular surface. The energy at the distal end of the fiber was adjusted with a variable attenuator. The pulse energy was set to 20 mJ, which corresponds t o fluences of about 7 J/cm2at the distal end of the fiber. The ablation threshold is about 1-4 J/cm2 i l l ] .
521
Visible radiation of a pulsed dye laser operated at a wavelength of 580 nm with a pulse width of about 10 ns FWHM was used t o illuminate the tissue surface for visualization of the ablation process. This system served as a flash light source for the imaging process. A fast photodiode in combination with a digital storing oscilloscope was used t o monitor the delay time between the ablation and the dye laser pulse. The ablation process was recorded by a CCD camera for each excimer laser pulse. Triggering and delaying of both laser systems were controlled by a PC via an electronic pulse generator (Hewlett-Packard)which allowed reproducible adjustment of delay times in the range of nanoseconds up to hundreds of microseconds. The CCD camera was attached to a PC-based imageprocessing system to record the photographed ablation process. The magnification factor of the microscope imaging system was 16. Study Design
Firstly, a probing pulse of the dye laser was used to image the selected sample segment before ablation. Care was taken t o photograph the ablation site without any shadowing of neighbouring tissue segments. Series of pictures were recorded from calcified and fatty plaques as well as from normal arterial wall at increasing delay times with a step width of 5 ps. Since the repetition rate of the excimer laser pumped dye laser is much too low for high speed cinematography, only single pictures at certain delay times have been recorded for each ablating pulse. By means of a micromanipulator fitted with a vernier micrometer the fiber tip was adjusted at a 0.5 mm distance t o the sample. In a second series the fiber was positioned in direct contact t o the vascular surface. Each laser pulse was applied to different positions of the same plaque specimen. This was achieved by moving the samples via the micromanipulator in steps of 800 pm. Histological Examination
The arterial segments irradiated were examined histologically t o specify the type of plaque which had been ablated. After fixation in a 10% formalin solution, the excised samples were embedded in paraffin and prepared for histological examination. The segments were cut into series of 4 pm thick slices and stained with Haematoxylin and Eosin or Elastica van Gieson. The plaques were classified in calcified and fatty plaques, depending on the major component of the plaque. Additionally, the arterial samples were examined
Preisack et al.
522
TABLE 1. Time Course of the Calculated Bubble Volume for Thirty Samples With Calcified and Fatty Plaques and for Normal Aortic Wall Calcified 0.04 f 0.07 40-50 1.31 f 0.43 130-160 30.2 t 9.2
Accuracy (mm3) Delay time at the maximal volume (ps) Maximal volume (mm3) Lifetime (ks) Volume increase after rise (cm3/s)
for sharpness of crater edges, carbonization of the surrounding tissue and debris within the crater lumen. Statistical Analysis
Fatty 0.06 t_ 0.08 30-55 0.34 f 0.11 100-120 11.4 2 4.9
ume V of the observed bubble can be described according to:
V
The volume of the observed bubble was calculated assuming rotational symmetry and using the model of a spherical segment. All values, unless otherwise indicated are expressed as mean 1 standard deviation. For these evaluated data a systematic error due t o the deviation from a spherical segment was not taken into account. To determine the accuracy of this determination, the deviation from a sphere was estimated by averaging the difference between the volume of a spherical segment completely enclosing a bubble and the volume of a spherical segment that is completely included in the same bubble (Table 1). Bubbles with an estimated error > 10% of the calculated volume were excluded from this analysis. An unpaired t-test was used to compare the maximal bubble volume of normal and atherosclerotic specimens. P values < .05 were regarded as significant.
Normal 0.04 f 0.05 50-70 0.32 t 0.25 100-140 5.4 t 1.9
rh
= 7(3r2+h2)
6
where r is the radius and h the height of the spherical segment. The volume-time relation for calcified, fatty, and normal tissue samples at a 0.5 mm distance is provided in Figure 3. The expansion of the calculated volume was different for normal tissue and calcified plaques resulting in different slopes of the figured curves. For the latter type of tissue a considerably steeper increase of the bubble volume was noted (P=O.OOl). The bubbles showed a maximum after 40-50 ps in case of calcified plaques and after 50-70 p s for irradiation of normal arterial wall (Table 1). The maximal extension of the bubble was considerably higher for calcified samples than for normal aortic wall (P= 0.0001). The life time of the bubble was up t o 160 ps for calcified and up to 140 p s for normal tissue. The evaluated data for irradiation of fatty plaques reveal a quite similar slope compared t o that of normal tissue. RESULTS In Figure 4 the ablation effect of direct conUltrafast Imaging tact between the fiber tip and the vascular wall is Thirty aortic specimen were subjected to ex- figured. After about 10 p s a bubble rises from the cimer laser ablation. Figure 2 shows a represen- tissue. This bubble has an irregular shape and, in tative series of photographs of an ablation process comparison to noncontact tissue ablation shown of normal arterial wall. The distance between the in Figure 2, appears somewhat suppressed. A fiber tip and the tissue surface was 0.5 mm, and maximum is already reached at a delay time of 40 the energy fluence was 7 J/cm2. Each photograph to 45 ps, the collapse is followed by an ejection of was taken at different delay times ranging up to small particles emerging from the irradiated several hundreds of microseconds after the ablat- area. ing laser pulse. The first photograph of this series shows the tissue surface before ablation. At about Histological Examination The consistence of the plaque irradiated was two microseconds after ablation is initiated a spherical bubble starts developing from the tissue confirmed by histological examination. In Figure surface. Growing steadily, a maximal extension is 5 a crater after excimer laser ablation (energy reached after 50 to 70 ps, then it begins to col- fluence 7 J/cm2, distance 0.5 mm) is shown. Only small disruption of crater edges and of the surlapse. Assuming rotational symmetry and using rounding tissue was apparent. Minimal thermal the simple model of a spherical segment, the vol- effects were found on the crater borders.
*
Tissue Ablation Imaging
523
Fig. 2. Series of photographs taken from normal tissue at different delay times with the experimental setup shown in Fig. 1(distance between fiber tip and tissue surface: 0.5 mm). The distal end of the fiber appears as a transparent bar pointing from the top in each photograph. The irregular horizontal
contour in front of the illuminated background is the lateral view of the tissue surface. The ablation process is figured as an increasing bubble appearing as a shadow. Delay time: a)0 ps, b) 2 ps, c ) 20 ps, d) 40 ks, e ) 60 ps, D 100 ps.
DISCUSSION
effectively treat patients with coronary artery disease [2-41. A variety of catheter delivery systems and different transmission devices has been used, yielding 60-70% acute success rates E21.
The coronary excimer laser angioplasty has, in the past two years, demonstrated its ability to
524
Preisack et al.
T
"f *
// I
py
\
{
1.0
I
E E
I
i
/i I I
c
I
>
f
,I
i
,
.o 2
/
10
I
I
I
1
I
I
....._......,
100
50
,
I
150
Delay time/ps Fig. 3. Mean bubble volume for the ablation of healthy tissue (n),calcified plaques (A)and fatty plaques (0)at a 0.5 mm distance to the fiber tip (energy fluence 7 J/cm2)for delay times up to 160 ps. The error bars refer to the mean error due to the deviation from a spherical segment.
Complications, such as severe dissections and vessel perforation have been reduced using pulsed laser energy [ll. However, the problem of restenosis still exists. Since the intimal smooth muscle cell proliferation was found to be the predominant mechanism of this process [12], several investigators attempt to prevent the proliferative response of the vessel wall by reducing vascular injury during ablation. The development of novel guiding systems t o discriminate atherosclerotic plaque from normal tissue during laser ablation has become a key issue of recent investigations [13,141. However, besides the study of prevention sources, the investigation of the ablation process and its potential of damage to the surrounding tissue is important for a better understanding of the underlying ablation mechanisms. The effects of laser light on tissue can be divided into differ-
ent ablation mechanisms: photochemical, thermal, photoablative and photomechanical effects [XI. This latter mechanism is governed by nonlinear interactions and depends on the formation of a plasma of free electrons, which propagates a shock wave through the target. A recent investigation has reported on excimer laser ablation under atmospheric conditions using the ultrafast imaging technique [ 5 ] .According t o this study, we found that excimer laser ablation results in an explosive ejection of ablated tissue material in form of small particles. This was thought to be the result of electronic excitations induced by laser photons producing extreme local heating. As a consequence rapid volume expansion with explosive ablation of vascular tissue takes place [16]. In the present study cavitation bubbles aris-
Tissue Ablation Imaging
525
Fig. 4. Ablation of a soft plaque at different delay times with the fiber being in direct contact with the tissue surface. Delay times: a) 0 ks, b) 10 ps, c ) 40 ps, d) 80 k s , e ) 120 k s , f, 200 I*s.
ing above the tissue surface have been visualized at about two microseconds after the ablating laser pulse. Cavitation bubbles often occur in liquids which are subject to a strong expansion or flow
[171. In this case, microbubbles resulting from gaseous ablation products, could be the nuclei for the cavitations. More or larger nuclei, i.e. more gaseous products, advance the appearance of cav-
526
Preisack et al.
Fig. 5. Histological examination of a sample after excimer laser irradiation of normal arterial wall shows smooth crater border, with only minimal disruption of the surrounding tissue resulting from ablation.
itation bubbles. The generation mechanism of cavitation bubbles after an excimer laser pulse is largely unknown. One hypothesis is that formation of cavitation is caused by evaporation of irradiated tissue due to surface heating. Moreover, ablation of vascular material mediated by a microexplosion of tissue particles and formation of a plasma of free electrons propagates a shock wave, which has been detected acoustically 118,191. Fujimoto et al. I201 have shown that this pressure wave induced by optical breakdown of water travels at supersonic velocity for several hundred micrometers before slowing t o normal acoustic transients. This strong expansion of the pressure wave within the saline solution may generate cavitation bubbles which were detected about 2 ps after the laser pulse in the present study. The potential of damage t o near solid surfaces induced
by cavitation bubbles during collapse has been examined in the past 1211. Previous investigations reported on a lack of ablation following laser irradiation in air compared to the tissue damage created under water. Initially, this was attributed to a poor acoustical coupling in air. Yashima et al. 1221 interpreted this phenomenon as an inertial confinement of material by a column of water, providing better acoustical matching. Nishioka et al. [231 also noted an enhancement of fragmentation of biliary calculi with tunable dye lasers in saline and concluded that a shock wave would be more efficiently coupled into a stone when it was immersed in saline. However, our observations suggest that for excimer laser ablation formation of cavitation bubbles above the tissue surface in saline may be responsible for a more efficient removal of material when laser irradiation is performed under water. Particle displacement occurring after the cavitation bubble had collapsed supports this suggestion. The fact that the volume of the bubble increases much faster for calcified plaques than for healthy tissue or fatty plaque is possibly based on the different amount of generated gaseous ablation products, which must strongly depend on the absorption of the irradiated material. Additionally, differences in viscoelastic properties of calcified and normal tissue could explain the different expansion modalities of the observed cavitations. Crazzolara et al. [19] have found that excimer laser irradiation of calcified plaque results in a shorter rise time and a higher rate of pressure increase of detected acoustic signals than normal arterial wall. They suggested that the different mechanical properties of hard calcified plaques and normal tissue result in different acoustic impedance. This leads to differences in acoustic coupling of the vessel to the water, providing a nearly total reflection of the acoustic signal in the case of hard plaque, whereas normal tissue is acoustically better matched to water, allowing a propagation of the pressure wave into the tissue. Therefore, it can be suspected that a corresponding relation exists for the increase in cavitation expansion described in our experiments. For ablation in a contact mode the spatial distribution was not analyzed due to the strong deviation of the bubble from the model of a sphere used in this study. However, maximal volume expansion was sooner reached for this ablation modus in direct contact to the vascular surface than for noncontact ablation, which may be due t o different energy fluences at the tissue surface.
Tissue Ablation Imaging
CONCLUSIONS
In conclusion. XeCl excimer laser irradiation may induce an explosive ejection of ablated tissue material. Spherical cavitation bubbles arising from the irradiated area have been recorded. The volume increase of this bubble was different for calcified plaque and normal arterial wall, suggesting that different specific tissue properties must be taken into account. Beside the known photomechanical mechanisms, the generation and collapse of cavitation bubbles in liquids might play an additional role for excimer laser induced ablation of vascular tissue.
REFERENCES 1. Haase KK, Wehrmann M, Duda S, Karsch KR: Experimental intracoronary excimer laser angioplasty. Z Kardiol 1990; 79(3):183-188. 2. Karsch Kr, Haase KK, Voelker W, Baumbach A, Mauser M, Seipel L: Percutaneous coronary excimer laser angioplasty in patients with stable and unstable angina pectoris. Acute results and incidence of restenosis during 6-month follow-up. Circulation 1990; 81:1849-1859. 3. Sanborn TA, Bittle JA, Hershman RA, Siege1 RM: Percutaneous coronary excimer laser-assisted angioplasty: Initial multicenter experience in 141 patients. JACC 1991; 17:169B-l73B. 4. Margolis JR, Litvack F, Grundfest W, Eigler N, Goldenberg T, Laudenslager J, Tsoi D, Wong S, Segalowitz J, Hestrin L, Rothbaum D, Linnemeier T, Helfant R, Forrester J : Excimer laser angioplasty: Results of a multicenter study. Circulation 1989; 80 (4):11-477. 5. Srinivasan R, Casey KG, Haller JD: Sub-nanosecond probing of the ablation of soft plaque from arterial wall by a 308 nm laser pulse delivered through a fiber. IEEE J Quantum Electr 1990; 262279-2283. 6. Izatt JA, Albagli D, Itzan I, Feld M: Pulsed laser ablation of calcified tissue: Physical mechanisms and fundamental parameters. In Jacques SL (ed): “Laser-Tissue Interaction 1990.” Proc. SPIE 1202, 133-140. 7. Vorwerk D, Zolotas G, Hessel S, Adam G, Guenther RG: Vascular tissue ablation by erbium-YAG laser: A fibertransmittable pulsed laser in the infrared range. Invest Radio1 1990; 25235-239. 8. Srinivasan R: Ablation of polymers and biological tissue by ultraviolet lasers. Science 1986; 234:559-565. 9. Leeuwen TG, van der Veen MJ, Verdaasdonk RM, Borst
527
C: Non-contact Tissue Ablation By Holmium: YSGG Laser Pulses in Blood. Lasers Surg Med 1991; 11:26-34. 10. Puliafito CA. Stern D, Krueger RR, Mandel ER: Highspeed photokaphy of excimei laser ablation of the cornea. Arch Ophthalmol 1987; 1051255-1259. 11. Singleton DL, Paraskevopoulos G, Jolly GS, Irwin RS, McKenney, Nip WS, Farrell EM, Higginson LAJ: Excimer laser in cardiovascular surgery: Ablation products and photoacoustic spectrum of arterial wall. Appl Phys Lett 1986; 482378-880. 12. Hanke H, Haase KK, Hanke S, Oberhoff M, Hassenstein S, Betz E, Karsch KR: Morphological changes and smooth muscle cell proliferation after experimental excimer laser treatment. Circulation 1991, 83:1380-1389. 13. Clarke RH, Isner JM, Gauthier T, Nakagawa K, Cerio F, Hanlon E, Gaffney E, Rouse E, Dejesus S: Spectroscopic characterization of cardiovascular tissue. Lasers Surg Med 1988; 8:45-59. 14. Cutruzzola FR, Stetz ML, OBrien KM, Gindi GR, Laifer LI, Garrand TJ, Deckelbaum LI: Change in laser-induced arterial fluorescence during ablation of atherosclerotic plaque. Lasers Surg Med 1989; 9:109-116. 15. Ischinger T, Coppenrath K, Weber H, Enders S, Unsold E, Hessel: Use of thermal laser effect of laser irradiation for cardiovascular applications exemplified by the Nd: YAG laser. Z Kardiol 1989; 78(11):689-700. 16. Husinsky W, Mitterer S, Grabner G, Baumgartner I: Photoablation by UV and visible laser radiation of native and doped biological tissue. Appl Phys 1989; B 49:463-467. 17. Hentschel W: On the transient response of gas bubbles in water. Acustica 1986; 6O:l-20. 18. Cross FW, Al-Dhahir RK, Dyer PE, MacRobert AJ: Time resolved photoacoustic studies of vascular tissue ablation at three laser wavelength. Appl Phys Lett 1987; 50: 1019-1021. 19. Crazzolara H, von Muench W, Rose C, Thiemann U, Haase KK, Ritter M, Karsch KR: Analysis of the acoustic response of vascular tissue irradiated by an ultraviolet laser pulse. J Appl Phys 1991; 70(3):1847-1849. 20. Fujimoto JG, Lin WZ, Ippen EP, Puliafito CA, Steinert RF: Time resolved studies of Nd:YAG laser-induced breakdown: Plasma formation, acoustic wave generation, and cavitation. Invest Ophthalmol Vis Sci 1985; 26:17711777. 21. Lauterborn W, Bolle H: Experimental investigations of cavitation bubble collapse in the neighborhood of a solid boundary. J Fluid Mech 1975; 72:391-399. 22. Yashima Y, McAuliffe DJ, Lacques SL, Flotte TJ: Laserinduced photoacoustic injury of skin: Effect of inertial confinement. Lasers Surg Med 1991; 11:62-68. 23. Nishioka DA, Lewins PC, Murray SC, Parrish JA, Anderson RR: Fragmentation of biliary calculi with tunable dye lasers. Gastroenterology 1987; 93250-255.
