Lasers in Surgery and Medicine 11:26-34 (1991)

Noncontact Tissue Ablation by Holmium:YSGG Laser Pulses in Blood Ton G. van Leeuwen, MSC, Maurits J. van der Veen, Rudolf M. Verdaasdonk, PhD, and Cornelius Borst, MD, PhD Department of Cardiology, Heart-Lung Institute, University Hospital Utrecht (T.G.v.L., M.J. v.d. K , C.B.), Interuniversity Cardiology Institute of the Netherlands (R.M. K), Utrecht, The Netherlands Key words: pulsed laser angioplasty, midinfrared laser ablation, transmission, Holmium laser

To assess the feasibility of intra-arterial tissue ablation by Holmium:YSGG laser pulses (2.1 pm) in a noncontact mode, the transmission of the laser pulses through saline and blood was measured. The temporal interaction between the 500 p s laser pulse and saline at the fiber tip was investigated with time-resolved flash photography. The penetration depth in blood and saline depended on the fiber output energy. In blood at 37”C, the penetration depth varied from 1.2 to 2.1 mm for intensities of 3.1 to 12.4 J/mm2 per pulse, respectively, whereas its theoretical value for water is 0.33 mm, which is based on the measured absorption coefficient of 3.0 2 O.l/mm. The large penetration depth was due to the development of a transparent vapour cavity around the fiber tip. In saline, its maximum length was 4.7 mm. Its maximum width was 2.8 mm. The lifetime of the cavity was 450 ps. In blood, ablation of porcine aorta was feasible at a distance of 3 mm. Large fissures observed in adjacent tissue are likely to be caused by the expansion of the vapour cavity. We conclude that, due to a “Moses effect in the microsecond region,” Holmium: YSGG tissue ablation is possible through at least 2.7 mm of blood. INTRODUCTION

etry, The presence and distribution of atherosclerotic plaque can be distinguished. Guidance of laser ablation of plaque by intravascular ultrasound [13,14] may reduce the risk of perforation o f the arterial wall. However, ablation guided by ultrasound imaging may require ablation in a noncontact mode and blood between the fiber tip and the target may impede effective tissue ablation. To assess the feasibility of intra-arterial tissue ablation in a noncontact mode, the energy transmission of the Holmium laser pulses through

For precise tissue ablation, pulsed lasers are t o be preferred t o continuous wave lasers [1,21. At present three types of pulsed lasers are used in angioplasty: the excimer lasers in the ultra violet region [3,4], the pulsed dye lasers in the visible region [5,6], and the pulsed Nd-YAG laser in the near infrared [71. Recently lasers with other wavelengths in the infrared region have become available [8,9]. In particular, the Holmium laser may be an alternative to the excimer laser for the ablation o f superficial tissue layers with little thermal injury t o adjacent tissue [10,111. Rather than breaking molecular bonds, the Holmium laser pulse vaporizes tissue water. The energy absorption in water of the Holmium laser wave- Accepted for publication October 1, 1990. length (2.1 km) is 100 times larger than that of Address reprint requests to Ton van Leeuwen, Experimental the Nd-YAG laser wavelength (1064 nm) [121. Cardiology Laboratory, Department of Cardiology, HeartIntravascular ultrasound provides detailed Lung Institute, Room E02.562, University Hospital Utrecht, images of the arterial lumen and the wall geom- Postbus 85500, 3508 GA Utrecht, The Netherlands. 0 1991 Wiley-Liss, Inc.

Ablation by Ho:YSGG saline and blood was measured for high-power densities. We also determined the absorption coefficient of Holmium laser light with low-power densities in saline. To explain discrepancies in the results, the interaction of laser light and water was investigated with time-resolved flash photography. Subsequently, the maximum distance between the fiber tip and the target that still allowed effective tissue ablation was determined. Tissue craters and adjacent tissue damage were investigated histologically. MATERIALS AND METHODS Laser

Pulses Through Blood 27 depth was determined by interpolating the transmission data. At low intensities the fiber tip was approximately 10 cm above the saline. The energy at the fiber tip was 0.50 J per pulse. To compensate for the energy losses due t o reflections at the airsaline, saline-glass, and glass-air boundaries, the initial energy per pulse was defined as the energy per pulse transmitted through a saline layer of minimal thickness. The absorption coefficient and the penetration depth were determined. At high intensities the fiber tip was submerged at least 30 mm in saline as well as in human blood. The transmitted energy per pulse was measured with a distance of 0-5 mm between the fiber tip and the bottom of the glass tray. The intensity at the fiber tip was varied from 3.1 t o 12.4 Jimm2 per pulse. The physiological media were circulated with a flow rate of 0.8 liter per minute. We compared the transmission and the penetration depth in saline at temperatures of 25°C and 37°C and in blood at 37°C. The influence of the hydrostatic pressure (0-150 mm Hg) on the transmission was measured.

All experiments were performed with a Holmium-YSGG laser (Schwartz ElectroOptics Inc., Orlando, FL). The laser generated 500 ps long series of ps pulses at a wavelength of 2.1 pm. The full width at half maximum of the envelope of these spikes was 200 bs. The experiments were performed at a repetition rate of 1 Hz. The laser beam was coupled into a low-OH optical fiber, which had a core diameter of 320 pm. The pulse energy at the fiber tip was varied from 0.25 t o 1.00 J . Pulse-to-pulse energy varia- Time Resolved Flash Photography tion was less than 5%. The laser pulse was reThe interaction of saline (or blood) and laser corded with a fast pyro-electric element (Molec- light at the fiber tip was investigated with time tron Detector, Inc., P5-01, Portland) incorporated resolved flash photography. in a circuit which had a rise time of 10 ns. A 25 ps flashlamp (Braun AG, 370 BVC, Frankfurt/Main, Germany) was triggered by a Energy Transmission pulse from the laser at an adjustable delay (0The energy transmission of the Holmium la- 1,500 ps) (Fig. 1).The light flash was detected by ser pulses through saline was measured for low a photodiode (BPW 24, AEG-Telefunken, Heilintensities (I < 0.01 J/mm2 per pulse) and high bronn, Germany) incorporated in a circuit which intensities (I > 1.00 J/mm2 per pulse). The energy had a rise time of 1.5 ps. Simultaneously, the latransmission through blood was measured for ser pulse was detected by the pyro-electric elehigh intensities (I > 1.00 J/mm2 per pulse). The ment. The delay time, defined as the period beenergy per pulse was measured with a laser en- tween the start of the laser pulse and the peak of ergy meter (Ophir Optics Jerusalem, Ltd. , Israel), the light flash, was measured with a dual trace positioned beneath a glass tray. The temporal storage oscilloscope (Tektronix Inc., 7623A, Beashape of the transmitted laser pulse was recorded verton, USA). The fiber tip was submerged at least 30 mm in saline. A glass tray (10 x 10 x 7 with the pyro-electric detector. The transmitted energy per pulse (the mean cm3) contained the saline. A millimeter scale next of ten pulses standard deviation) was measured to the fiber tip was used as reference. During the as a function of the thickness of the saline or blood light flash the photographic emulsion of a photo layer between the fiber tip and the bottom of the camera (Minolta Camera Co., Ltd., X700, Osaka, tray. The transmission was defined as the ratio of Japan) or a CCD chip of the video camera (Panathe energy per pulse transmitted through the me- sonic WV CD132E, Osaka, Japan) was exposed. Acoustic signals were recorded by a hydrodium to the initial energy per pulse. The penetration depth was defined as the depth at which the phone (1-7 MHz, Mediscan, Inc., Connecticut) transmitted energy per pulse had decreased to lie submerged in the saline at a distance of 5 mm of the initial energy per pulse. The penetration from the fiber tip.

*

van Leeuwen et al.

28

-

fiber

4.

,

I

laser

camera 0

100

200

300

400

500

delay

time [microseconds] osci I loscope

Fig. 1. The experimental setup for the time resolved flash photography of the interaction between Holmium laser pulses and saline

Fig. 2. The mean temporal shape of a 0.75 J Ho1mium:YSGG laser pulse (n = 161, represented as the output power (in kW) as a function of time. The dotted line represents the mean temporal shape of a transmitted laser pulse through 1mm saline. T, is the time at which the cavity reached the bottom of the glass tray and T, is the time at which the cavity collapsed. 100 U

In Vitro Noncontact Tissue Ablation

Within 6 hours after death, the thoracic aorta of a .pig.was irradiated oemendicular to the luminal surface. Thereafter the aorta was immersed in formalin 4%. After 24 hours the aortic segments were dehydrated and embedded in paraffin. Sections of 5 pm thickness were cut at intervals of 50 pm. The sections were stained with Haematoxylin and Eosin or Elastin von Gieson and analyzed quantitatively. The ablation threshold of the aorta for Holmium laser pulses was defined as the intensity required for tissue removal. The ablation threshold was determined histologically. An aorta was also ablated through 2 and 3 mm of blood. The output energy at the fiber tip was 1.00 J per pulse. The craters and adjacent tissue were examined histologically as described above. I

"

*

I

s

Y

c

0 m cn Ecn c a L

t-

0

1

2

3

4

5

depth [mm] Fig. 3. The transmission (in percent) of the Holmium laser pulses through saline at 37°C as function of the depth for various laser intensities per pulse. + , I < 0.01 Jimm2; 0, I = 3.1 Jimm' (0.25 Jipulse); A,I = 6.2 J/mm' (0.50 J/pulse); V, I = 9.3 J/mm2 (0.75 Jipulse); 0 , I = 12.4 J/mm2 (1.00 J i pulse).

RESULTS

The mean temporal shape of 16 Holmium: YSGG laser pulses (0.75 J per pulse) is shown in Figure 2. For intensities less than 0.01 J/mm2 per pulse, the transmitted energy per pulse in saline decreased, according to Beer's law, exponentially with distance (Fig. 3, crosses). The absorption coefficient was 3.0 L 0.1 /mm (n=4). Therefore the penetration depth was 0.33 0.01 mm. At high intensities the transmitted energy did not agree with Beer's law (Fig. 4). For intensities between 3.1 and 12.4 J/mm2 per pulse, the

*

penetration depth in saline at 25°C varied from 1.3 t o 2.2 mm, respectively. In saline at 37°C the penetration depth varied from 1.3 t o 2.3 mm, respectively (Figs. 3 and 4, open markers). Note that the initially transmitted energies shown in Figure 4 correspond with 100% in Figure 3. In blood at 37°C we measured the transmitted energy per pulse for intensities between 3.1 and 12.4 J/mm2 (Fig. 4,filled markers). The penetration depth varied from 1.2 t o 2.1 mm, respectively. The penetration depth of the Holmium:

Ablation by Ho:YSGG Pulses Through Blood

L I"

+

0

1

2

3

4

5

depth [mm] Fig. 4. The transmitted energy [mJl per pulse through saline (open markers) and blood (filled markers), both at 37"C, as function of the depth for various laser intensities at the 320 p,m fiber tip. 0 , I = 3.1 Jimm' (0.25 Jipulse); A, I = 6.2 Jimm' (0.50 J/pulse); V, I = 9.3 Jimm' (0.75 Jipulse); 0, I = 12.4 Jimm' (1.00 Jipulse).

YSGG laser pulses in saline at 25°C and 37°C and in blood at 37°C are listed in Table 1. Time-resolved flash photography (Figure 5) revealed vaporization of water in front of the fiber tip and the formation of a vapour cavity during the laser pulse. The water vapour in this cavity was transparent for Holmium laser light [151. The expanding (Fig. 5A-E) and imploding (Fig. 5F-H) cavity was observed in saline at room temperature at varous delay times (0-1,500 ps). The fiber output was 0.75 J per pulse. The maximum horizontal radius of the cavity measured 1.4 mm and was reached at 250 p s (Fig. 5E). At about 300 ps after the beginning of the laser pulse, the cavity reached a maximum length of 4.7 mm (Fig. 5F). After the implosion of the cavity, at approximately 450 ps (Fig. 5H) additional cavities were formed, which subsequently imploded at about 750 ps. From 35 exposures the distance between the bottom of the vapour cavity and the fiber tip as well as the horizontal radius of the vapour cavity were measured as a function of the delay time (Fig. 6). The fiber output was 0.75 J per pulse. The maximum depth of the vapour cavity was reached at 350 p s and it measured 3.4 mm from the bottom of the cavity t o the fiber tip. The dotted line in Figure 2 represents the temporal shape of a 0.75 J laser pulse after transmission through 1mm of saline. The timing of the onset and the end of complete vapour transmis-

29

sion, T, and T,, respectively, were measured for a 0.75 J laser pulse transmitted through various depths of saline and blood (Table 2). Only small acoustic signals were recorded by the hydrophone. The most intense acoustic signals were recorded upon the collapse of the vapour cavity (Fig. 5H) and not at the start of the laser pulse. The influence of the hydrostatic pressure on the transmission of the laser pulses was negligible. We determined histologically that the ablation threshold for porcine aorta was 0.75 t o 0.85 J/mm2. Tissue ablation by single Holmium laser pulses through 2 (Fig. 7) and 3 mm blood was accomplished with a fiber output of 1.00 J/pulse. The depth of the crater after one 1.00 J laser pulse (Fig. 7A) was 0.2 mm. The diameter of the crater was 0.3 mm and the damage zone had a diameter of about 1.0 mm. The crater after two 1.00 J laser pulses through 2 mm of blood is shown in Figure 7B. The depth and width were approximately 0.5 mm and 0.2 mm, respectively. The diameter of the damage zone consisting of fissures and irregularities in surrounding tissue was about 1.7 mm.

DISCUSSION

The aim of this study was t o assess the feasibility of tissue ablation by Ho1mium:YSGG laser pulses in a noncontact mode, which may be useful for ablation guided by intravascular ultrasound. The principal finding of this study was that in blood the penetration depth of high-intensity Holmium laser pulses was three t o seven times larger than the expected 0.33 mm (Table 11, the depth corresponding with the absorption coefficient of water of 3.0 k 0.1 /mm. The discrepancy is due t o a cavity created around the fiber during the laser pulse (Fig. 5). The water vapour in the cavity is transparent t o Holmium laser light 1151. The divergence (4") of the laser beam was neglected in measuring the thickness of the liquid layer. This systematic error (

Noncontact tissue ablation by holmium:YSGG laser pulses in blood.

To assess the feasibility of intra-arterial tissue ablation by Holmium:YSGG laser pulses (2.1 microns) in a noncontact mode, the transmission of the l...
857KB Sizes 0 Downloads 0 Views