Lasers in Surgery and Medicine 12:576584 (1992)

Effect of Force on Ablation Depth for a XeCl Excimer Laser Beam Delivered by an Optical Fiber in Contact With Arterial Tissue Under Saline Geett H.M. Gijsbers, PhD, Duco G. van den Broecke, Rene L.H. Sprangers, MD, PhD, and Martin J.C. van Gemert, PhD Laser Center, Academic Medical Center, Amsterdam, The Netherlands

The effect of force applied to a 430 pm single fiber, delivering 60 pulses of 308 nm XeCl laser radiation at 20 Hz, on the ablation depth in porcine aortic tissue under saline has been investigated. Energy densities of 8,15,25,28,31,37, and 45 mJ/mm2were used. Force was applied by adding weights from 0 to 10 grams to the fiber. The fiber penetration was monitored by means of a position transducer. At 0 grams, the ablation depth increased linearly with incident energy density, but the fiber did not penetrate the tissue; with any weight added, the fiber penetrated the tissue at energy densities above 15 mJ/mm2. The fiber did not penetrate during the first several pulses, possibly due to gas trapped under the fiber. After these first pulses, a smooth linear advancement of the fiber began, which lasted until the pulse train stopped. The ablation depth increased with increasing energy densities and weights. This effect was lar est above 25 mJ/mm2where the ablation efficiencies (unit mm /J),with weights added to the fiber, were substantially larger than values found in 308 nm ablation experiments described in the literature, which were conducted with either a focused laser beam or a fiber without additional force. The results imply that in 308 nm excimer laser angioplasty, force must be applied to the beam delivery catheter for efficient recanalization, and that experiments performed with a focused beam or without actual penetration of the fiber do not represent the situation encountered in excimer laser angioplasty.

B

0 1992 Wiley-Liss, Inc.

Key words: excimer laser angioplasty, in vitro, contact ablation

tial thermal damage t o the remaining tissue. In these in vitro experiments, the laser beam was During the past few years, excimer laser cor- focused on tissue in air. This invariably resulted in onary angioplasty (ELCA) has become one of the deep craters with very sharp boundaries where the major new modalities under investigation in the crater depth varied linearly with the number of field of interventional cardiology. The results of > pulses delivered to the tissue. Ablation by excimer 2,000 treated patients [l-41have been published, lasers without adjacent tissue injury was considand the clinical results are at least comparable t o ered desirable for removal of atherosclerotic those of conventional PTCA in terms of efficacy plaque. The results of these investigations have and safety. led t o the development of several commercial clinThe development of clinical excimer laser systems has been initiated by investigations of several groups in the mid-1980s [5-111. With Accepted for publication August 12, 1992. pulsed excimer laser beams, these groups all found Address reprint requests to G.H.M. Gijsbers, Laser Center, that it was possible to ablate atherosclerotic Academic Medical Center, Meibergdreef 9, 1105 A2 Amsterplaque efficiently and precisely without substan- dam, The Netherlands. INTRODUCTION

0 1992 Wiley-Liss, Inc.

Force Effect on Ablation Depth for a XeCl Excimer Laser

ical excimer laser systems. All systems employ a XeCl excimer laser emitting 308 nm radiation with “stretched pulses” t o facilitate fiber optic beam delivery. In contrast to the initial in vitro experiments [5-111, in ELCA the laser beam is applied by means of optical fibers or light guides that are brought in contact with the tissue in a liquid blood environment. It has been found [12,131that immersion of the tissue in saline decreases the ablation efficiency considerably when the beam is directed to the tissue through the liquid. In contrast, in ELCA the beam delivery catheter is brought in contact with the tissue and manually advanced during application of the laser pulses. One would expect that in this case the negative effect of the liquid environment on the ablation efficiency would be reduced. In preliminary experiments by our group [131, it was found that excimer laser pulses delivered by a clinical catheter or a bare fiber produced a ragged crater wall in aortic tissue adjacent t o the ablation crater. We believe that this was due to the trapping of gaseous debris under the delivery system. Furthermore, a dramatic increase in ablation efficiency occurred when the fiber was pushed onto the tissue during application of laser energy compared to a fiber that was kept just above the tissue. In ELCA, the laser pulse energy densities that were used by the various groups to remove atherosclerotic plaque show substantial variation. In an early clinical study [ll, energy densities of 30 mJ/mm2 were used. In other studies [2,14-161, 30-64 mJ/mm2 was used, because clinical “experience showed that higher energy densities are often required for successful ablation, particularly with calcified lesions” [21; however, this statement was not substantiated by controlled experiments. Therefore we evaluated the effects of force applied to a single fiber delivering a 308 nm excimer laser beam in contact with cardiovascular tissue and the influence of incident energy density on the ablation efficiency. We used a 430 pm single fiber and porcine aortic tissue. Ablation depths, defined as the maximum depths of the thus created craters, were measured by means of light microscopy and the movement of the fiber was recorded by a position transducer. MATERIALS AND METHODS Laser

The laser system used was the Dymer 200 XeCl Excimer laser system (Advanced Interven-

577

Fig. 1. Scheme of the experimental setup. A 430-pm core diameter fiber delivering 308 nm excimer laser pulses is fixed in a shaft and put on the tissue immersed in saline. Different weights can be added to the shaft to control the force of the fiber on the tissue. The movement of the shaft is recorded by a position transducer connected to a personal computer.

tional Systems (AIS), Irvine, CAI. The laser emitted pulses of 220 ns (FWHM) at a wavelength of 308 nm. The pulsed laser beam was delivered to the tissue by means of a single quartz fiber with a core diameter of 430 pm, which was stripped at the end. Pulses were delivered at a frequency of 20 Hz during 3 seconds, as recommended by AIS in their clinical protocol. Energy densities of 8, 15, 25, 28, 31, 37, and 45 mJ/mm2 at the distal fiber end were used, starting at the threshold energy density of about 8 mJ/mm2; 37 and 45 mJ/mm2 were used because of their recommended use in clinical ELCA. Pulse energies were measured with a Molectron radiometer supplied with the laser system. The energy measurements were believed to be accurate within 10%. Measurement of the Ablation Depth

The experimental setup is shown in Figure 1.The arrangement is similar to a system used by Verdaasdonk et al. [17] in their study of the penetration properties of hot tips and transparent contact probes delivering CW laser beams. The fiber delivering the laser pulses was fixed in a shaft. The distal end of the fiber was free t o be in contact with the tissue. The shaft moved freely in the vertical direction and was kept in vertical position by means of a thin wire guided by pulleys and a counter weight. On a table fixed on the shaft, weights were added to control the force

Gijsbers et al. line represents the line of both measurements producing equal depths. These results show a quite reasonable agreement between the two measurements with differences up to about 28% at the larger depths, but with an average difference (determined for all points) of less than 10%. Therefore we considered it reasonable to extrapolate the penetration curves to estimate the ablation depths after 60 pulses at incident energy densities and weights where perforation occurred at less than 60 pulses. In these cases the linear descents as observed in the penetration curves, were 0 0.2 0.4 0.6 0.8 1 1.2 1.4 assumed t o last until 3 seconds after the start of depth by microscopy (mm) the laser pulse train. The ablation depth was calculated to be the penetration extrapolated at 3 Fig. 2. Maximum penetration depths determined from the seconds plus the contribution of a delayed peneposition transducer graphs vs. ablation depths obtained by tration of 150 pm, which was consistently found light microscopy. The solid line represents the line where to occur (see Results). both depths are equal. At each set of incident energy density and weight, the penetration behavior and ablation with which the fiber was pushed onto the tissue. depth were determined 10 times in a single piece The weights used were 0,2,4,6,8, and 10 g. With of aortic wall tissue. All data presented are given 0 g, the fiber was put in contact with the tissue. as the average of 10 measurements -t the stanDuring application of the laser pulses the vertical dard deviation. movement of the fiber into the tissue was recorded by a position transducer (7 DCDT-250, Tissue Porcine aortas were obtained from a slaughHewlett Packard, Palo Alto, CA), connected t o a data recorder (Keithley System 570, Keithley In- terhouse and kept deep frozen. Before the experstruments, Cleveland, OH) and a personal com- iments the tissue was defrosted and kept in saputer. The tissue was fixed on a holder that was line. Porcine aortic tissue was used instead of postmortem human aortic tissue since it was immersed in saline in a glass reservoir. When no perforation of the tissue occurred, much more homogeneous and reproducible than the ablation depths were measured by means of the human tissue. The vessel wall thickness varincident light microscopy at a total magnification ied between 1.3 and 1.6 mm. The fiber was placed of 100, immediately after application of the laser on the intimal side. pulses. The craters were colored with a drop of Evan's Blue for increased contrast, the tissue RESULTS samples were illuminated from below, and a At the start of the experiments the threshold strong light source was directed on the surface of the samples. With the bottom and the rim of the energy density for ablation was determined. This crater now clearly visible, the microscope stage threshold was defined as the incident energy denwas adjusted to give a sharp image of the rim and sity that produced a crater that was just observed the bottom. The difference in the microscope stage by incident light microscopy after 60 pulses and 0 g of weight. The energy threshold was established adjustment yielded the ablation depth. When perforation occurred before 60 pulses t o be 8 mJ/mm2. To evaluate the cutting effect of the bare fiwere completed, the depth that would have been ber itself, the maximum weight of 10 g was added observed after 60 pulses was extrapolated from the penetration graphs measured with the posi- to the fiber without firing the laser. After applytion transducer. This depth was found to increase ing the fiber to the aortic tissue for several minlinearly with the number of pulses once penetra- utes, an impression of the fiber in the tissue was tion had started. In Figure 2, the depth after 60 noted by bare sight. However, after a few minutes pulses as measured from the penetration graphs this impression had disappeared, and even under is compared with the corresponding ablation the microscope the impression could no longer be depths determined by light microscopy. The solid seen. This was also true for applying the fiber

578

depth by position transducer (mm) 1.4 i /'

I

Force Effect on Ablation Depth for a XeCl Excimer Laser 579 with a 10 g weight on the media from which the 0.2 intimal layer and the internal elastic lamina had 0been removed by blunt dissection. Therefore we -0.2 neglected possible cutting effects of the fiber. The start of the pulse train was accompanied -0.4 by the release of small bubbles of insoluble gas a) -0.6from the tissue at the site of the fiber. Sometimes -0.8we even noted some gas bubbles escaping from the side of the tissue sample, several millimeters from 25 mJ/mm2, 2 g the fiber site. The release of gas not only persisted -1.2 , during the pulse train, but some gas escaped also 0 2 4 6 8 10 12 14 16 time (s) during a few seconds after the pulse train. At 0 g and for all incident energy densities, the fiber did not penetrate the tissue, as observed from the recordings of the position transducer sigPenetration (mm) 0.2 1 nal. At 8 and 15 mJ/mm2the fiber penetrated less than 50 km at all weights; this was difficult to distinguish from the noise in the position transducer signal. Therefore these data were omitted -0.4 in the following analysis. Three representative examples of penetration of the fiber at 25, 37, and 45 mJ/mm2 and -1 with 2 g of weight added, as derived from the po37 mJ/mm2, 2 g sition transducer, are shown in Figure 3a-c. The pulse train was started at time zero and was finished at 3 seconds. Immediately after the start of the pulse train, the fiber was lifted somewhat (Fig. 3a,b) or stayed in position. After several penetration (mm) pulses, depending on incident energy density and o.2 1 weight, a linear descent of the fiber began that lasted until the end of the pulse train. After the laser stopped, there was still some residual pene-o.21 -0.4 tration movement. The penetration graphs are analyzed according t o Figure 4. The time before the fiber started t o penetrate is given as t,,. This time is related to -1 the 5% penetration compared to the total penetra1.2 0 2 4 6 8 10 45 12 mJ:mm 14’, 2 g16 c, tion depth d,,. d, is defined as the average of the position transducer signal between 6 and 16 time (s) seconds after the start of the pulse train, relative to the zero level before the pulse train. The 5% Fig. 3. Examples of the penetration behavior of the fiber as a point could accurately be determined relative t o function of time. At time zero a train of excimer laser pulses the noise in the transducer signal. is applied during 3 seconds at 20 Hz. A weight of 2 g is used. The penetration of the fiber from t,, to the (a) Energy density 25 mJ/mm2. (b) Energy density 37 mJ/ moment the laser was stopped at 3 seconds is de- mm2. (c) Energy density 45 mJ/mm2. fined as dlaser.The penetration velocity of the linear descent during this time is given as:

-I

-o.21 i

\,

:

penetration velocity

:

=

dlaser (mds) 3 sec-t,,

The late penetration of the fiber after the laser was stopped is defined as dmax-dss,where d,, is the

I

fiber penetration at the moment the pulse train ends. The “time to start penetration”, tsp,as a function of weight added t o the fiber at 25,28,31, 37 and 45 mJ/mm2 is shown in Figure 5. Even at

Gijsbers et al.

580 0.2

penetration (mm)

1

2

45 mJlmm

0.6

/h''

~

2

37 m J I m m

2

31 m J I m m 2

2 8 mJlmm 2

2 5 mJImm

0

start laser

~

stop laser

4

6

8

10

12

14

16

-

time (s)

0

time to start penetration

I 1.5

'I

(5)

T

7

25 mJ/mm

2

28 mJ/mm2 31 mJ/mm'

0.5

00

37 mJ/mm2 4 5 mJ/mm2

- r

2

4

6

8

7

10

12

weight (gr)

Fig. 4. Parameters used to analyze the penetration graphs. d,,: the total penetration compared to the zero level. d38:the fiber penetration at the end of the 3 second pulse train. d,,d3s:the late penetration that occurs after the end of the pulse train. tSp:the time where the fiber starts to penetrate, at 5% of the total penetration d,,. dlaser:penetration after the uniform descent.

2.5 -1

2

4

~

6

8

10

12

weight (9)

Fig. 5. The time between the start of the pulse train and the actual fiber penetration as a function of weight added to the fiber at different energy densities (data points: mean 2 s.d., n = 10).

45 mJ/mm2 and using 10 g it took at least 4 pulses t o start the penetration process. The measured penetration velocities as a function of weight added to the fiber at the same incident energy densities are presented in Figure 6. At the lower energy densities (25, 28, and 31 mJ/mm2) the velocity was only very slightly dependent on the applied weight. At 37 mJ/mm2 there was a marked increase in velocity when increasing the weight from 2 t o 10 g. At 45 mJ/mm2 there was no statistically significant change in the penetration velocity at weights above 6 g. The late penetrations dmax-dss,as a function

Fig. 6. The penetration velocity as a function of weight at different energy densities. The penetration velocity is defined as

dlaser (see Fig. 3) (data points: mean 3 sec-t,,

2

s.d., n = 10).

of weight, again at 2 5 , 2 8 , 3 1 , 3 7 , and 45 mJ/mm2 are shown in Figure 7. These late penetrations of - 150 pm were remarkably independent of weight and incident energy density within measurement uncertainty. The ablation depth after 60 pulses as a function of incident energy density per pulse is shown in Figure 8. For energy densities up to 31 mJ/ mm2, it was possible t o use weights up to 10 g without causing perforation of the aortic wall after 60 pulses. At 37 mJ/mm2, weights of 8 and 10 g always resulted in perforation, whereas at 45 mJ/mm2 perforation occurred for weights L 4 g. In these cases the ablation depths after 60 pulses were extrapolated from the penetration curves as explained in the Materials and Methods section. The extrapolated data are connected by dashed lines in Figures 8 and 9. Figure 9 presents the same data as Figure 8, but now the ablation depths are given as a function of weight added to the fiber. DISCUSSION

When force was applied to the fiber, the ablation depth was always larger than the depth using no additional weight (Figs. 8, 9). We hypothesize that force is necessary to push aside the gas and debris that is created below the fiber tip. For example, at 37 mJ/mm2 the use of 2 g increased the ablation depth by a factor of 2.8 compared t o the depth at 0 g, and even a factor of 3.1 at 45 mJ/mm2. For 4 g, these factors were 3.9 and

Force Effect on Ablation Depth for a XeCl Excimer Laser 250

31

1

2.5

T

1.5

1504

50

T

1

21

200

100

581

ablation depth (mm)

displacement after laser (micrometers)

-1

f

4 6 mJlrnm’

++ 3 7 m J l m m Z

-I

-+ 31 m J l m m Z

0

4 2 8 mJlmmZ

+ 25 0

7

0

2

4

8

4 6 weight ( g r )

a

10

mJlmm’

, 6

2

I

10

12

weight (gr)

Fig. 7. The late penetration of the fiber after the end of the pulse train (dmax-ds8, see Fig. 3) as a function of the weight added to the fiber (data points: mean s.d., n = 10).

Fig. 9. Ablation depth after 60 pulses as a function of weight added to the fiber a t different energy densities. The values connected by the dotted lines are derived by extrapolation of the penetration curves from the position transducer (data points: mean s.d,, = -+

*

ablation depth (mm)

2.5

4 I

w.

2 Qr.

0.5

0 gr. 1

0

10 20 30 40 energy density (mJ/mm 2/pulse)

50

Fig. 8. Ablation depth after 60 pulses as a function of energy density per pulse a t different weights added to the fiber. The values connected by the dotted lines are derived by extrapolation of the penetration curves from the position transducer (data points: mean ? s.d., n = 10).

6.3, respectively. The ablation depths increased more or less linearly with weight, except at 45 mJ/mm2 where the ablation depth is constant at weights of 6 g and more. The slope of the ablation depth versus incident energy density curves (Fig. 8) can be interpreted as an “ablation efficiency” (unit mm3/J), i.e., the increase of depth with increasing energy density. The larger the slope, the more efficient the ablation process. When no weight is added, the ablation efficiency is nearly constant over the range of incident energy densities we studied. However, when weights are added to the fiber, the ablation efficiency is distinctly increased at incident energy densities above 25 mJ/mm2 relative

to the ablation efficiency at lower incident energy densities. It seems as if the tissue “gives way” much more easily with the fiber pushed into the tissue at the larger energy densities. Preliminary results of our group (unpublished) indicates that above 25 mJ/mm2 gas production increases more than linearly with the incident energy density. This may point to an additional laser tissue interaction mechanism at these larger energy densities. This increased gas production would increase the mechanical strain on the tissue, and this may explain the observed increased ablation efficiencies at the larger energy densities. Experiments to quantify this are underway. We have shown that it takes a number of pulses t o start the actual penetration of the fiber into the tissue (Fig. 5 ) . We hypothesize that this is due to the initial trapping of the gaseous debris under the fiber. Only when the gas volume is large enough can it push aside the remaining tissue, allowing the fiber to descend. The deeper tissue layers have already been affected by the laser light and the mechanical strain of the gas and are less resistant t o the fiber. When this happens, the fiber can penetrate into the tissue and the deeper layers can also be ablated. Once fiber advancement has started, a moving boundary of gaseous debris is created in front of the fiber. At the end of the pulse train some gas is still present under the fiber which will eventually escape. This may explain the late penetration (Fig. 7) after the pulse train. According t o Figure 7, the thickness of the gaseous layer must be about 150 pm. With increasing force the gaseous debris will

Gijsbers et al.

582

TABLE 1. Ablation Efficiencies of XeCl Lasers for Soft Tissues*

Laser

Tissue

Ablation efficiency mm3/J 0.1 0.1

Range of linearity mJ/mm2 25 40

Comments One energy used, ablation efficiency calculated as depth per pulsehergy density One energy used, ablation efficiency calculated as depth per pulsehergy density Below 20 mJ/mm2 smaller slope; ablation efficiency difficult t o estimate Below 12 mJ/mm2 slightly smaller slope. Two ranges where depth is linear.

XeCl40 ns 10 Hz, 1 mm fiber touching tissue [121

porcine aorta in air

XeCl40 ns 10 Hz, 1 mm fiber touching tissue [121

porcine aorta under saline

0.015 0.015

25 40

XeCl 15 ns 2 Hz focused beam [181

bovine cartilage in air

0.23

20-64

XeCl 15 ns 2 Hz focused beam [181 XeCl 17 ns focused beam [19,201 XeCl 20 ns 25 Hz focused beam [211 XeCl85 ns 20 Hz 400 pm fiber touching tissue [221 XeCl35 ns 10 Hz focused beam [231 XeCl35 ns 10 Hz focused beam [231 XeCl220 ns 20 Hz, 430 pm fiber in contact this work

human normal aorta in air

0.38

12-52

human aorta in air

0.19 0.43 0.29

9-20 20-60 12-24

human intervertebral disc in air human aorta in air

0.22

6-55

0.33

16-60

human aorta in air

0.25

20-70

Aorta with visible intimal thickening Yellow opaque aorta, no calcium

porcine aorta under saline

0.14

8-45

og

0.16 0.68 0.28 1.78 0.40 2.18

8-25 31-45 8-25 31-45 8-25 31-45

2g 2g 4g 4g 6g 6g 8 and 10 g comparable to 6 g

human skin in air

'The ablation efficiency is taken to be the slope of the crater depth per pulse as a function of the laser energy density. Where this was not possible, the ablation efficiency is given as the depth per pulsehergy density.

sooner be forced t o escape out of the region in front of the fiber tip, and the moving boundary will be established earlier. Obviously, the increase in total penetration with weight is partly due t o a decrease in the number of pulses after which penetration actually starts (Fig. 5 ) and partly due to the increase in penetration velocity of the fiber (Fig. 6). The dependence of the penetration velocity on the added weight (Fig. 61, especially the flattening of the curve at 45 mJ/mm2, is not yet understood but is undoubtedly related t o the mechanical properties of the tissue. To compare our results with those of other groups, we have summarized in Table 1the ablation eficiencies as defined above, from the published ablation depth vs. incident energy density curves from those groups and from our results (Fig. 8). All authors except Taylor et al. [121 studied ablation in air and without added force when a fiber was used. Interestingly, Wieshammer et

al. [18] and Harnoss et al. [19,20] found in air an increased ablation efficiency at larger incident energy densities as we have also observed when we added weight t o the fiber. This might again indicate a change in the ablation mechanism. However, Kaufmann and Hibst [21], Wolgin et al. [22], and Taylor et al. [23] found a linear relation in the incident energy density ranges studied. It is not clear what causes these differences in ablational behavior. The ablation efficiencies in air from Taylor et al. [12] are small compared to the efficiencies by other authors, but in later work [231 they found ablation efficiencies in the same range as found by other authors (see Table 1).Striking is that the ablation efficiencies of tissue under saline found by Taylor et al. [12] are about seven times smaller than those of tissue in air. The authors suggest that the liquid acts as a barrier to suppress the rapid ejection of ablated material.

Force Effect on Ablation Depth for a XeCl Excimer Laser 583 The ablation efficiencies in air measured by the Liem for adapting the data handling software to other authors are in the range of 0.19-0.43 mm3/ our needs, and Mr. P. van Rijn for providing the J, which is larger than the 0.14 mm3/J measured porcine aortic tissues when needed. under saline in this work, when no weight was added to the fiber. This might also be due t o the dampening effect of the surrounding liquid. How- REFERENCES ever, when a weight is applied, the ablation effi- 1. Karsch KR, Haase KK, Voelker W, Baumbach A, Mauser ciencies are also in the same range as found for M, Seipel L. Percutaneous coronary excimer laser angioplasty in patients with stable and unstable angina pecthe tissues in air for incident energy densities betoris: Acute results and incidence of restenosis during low 25 mJ/mm2. Above 25 mJ/mm2, applying 6-month follow up. Circulation 1990; 81:1849-1859. force to the fiber enhanced the ablation efficien- 2. Cook SL, Eigler NL, Shefer A, Goldenberg T, Forrester cies substantially, even far beyond the values obJS, Litvack F. Percutaneous excimer laser coronary angioplasty of lesions not ideal for balloon angioplasty. Cirserved in air. This suggests that experiments perculation 1991; 84:632-643. formed with a focused beam or without actual penetration of the fiber do not represent the situ- 3. Holmes DR, Litvack F, Goldenberg T, Bresnahan JF, Cummins FE, Margolis JR. Excimer coronary laser anation encountered in excimer laser angioplasty. gioplasty (ELCA) registry: Lesion length and outcome. CONCLUSIONS

Penetration of a single fiber in contact with porcine aortic tissue that delivers XeCl excimer laser pulses increases with increasing weights applied t o the fiber. This effect is largest for weights up t o 4 g. The ablation efficiency (in mm3/J) increases at incident energy densities above 25 mJ/ mm2 when weights are applied, which may indicate a change in the ablation mechanism. It takes several pulses before fiber penetration starts. This is attributed to the gas trapped under the fiber. Once penetration starts, the penetration velocity is constant during the remainder of the pulse train. Clinically, our results imply that in order t o cross a lesion with a laser beam delivering catheter, at least some force on the catheter tip must be applied. We have seen that for a single 430-pm fiber, the ablation efficiency at 37 mJ/mm2 is approximately the same as at 45 mJ/mm2. This does not explain why the use of energy densities of 45 mJ/mm2 and higher in ELCA give better clinical results [2]. However, we have seen that the increase in penetration depth when adding weight from 0 to 4 g is a factor of 3.9 at 37 mJ/mm2 and a factor of 6.3 at 45 mJ/mm2.This means that the effect of force applied t o the laser beam delivery system is much more profound at larger energy densities. This might make the difference in the ability t o ablate hard fibrous or calcified plaques and therefore in obtaining clinical success. ACKNOWLEDGMENTS

The authors thank Arie Steenbeek for the excellent construction of the equipment, Koen

Circulation 1991; 84 (suppl 1I):II-362 (abstract). 4. Sanborn TA, Bittl JA, Siege1 RM, Kramer BL, Tcheng JE. Lack of effect of lesion severity on clinical success and complication rates with percutaneous excimer laser coronary angioplasty (PELCA). Circulation 1991; 84 (suppl II):I1-362 (abstract). 5. Linsker R, Srinivasan R, Wynne JJ, Alonso DR. Far-ultraviolet laser ablation of atherosclerotic lesions. Lasers Surg Med 1984; 4:201-206. 6. Grundfest WS, Litvack F, Goldenberg T, Sherman T, Morgenstern L, Carroll R, Fishbein M, Forrester J , Margitan J , McDermid S, Pacala TJ, Rider DM, Laudenslager JB. Pulses ultraviolet lasers and the potential for safe laser angioplasty. Am J Surg 1985; 150:220-226. 7. Grundfest WS, Litvack F, Forrester JS, Goldenberg T, Swan HJC, Morgenstern L, Fishbein M, McDermid IS, Rider DM, Pacala TJ, Laudenslager JB. Laser ablation of human atherosclerotic plaque without adjacent tissue injury. J Am Coll Cardiol 1985; 5:929-33. 8. Isner JM, Donaldson RF, Deckelbaum LI, Clarke RH, Laliberte SM, Ucci AA, Salem DN, Konstam MA. The excimer laser: Gross, light microscopic and ultrastructural analysis of potential advantages for use in laser therapy of cardiovascular disease. J Am Coll Cardiol 1985; 6:1102-1109. 9. Sartori M, Henry PD, Sauerbrey R, Tittel FK, Weilbaecher D, Roberts R. Tissue interactions and measurement of ablation rates with ultraviolet and visible lasers in canine and human arteries. Lasers Surg Med 1987; 7:300-306. 10. Singleton DL, Paraskevopoulos G, Taylor RS, Higginson LA. Excimer laser angioplasty: Tissue ablation, arterial response, and fiber optic delivery. IEEE J Quant Elec, QE-23, 1987; 1772-1782. 11. Litvack F, Grundfest WS, Goldenberg T, Laudenslager J, Pacala T, Segalowitz J, Forrester JS. Pulsed laser angioplasty: Wavelength power and energy dependencies relevant to clinical application. Lasers Surg Med 1988; 8: 60-65. 12. Taylor RS, Leopold KE, Walley VM, Higginson LAJ. XeCl laser ablation of cardiovascular tissue: practical considerations. Lasers Life Sci 1988; 2:227-241. 13. Gijsbers GHM, Sprangers RLH, Keijzer M, De Bakker JMT, Van Leeuwen TG, Verdaasdonk RM, Borst C, Van Gemert MJC. Some laser-tissue interactions in 308 nm

584

14.

15.

16.

17.

18.

Gijsbers et al.

excimer laser coronary angioplasty. J Interven Cardiol 1990; 3~231-241. Sanborn TA, Bittl JA, Siege1 RM, Kramer BL, Tcheng JE. Lack of effect of lesion severity on clinical success and complication rates with percutaneous excimer laser coronary angioplasty (PELCA). Circulation 1991; 84 (suppl 1I):II-362 (abstract). Buchwald A, Werner GS, Unterberg C, Wiegand V. Restenosis after excimer laser angioplasty of coronary stenoses and chronic total occlusions. Circulation 1990; 82 (suppl 1II):III-313 (abstract). Eigler N, Cook S , Kent K, Margolis J, Rothbaum D, Webb-Peploe M, Smith D, Vater M, Hestrin L, Litvack F. Excimer laser angioplasty of ostial coronary stenosis: results of a multicenter study. Circulation 1990; 82 (SUPPl 1II):III-1 (abstract). Verdaasdonk RM, Jansen ED, Holstege FC, Borst C. Mechanism of CW Nd:YAG laser recanalization with modified fiber tips: Influence of temperature and axial force on tissue penetration in vitro. Lasers Surg Med 1991; 11~204-212. Wieshammer S, Hibst R, Bellekens M, Steiner R. Ultraviolet laser ablation of biologic tissue: Quantitation of

19.

20.

21. 22.

23,

etch rate as a function of incident fluence. Lasers Life Sci 1988; 2:125-135. Harnoss BM, Kar H, Zuhlke H, Berlien HP, Muller G, Haring R. A comparative study on the physical aspects of the ablation behavior of short-pulsed laser in arteriosclerotic vessels in vitro. Lasers in Medicine and Surgery (Germany) 1990; 3:105-110. Muller G, Harnoss M, Kar H, Dorschel K, Berlien HP. Photoablation a question of wavelength? In: J. Marshall, ed. “Laser Technology in Ophthalmology.” Berkeley: Kugler, 1988, pp 221-227. Kaufmann R, Hibst R. Pulsed Er:YAG- and 308 nm UVexcimer laser: an in vitro and in vivo study of skin ablative effects. Lasers Surg Med 1989; 9:132-140. Wolgin M, Finkenberg J, Papaioannou T, Segil C, Soma C, Grundfest W. Excimer ablation of human intervertebra1 disc at 308 nanometers. Lasers Surg Med 1989; 9: 124-131. Taylor RS, Higginson LAJ, Leopold KE. Dependence of the XeCl laser cut rate of plaque on the degree of calcification, laser fluence, and optical duration. Lasers Surg Med 1990; 10:414-419.