Lasers in Surgery and Medicine 10:405413 (1990)
Holmium:YAG Laser Ablation of Vascular Tissue George E. Kopchok, BS, Rodney A. White, MD, Marwan Tabbara, Vahid Saadatmanesh, MS, and Shi-Kaung Peng, MD, PhD
MD,
Depaflments of Surgery (G.E. K., R.A.W., M. T , V.S.) and Pathology (S-k.P), Harbor-UCLA Medical Center, Torrance, California 90509
The ablation of atherosclerotic lesions without collateral thermal or shock wave damage is thought to be a key element for successful laser angioplasty. This study evaluated the effectiveness of pulsed ho1mium:YAG laser (2.1 pm wavelength) for this application. Fresh normal tissue (n = 139) and arteriosclerotic canine arteries (n = 21) as well as formalin-preserved normal canine (n = 31) and atherosclerotic human arteries (n = 177) were irradiated under saline via a 600 pm diameter fiber placed perpendicular to the intimal surface with 0-10 gm of force. The laser was operated in the free running mode (FRM; 250 psec pulsewidth, 5 Hz, 30-7,100 mJ/mm2) and in the Q-switched mode (QSM; 200 nsec pulsewidth, 6 Hz, 30-1,100 mJ/mm2).Following the experiments, the samples were prepared for histologic and morphometric analysis. Ablation thresholds in the FRM were 60 and 180 mJ/mm2 in fresh and preserved canine tissue, respectively. Ablation thresholds in the QSM for fresh and preserved canine tissues were 75 and 180 mJ/mm2, respectively. Thresholds for human atherosclerotic tissue were dependent on the amount of calcification. In the QSM and FRM, there were no samples that could not be penetrated at 1,100 mJ/mm2 and above. Histologic examination of the FRM samples revealed confined columns of tissue ablation, with approximately 55-250 pm and 70-140 pm zones of thermal effect being apparent in the fresh and formalinpreserved samples, respectively. The QSM samples (fresh and preserved) revealed minimal thermal effect (0-20 pm) to the intima and media (smooth muscle cells) between 300 and 1,100 mJ/ mm2, whereas thermal effect in the adventitia (collagen) was 64172 pm with the same energy fluence. We conclude 1) that the thermal effect can be optimized by shortening the pulsewidth (Q switching at 200 nsec) or by limiting the total energy delivered and 2) that controlled ablation of both normal and atherosclerotic vascular tissue can be accomplished with the ho1mium:YAG laser. Key words: Ho:YAG laser, ablation threshold, vascular changes, Q-switchedmode, free running mode
INTRODUCTION
The ability to deliver energy through a thin flexible fiberoptic is a key prerequisite for the use of lasers in vascular surgery. A laser that can penetrate or ablate various components of atheroSClerOtiC plaque including Calcium with minimal for arOr acoustic effect may be terial angioplasty. Pulsed lasers such as Excimer [l,2], Er:YAG [3], ho1mium:YAG [4],and hol0 1990 Wiley-Liss, Inc.
mium:YSGG [51 have been shown t o meet these criteria. The ho1mium:YAG laser, with its wavelength of 2.1 ~ m may , also satisfy these condi Accepted for publication June 19, 1990, Address reprint requests to Rodney A. White, MD, Chief, Vascular Surgery, Harbor-UCLA Medical Center, 1000 West Carson Street, Torrance, CA 90509.
406
Kopchok et al.
tions. It has a n absorption coefficient (a)in pure water of 30 cm-' and a depth of penetration of 333 pm, which accentuates the tissue effects at this wavelength. The potential of shallow penetration depth coupled with high pulse energies makes the Ho:YAG laser a promising candidate for laser angioplasty . This study evaluated the Ho:YAG (2.1 pm) laser for use in vascular angioplasty. In vitro tissue (fresh and formalin-preserved) effects were investigated in the free running mode (FRM; 250 psec pulsewidth) and Q-switched mode (QSM; 200 nsec pulsewidth). Normal and arteriosclerotic canine and atherosclerotic human tissues were evaluated. MATERIALS AND METHODS Tissue Samples
Both canine and human tissues were evaluated. Fresh samples were studied immediately following removal, and preserved samples were treated with 10% formalin before being exposed to laser energy. Canine arteriosclerotic arteries were created by a combination of mechanical stripping of the intima and breaking of the elastic lamina of the iliac and femoral arteries and a high-cholesterol diet [6]. Human arteries were obtained from surgical specimens following disposal or from cadavers donated for medical teaching and research. All animals received humane care as outlined by the National Research Council's Guide for the Care and Use of Laboratory Animals (NIH Publication 62-43, revised 1985) and the Principles of Laboratory Animal Care formulated by the National Society of Medical Research. Laser
The Ho:YAG laser system (Trimedyne, Inc., Tustin, CA) (2.1 pm wavelength) was used in two different modes of operation. In the FRM, the laser energy was delivered with 250 psec pulsewidths (full width at half maximum) at 5 Hz. Six hundred micrometer diameter low-OH silica f i bers were used to deliver the energy. The energy fluences used ranged from 30 to 7,100 mJ/mm2 per pulse. In the QSM, the energy was delivered in a 200 nsec pulsewidth (full width at half maximum) at 6 Hz. A polished 600 pm diameter lowOH silica fiber was used to deliver the energy fluence, which ranged from 30 to 1,100 mJ/mm2 per pulse. In both modes, 100 pulses of laser energy was applied at five or six pulses per second.
Fig. 1. Experimental apparatus used to hold the vascular tissue samples perpendicular to the laser fiber a t a constant force. Zero or ten grams of force was added to the weight of the fiber.
Ablation Thresholds
To control the precision of energy delivery and the amount of force exerted on the tissue, all samples were placed in a specially designed fixture, which controlled the fiber alignment and applied a force of 0 or 10 gm (Fig. 1)to the fiber. The vascular tissues were cut into pieces with dimensions of approximately 1.5 x 2 cm and placed in a holder with the laser energy delivered perpendicular to the tissue. All specimens were immersed in room temperature saline with the fiber in contact with the intimal surface of the arteries prior to laser activation. Morphometric Analysis of Laser-Tissue Interactions
Following laser exposure, tissue effects were observed grossly at x 2 0 and compared to specimens that were prepared for histologic examination using trichrome and Verhoeff VanGieson (elastin) stains. Three or four samples were evaluated at each energy level. The majority of the specimens were cut parallel to the axis of the holes to determine the depth of penetration and the effect of laser energy on different layers of the vessel. To define further the dimensions and configuration of the holes (circular vs. eliptical), selected specimens were cut in a plane perpendicular to the axis of the zone of ablation. The thermal effect was quantified by measuring the maximum lateral distance of abnormal morphology from the crater edge in the media of the vessel. The two
Ho1mium:YAG Laser Ablation
407
TABLE 1. Ablation Threshold (mJ/mm2) Canine Fresha Preserved Free running mode (250 psec) Q-switched mode (200 nsec)
60 75
180 180
Human atherosclerotic Noncalcified Calcified 160 280
700-1,100 480-1,100
"No difference between normal and arteriosclerotic tissues.
types of degenerative effects seen were coagula- through the intima and media but slowed and tion and vacuolization. Distances were measured took much longer to pass through the adventitia. at x 100 using a microplan I1 digital pad (LaboThe configuration of the ablation zone (cirratory Computer Systems, Inc.). cular vs. eliptical) was inconsistant in both the FRM and QSM. Most of the holes were circular in shape (Fig. 51, but frequently identical parameRESULTS ters produced both circular and eliptical holes. Free Running Mode The minimum hole diameters created with a The ablation thresholds in the FRM at 5 Hz 600 pm fiber in fresh and fixed canine tissues were determined to be approximately 60 and 180 ranged from 220 t o 460 pm. mJ/mm2 in fresh and fixed normal canine arterial In the FRM, the energy required to ablate tissues, respectively (Table 1). The amount of tis- the human tissue varied due to differences in sue removed with 100 pulses at this energy was calcification. In the three groups of noncalcified, minimal, and virtually no degenerative changes moderately calcified, and heavily calcified tissue, (thermal or shockwave injury) could be seen mi- the energy fluences required to pass through the croscopically. Increasing the energy fluence from tissues (ablation threshold) were approximately threshold to 1,500 mJ/mm2 appeared to produce 160, 700, and 1,100 mJ/mm2. There were no an initial rise, then a leveling off, in the amount samples found that could not be penetrated from of tissue effect. Smaller zones of coagulation could 1,100 t o 1,800 mJ/mm2. In the higher energy be measured in the intima and media compared to fluence range, the amounts of degenerative the adventitia. The smallest areas of effect seen in changes seen in fresh canine atherosclerotic fresh tissues were 55 pm at 60 mJ/mm2, 60 pm at tissue (noncalcified) were 140 pm at 950 mJ/ 130 mJ/mm2, and 105 pm at 640 mJ/mm2(Fig. 2). mm2, 200 pm at 1,500 mJ/mm2, and 265 pm at The greatest areas of thermal effect seen in the 4,700 mJ/mm2. fresh tissue at 180, 390, and 4,700 mJ/mm2 were 185, 225, and 265 pm, respectively. There was @Switched Mode The ablation thresholds in the QSM were apalso no difference noted in normal vs. arteriosclerotic canine tissues. In fixed canine tissues the proximately 75 and 180 mJ/mm2 in fresh and prethermal coagulation was consistently less than served canine tissues, respectively. As in the with the fresh tissue, with minimal cell damage FRM, the amount of tissue removed and the ther(
Holmium:YAG Laser Ablation of Vascular Tissue George E. Kopchok, BS, Rodney A. White, MD, Marwan Tabbara, Vahid Saadatmanesh, MS, and Shi-Kaung Peng, MD, PhD
MD,
Depaflments of Surgery (G.E. K., R.A.W., M. T , V.S.) and Pathology (S-k.P), Harbor-UCLA Medical Center, Torrance, California 90509
The ablation of atherosclerotic lesions without collateral thermal or shock wave damage is thought to be a key element for successful laser angioplasty. This study evaluated the effectiveness of pulsed ho1mium:YAG laser (2.1 pm wavelength) for this application. Fresh normal tissue (n = 139) and arteriosclerotic canine arteries (n = 21) as well as formalin-preserved normal canine (n = 31) and atherosclerotic human arteries (n = 177) were irradiated under saline via a 600 pm diameter fiber placed perpendicular to the intimal surface with 0-10 gm of force. The laser was operated in the free running mode (FRM; 250 psec pulsewidth, 5 Hz, 30-7,100 mJ/mm2) and in the Q-switched mode (QSM; 200 nsec pulsewidth, 6 Hz, 30-1,100 mJ/mm2).Following the experiments, the samples were prepared for histologic and morphometric analysis. Ablation thresholds in the FRM were 60 and 180 mJ/mm2 in fresh and preserved canine tissue, respectively. Ablation thresholds in the QSM for fresh and preserved canine tissues were 75 and 180 mJ/mm2, respectively. Thresholds for human atherosclerotic tissue were dependent on the amount of calcification. In the QSM and FRM, there were no samples that could not be penetrated at 1,100 mJ/mm2 and above. Histologic examination of the FRM samples revealed confined columns of tissue ablation, with approximately 55-250 pm and 70-140 pm zones of thermal effect being apparent in the fresh and formalinpreserved samples, respectively. The QSM samples (fresh and preserved) revealed minimal thermal effect (0-20 pm) to the intima and media (smooth muscle cells) between 300 and 1,100 mJ/ mm2, whereas thermal effect in the adventitia (collagen) was 64172 pm with the same energy fluence. We conclude 1) that the thermal effect can be optimized by shortening the pulsewidth (Q switching at 200 nsec) or by limiting the total energy delivered and 2) that controlled ablation of both normal and atherosclerotic vascular tissue can be accomplished with the ho1mium:YAG laser. Key words: Ho:YAG laser, ablation threshold, vascular changes, Q-switchedmode, free running mode
INTRODUCTION
The ability to deliver energy through a thin flexible fiberoptic is a key prerequisite for the use of lasers in vascular surgery. A laser that can penetrate or ablate various components of atheroSClerOtiC plaque including Calcium with minimal for arOr acoustic effect may be terial angioplasty. Pulsed lasers such as Excimer [l,2], Er:YAG [3], ho1mium:YAG [4],and hol0 1990 Wiley-Liss, Inc.
mium:YSGG [51 have been shown t o meet these criteria. The ho1mium:YAG laser, with its wavelength of 2.1 ~ m may , also satisfy these condi Accepted for publication June 19, 1990, Address reprint requests to Rodney A. White, MD, Chief, Vascular Surgery, Harbor-UCLA Medical Center, 1000 West Carson Street, Torrance, CA 90509.
406
Kopchok et al.
tions. It has a n absorption coefficient (a)in pure water of 30 cm-' and a depth of penetration of 333 pm, which accentuates the tissue effects at this wavelength. The potential of shallow penetration depth coupled with high pulse energies makes the Ho:YAG laser a promising candidate for laser angioplasty . This study evaluated the Ho:YAG (2.1 pm) laser for use in vascular angioplasty. In vitro tissue (fresh and formalin-preserved) effects were investigated in the free running mode (FRM; 250 psec pulsewidth) and Q-switched mode (QSM; 200 nsec pulsewidth). Normal and arteriosclerotic canine and atherosclerotic human tissues were evaluated. MATERIALS AND METHODS Tissue Samples
Both canine and human tissues were evaluated. Fresh samples were studied immediately following removal, and preserved samples were treated with 10% formalin before being exposed to laser energy. Canine arteriosclerotic arteries were created by a combination of mechanical stripping of the intima and breaking of the elastic lamina of the iliac and femoral arteries and a high-cholesterol diet [6]. Human arteries were obtained from surgical specimens following disposal or from cadavers donated for medical teaching and research. All animals received humane care as outlined by the National Research Council's Guide for the Care and Use of Laboratory Animals (NIH Publication 62-43, revised 1985) and the Principles of Laboratory Animal Care formulated by the National Society of Medical Research. Laser
The Ho:YAG laser system (Trimedyne, Inc., Tustin, CA) (2.1 pm wavelength) was used in two different modes of operation. In the FRM, the laser energy was delivered with 250 psec pulsewidths (full width at half maximum) at 5 Hz. Six hundred micrometer diameter low-OH silica f i bers were used to deliver the energy. The energy fluences used ranged from 30 to 7,100 mJ/mm2 per pulse. In the QSM, the energy was delivered in a 200 nsec pulsewidth (full width at half maximum) at 6 Hz. A polished 600 pm diameter lowOH silica fiber was used to deliver the energy fluence, which ranged from 30 to 1,100 mJ/mm2 per pulse. In both modes, 100 pulses of laser energy was applied at five or six pulses per second.
Fig. 1. Experimental apparatus used to hold the vascular tissue samples perpendicular to the laser fiber a t a constant force. Zero or ten grams of force was added to the weight of the fiber.
Ablation Thresholds
To control the precision of energy delivery and the amount of force exerted on the tissue, all samples were placed in a specially designed fixture, which controlled the fiber alignment and applied a force of 0 or 10 gm (Fig. 1)to the fiber. The vascular tissues were cut into pieces with dimensions of approximately 1.5 x 2 cm and placed in a holder with the laser energy delivered perpendicular to the tissue. All specimens were immersed in room temperature saline with the fiber in contact with the intimal surface of the arteries prior to laser activation. Morphometric Analysis of Laser-Tissue Interactions
Following laser exposure, tissue effects were observed grossly at x 2 0 and compared to specimens that were prepared for histologic examination using trichrome and Verhoeff VanGieson (elastin) stains. Three or four samples were evaluated at each energy level. The majority of the specimens were cut parallel to the axis of the holes to determine the depth of penetration and the effect of laser energy on different layers of the vessel. To define further the dimensions and configuration of the holes (circular vs. eliptical), selected specimens were cut in a plane perpendicular to the axis of the zone of ablation. The thermal effect was quantified by measuring the maximum lateral distance of abnormal morphology from the crater edge in the media of the vessel. The two
Ho1mium:YAG Laser Ablation
407
TABLE 1. Ablation Threshold (mJ/mm2) Canine Fresha Preserved Free running mode (250 psec) Q-switched mode (200 nsec)
60 75
180 180
Human atherosclerotic Noncalcified Calcified 160 280
700-1,100 480-1,100
"No difference between normal and arteriosclerotic tissues.
types of degenerative effects seen were coagula- through the intima and media but slowed and tion and vacuolization. Distances were measured took much longer to pass through the adventitia. at x 100 using a microplan I1 digital pad (LaboThe configuration of the ablation zone (cirratory Computer Systems, Inc.). cular vs. eliptical) was inconsistant in both the FRM and QSM. Most of the holes were circular in shape (Fig. 51, but frequently identical parameRESULTS ters produced both circular and eliptical holes. Free Running Mode The minimum hole diameters created with a The ablation thresholds in the FRM at 5 Hz 600 pm fiber in fresh and fixed canine tissues were determined to be approximately 60 and 180 ranged from 220 t o 460 pm. mJ/mm2 in fresh and fixed normal canine arterial In the FRM, the energy required to ablate tissues, respectively (Table 1). The amount of tis- the human tissue varied due to differences in sue removed with 100 pulses at this energy was calcification. In the three groups of noncalcified, minimal, and virtually no degenerative changes moderately calcified, and heavily calcified tissue, (thermal or shockwave injury) could be seen mi- the energy fluences required to pass through the croscopically. Increasing the energy fluence from tissues (ablation threshold) were approximately threshold to 1,500 mJ/mm2 appeared to produce 160, 700, and 1,100 mJ/mm2. There were no an initial rise, then a leveling off, in the amount samples found that could not be penetrated from of tissue effect. Smaller zones of coagulation could 1,100 t o 1,800 mJ/mm2. In the higher energy be measured in the intima and media compared to fluence range, the amounts of degenerative the adventitia. The smallest areas of effect seen in changes seen in fresh canine atherosclerotic fresh tissues were 55 pm at 60 mJ/mm2, 60 pm at tissue (noncalcified) were 140 pm at 950 mJ/ 130 mJ/mm2, and 105 pm at 640 mJ/mm2(Fig. 2). mm2, 200 pm at 1,500 mJ/mm2, and 265 pm at The greatest areas of thermal effect seen in the 4,700 mJ/mm2. fresh tissue at 180, 390, and 4,700 mJ/mm2 were 185, 225, and 265 pm, respectively. There was @Switched Mode The ablation thresholds in the QSM were apalso no difference noted in normal vs. arteriosclerotic canine tissues. In fixed canine tissues the proximately 75 and 180 mJ/mm2 in fresh and prethermal coagulation was consistently less than served canine tissues, respectively. As in the with the fresh tissue, with minimal cell damage FRM, the amount of tissue removed and the ther(
