Lasers in Surgery and Medicine 19:23—31 (1996)
`
`Time-Resolved Observations of Shock
`
`Waves and Cavitation Bubbles Generated
`
`by Femtosecond Laser Pulses in Corneal
`Tissue and Water
`
`Tibor Juhasz, PhD, George A. Kastis, as, Carlos Suarez, MS, Zsolt Bor, PhD,
`and Walter E. Bron, PhD
`
`Department of Physics, University of California, lrvine (T.J., G.A.K., 0.8., 2.8., W.E.B.);
`Intelligent Surgical Lasers, San Diego, California 92121 (T.J.); Department of Optics and
`Quantum Electronics, JA TE University, Dom ter 9, Hungary (2.3.)
`
`Background and Objective: Photodisruption in ocular media
`with high power pulsed lasers working at non-absorbing fre-
`quencies have become a well established surgical tool since the
`late seventies. Shock waves and cavitation bubbles generated by
`the optical breakdown may strongly influence the surgical effect
`of photodisruptive lasers. We have investigated the shock wave
`and cavitation bubble effects of femtosecond laser pulses gener-
`ated during photodisruption in corneal tissue and water. The re-
`sults are compared to those obtained with longer laser pulses.
`Study Design/Materials and Methods: Laser pulses with 150 fs
`duration at ~620 nm wavelength have been focused into corneal
`tissue and water to create optical breakdown. Time-resolved
`flash photography has been used to investigate the dynamics of
`the generated shock waves and cavitation bubbles.
`Results: A rapid decay of the shock waves is observed in both
`materials with similar temporal characteristics and with a spatial
`range considerably smaller than that of shock waves induced by
`picosecond (or nanosecond) optical breakdown. Cavitation bub-
`bles are observed to develop more rapidly and to reach smaller
`maximum diameter than those generated by longer pulses. In
`corneal tissue, single intrastromal cavitation bubbles generated
`by femtosecond pulses disappear within a few tens of seconds,
`notably faster than cavitation bubbles generated by picosecond
`pu ses.
`Conclusions: The reduced shock wave and cavitation bubble ef-
`fects of the femtosecond laser result in more localized tissue dam-
`age. Therefore, a more confined surgical effect should be ex-
`pected from a femtosecond laser than that from picosecond (or
`nanosecond) lasers. This indicates a potential benefit from the
`applications of femtosecond laser technology to intraocular mi-
`crosurgery.
`© 1996 Wiley-Liss, Inc.
`
`Key words: laser-tissue interactions, laser-generated shock waves and cavitation
`bubbles in cornea and water, femtosecond lasers, time-resolved flash
`photography,
`intraocular
`surgery,
`intrastromal photorefractive
`keratectomy
`
`INTRODUCTION
`
`Photodisruption in ocular media generated Accepted for publication March 20, 1995.
`with high power pulsed lasers operating at non-
`Address reprint requests to Tibor Juhasz, Department of
`Physics, University of California, Irvine, CA 92717.
`absorbing frequencies (at least to first order) has
`
`© 1996 Wiley-Liss, Inc.
`
`Alcon Research, Ltd.
`Exhibit 1022 - Page 1
`LENSX0054450
`
`Alcon Research, Ltd.
`Exhibit 1022 - Page 1
`
`

`

`24
`
`Juhasz et a1.
`
`become a well-established tool for intraocular mi-
`
`MATERIALS AND METHODS
`
`crosurgery since the 19708 [1—3]. Damage from
`a photodisruptive laser is initiated by optical
`breakdown at the focus of the laser beam, and the
`plasma created in this way initially expands with
`hypersonic velocity as a result of the plasma pres—
`sure and temperature [4]. A shock wave is emit—
`ted when the plasma expansion decreases to
`subsonic velocity [5]. Further expansion of the
`plasma results in the creation of a cavitation
`bubble [6—11], followed by the implosion of the
`cavitation bubble, which may create secondary
`acoustic transients, such as radially propagating
`acoustic waves [12]. There are indications that
`the primary surgical effect of the photodisruptive
`lasers is due to tissue evaporation by the laser
`plasma [8—11,13]. Shock waves and cavitation
`bubbles may cause extended collateral
`tissue
`damage if they reach large magnitudes. In the
`case of nanosecond pulses, with energies in the
`millijoule range, collateral tissue damage has
`been observed in a large volume due to acoustic
`side effects [8,9]. A strong reduction of the collat—
`eral tissue damage has been achieved with pico—
`second pulses of energies typically in the range of
`a few tens of microjoules [8—11,13].
`Recent developments in laser technology
`have demonstrated that stable generation of fem-
`tosecond laser pulses is possible with solid-state
`lasers. The availability of stable amplified femto—
`second laser pulses suggests, on the one hand,
`their applications to intraocular surgery and, on
`the other hand, generates questions of the possi—
`ble advantage compared to already existing laser
`surgical
`techniques. As previous experiments
`have indicated, the threshold energy for optical
`breakdown is proportional to the square root of
`the duration of the laser pulse in ocular media
`[13,14]. Therefore, optical breakdown can be
`achieved with even smaller energies with femto-
`second laser pulses. This indicates that shock
`wave and cavitation bubble effects and heat
`
`transfer to the tissue may be further reduced if
`femtosecond laser technology is used during in—
`traocular surgery.
`To verify the possible benefit that femto—
`second lasers bring to intraocular surgery and to
`evaluate the application of femtosecond lasers
`in comparison to the techniques currently in use,
`we have investigated the shock wave emission
`and cavitation bubble formation after femtosec-
`
`ond optical breakdown in cornea and in water
`using a time-resolved flash photography tech-
`nique.
`
`Experimental Apparatus
`
`A laser system generating amplified femto-
`second and picosecond laser pulses simultane-
`ously at different wavelengths is used to perform
`the experiments [15]. The femtosecond pulses are
`used to create optical breakdown in the samples,
`whereas the picosecond pulses provide the flash
`for the time-resolved photography. The experi-
`mental arrangement is outlined in Figure 1. The
`system begins with a cw-pumped, actively mode-
`locked, NszAG master oscillator, which delivers
`pulses of ~100 ps duration at 1.064 um wave-
`length at a 76 MHz repetition rate. After fre-
`quency doubling in a KTP crystal, 1 W of 532 nm
`radiation is obtained, which is used to synchro-
`nously to pump a linear cavity femtosecond dye
`laser [15]. The standard six-mirror linear cavity
`femtosecond dye laser is equipped with intracav-
`ity group velocity dispersion control and delivers
`pulses of ~130 fs duration at ~620 nm wave-
`length with an average power of ~50 mW. The
`output of the femtosecond dye laser is synchro-
`nously amplified through a two-stage dye ampli-
`fier chain using quartz cuvettes with flowing dyes
`of Kiton Red at a concentration of ~10‘4 moles
`
`in water. A NszAG regenerative amplifier is
`seeded with a small portion of the IR output of the
`master oscillator. The frequency doubled output
`of the regenerative amplifier is then used to pump
`the dye amplifier chain. In order to suppress the
`amplified spontaneous emission and the unampli-
`fied background, a saturable absorber dye jet is
`placed between the two amplifier stages. After
`amplification, the pulse energy and the pulse du-
`ration are measured to be ~3 uJ and ~150 fs,
`respectively. The repetition rate of the amplified
`pulses can be varied between 25 Hz and 1 kHz.
`The amplified femtosecond pulses are transmitted
`through a variable density optical filter for fine
`adjustment of the energy before they are applied
`to the samples.
`In order to provide the synchronized flash
`pulses used for the time-resolved photography,
`pulses from the output of the master oscillator are
`amplified in a second regenerative amplifier. A
`second harmonic generator unit is used to convert
`the amplified near-infrared laser pulses into vis-
`ible light at 532 nm wavelength. The duration of
`the green pulses is measured to be ~60 ps, which
`determines the time resolution of the system. The
`relative timing between the two regenerative am-
`plifiers is adjusted by steps corresponding to the
`
`Alcon Research, Ltd.
`Exhibit 1022 - Page 2
`LENSX0054451
`
`Alcon Research, Ltd.
`Exhibit 1022 - Page 2
`
`

`

`Femtosecond Laser Pulses in Corneal Tissue
`
`25
`
`
`
`
`
`
`
`
`Fig. 1. The schematic diagram of the experimental appara-
`tus. MO, Nd2YAG master oscillator; SHG's, second harmonic
`generator crystals, FL, linear cavity synchronously pumped
`femtosecond dye laser; RAs, Nd:YAG regenerative amplifiers;
`DD, digital delay generator; FA, synchronously pumped fem-
`tosecond dye amplifiers; OD, variable optical delay; L, lens
`(f = —30 cm); F1 variable density filter; DM, dichroic mirror;
`M, microscope objective (magnification: 16 X); S, sample; F2,
`interference filter; IO, imaging optics; C, CCD camera.
`
`time interval between two oscillator pulses (13
`us) by means of a digital delay generator. A fine
`adjustment of the time delay below 13 ns is pro-
`vided by a variable optical delay.
`Bovine corneas are used in our study. The
`eyes had been enucleated several hours prior to
`the experiments. The corneas were removed a few
`minutes before the laser illumination and placed
`between two glass plates, which prevented the
`loss of water during the experiments. An outline
`of the photographic apparatus is displayed in Fig—
`ure 1. The femtosecond surgical beam is at near
`normal incidence to the corneal surface and is fo-
`
`cused to a diameter of ~5 pm in the cornea. Mea-
`surements of the spot size have been carried out
`with a knife edge. A divergent illumination beam
`is directed to the same beam pass as the surgical
`beam with a dichroic mirror and is focused into
`
`the cornea with the same objective. Due to its
`divergence, the focal plane of the illumination
`beam is behind that of the surgical beam. There-
`fore the green beam illuminates the focal spot of
`the surgical pulse and its surrounding area. An
`interference filter blocks out the surgical beam
`behind the sample and the green beam is imaged
`to a CCD camera (ELMO model MN401E) with
`the help of two microscope objectives. The picture
`provided by the CCD camera is recorded by a
`video recorder. The spatial resolution of the im-
`aging system is estimated to be ~2 am. In order
`
`to ensure recording single events in each video
`frame with our simple nongated camera, electron-
`ically triggered shutters select single pulses from
`the surgical and illumination beams.
`The cornea is replaced with a rectangular
`glass cuvette for comparative studies in water.
`Repetitive measurements were taken since both
`the shock wave and cavitation are completely re-
`laxed in water by the time the next surgical laser
`pulse arrives at the target. Contrary to this, the
`laser pulses introduce irreversible changes in tis—
`sue. Therefore the corneal samples are moved
`across the laser beams by a stepper motor to en-
`sure that each pulse interacts with undamaged
`tissue.
`
`RESULTS
`
`We first determined the threshold fluence of
`
`the optical breakdown inside the cornea and in
`tap water. Since the fluorescent light of the fem-
`tosecond optical breakdown is rather weak at
`intensities around the threshold, we used the ap—
`pearance of the cavitation bubbles as an indica—
`tion of optical breakdown. In order to complete
`the measurements, the fluence of the surgical la—
`ser pulse is slowly increased from below threshold
`level until the appearance of the cavitation bub—
`bles is observed with the high resolution imaging
`apparatus. The disappearance of cavitation bub—
`ble generation is observed with decreasing pulse
`energy. The threshold fluences in cornea and in
`tap water are 1.3 J/cm2 and 0.87 J/cmz, respec-
`tively, as calculated from the measurements of
`the pulse energy and the spot size. The threshold
`fluence in the stroma is only slightly higher than
`that measured on the corneal surface [13]. As
`compared to other threshold measurements with
`longer pulses [5,13,14,16], both measurements
`support earlier observations that
`the optical
`breakdown threshold is proportional to the square
`root of the pulse duration.
`The dynamics of the shock waves are in-
`vestigated with a series of pictures taken with
`increasing time delay between the optical break-
`down and the exposure of the picture. Photo-
`graphs of shock waves obtained in bovine cornea
`and water are shown a and b, respectively, in Fig—
`ure 2. The density of the material, and therefore
`the optical refraction, is changed in the shock
`wave front, which deflects some light out from the
`aperture of the imaging lens system. Thus the
`shock wave appears as a dark ring in the pictures.
`The distance of the shock front from its cen-
`
`Alcon Research, Ltd.
`Exhibit 1022 - Page 3
`LENSX0054452
`
`Alcon Research, Ltd.
`Exhibit 1022 - Page 3
`
`

`

`26
`
`Juhasz et a1.
`
`
`
`Fig. 2. Shock waves photographed 8 ms after the optical breakdown in bovine cornea (a), and
`12.8 ns after the optical breakdown in water (b). The laser fluences are 15 J/cm2 and 9 J/cm2,
`respectively. Bar = 20 um.
`
`ter, defined as the radius of the shock wave, is
`measured in directions perpendicular to the prop—
`agation direction of the light. Figure 3 displays
`the shock wave radius as a function of time. The
`
`displayed data are taken in bovine cornea sam-
`ples (open circles) and in water (filled circles) with
`laser fluences 15 J/cm2 and 9 J/cm2, respectively.
`Each data point is an average of 8—10 measure-
`ments obtained in separate events. The error bars
`are the standard deviations. The sizes of the error
`
`bars are ~2 pm, which is consistent with the spa-
`tial resolution of the system. (The error bars are
`not visible in Fig. 3 since their sizes are slightly
`smaller than those of the symbols.) Nonlinear
`time dependence of the shock wave radius is ob—
`served only in the first ~10 us after the optical
`breakdown indicating the rapid decay of the
`shock wave. The solid line represents a linear fit
`to the data obtained in water with delays longer
`than ~10 ns.
`
`
`
`60
`
`50
`
`4-0
`
`(pm) 10 I—L_L
`SHOCKWAVERADIUS
`
`
`30
`
`20
`
`0
`
`5
`
`10
`
`15
`
`I__i_|___lu
`20
`25
`30
`35
`40
`
`TIME DELAY (ns)
`
`Fig. 3. The shock wave radius as a function of time in bovine
`cornea (open circles), and in water (filled circles). The laser
`fluences are 15 J/cm2 and 9 Jicmz, respectively. The solid line
`is a linear fit to the data obtained in water ~10 us after the
`impact of the laser. The velocity of the sound in water is
`measured to be v5 = 1,488 m/s from the slope of the line.
`
`The velocity of the shock wave as a function
`of the shock wave radius is shown in Figure 4.
`The open circles represent results obtained in cor-
`
`neal samples, and the filled circles display results
`in water. The shock wave velocity is obtained by a
`simple numerical differentiation with respect to
`
`Alcon Research, Ltd.
`Exhibit 1022 - Page 4
`LENSX0054453
`
`Alcon Research, Ltd.
`Exhibit 1022 - Page 4
`
`

`

`Femtosecond Laser Pulses in Corneal Tissue 27
`
`
` I 1 ‘1' I‘ '1
`l
`I
`l'
`'—I'
`I
`
`
`
`
`T
`
`
`
`
`
`SHOCKWAVEVELOCITY(m/s)
`
`M ()1 O O
`
`N O O O
`
`1500
`
`0
`
`J
`
`
`
`_l_
`
`J
`
`
`
`
`
`SHOCKWAVEPRESSUREx10a(Pa)
`
`ii
`
`l
`1o
`
`l
`
`I
`30
`
`L
`20
`
`I
`4o
`
`I
`50
`
`I— L
`O
`20
`
`.1
`4O
`
`60
`
`80
`
`100
`
`_|_
`120
`
`SHOCK WAVE RADIUS (,um)
`
`SHOCK WAVE RADIUS (urn)
`
`Fig. 4. The shock wave velocity as a function of the shock
`wave radius in bovine cornea (open circles) and in water
`(filled circles) as calculated from the data displayed in Figure
`3. The solid line represents the one parameter fit of Eq. 1 to
`the data obtained in water.
`
`the time of the data displayed in Figure 3. The
`differentiation is defined as Ar/At = (ri +1 — ri)/
`(ti + 1 — ti) where ri is the radius of the shock wave
`at time ti. The procedure starts at the first mea-
`sured value of ri. The above derivation of the
`shock wave velocity results in a superposition of
`the experimental uncertainties of ri+1 and ri,
`which is indicated by the increased sizes of the
`error bars in Figure 4.
`As the high velocity shock wave propagates
`through the medium, it loses energy and decays to
`a harmless acoustic wave that propagates with
`the velocity of the sound. The velocity of sound in
`water, as is determined from the slope of the line
`fit in Figure 3, is found to be 1,488 m/s. The ve-
`locity of sound in cornea is found to be 1,463 m/s
`using the same method. The dissipation of the
`shock waves takes place within ~10 ns in both
`media, corresponding to a radius of ~20 pm. The
`above observations indicate that the dynamics of
`the shock waves generated by femtosecond optical
`breakdown is similar in corneal tissue and in wa-
`
`ter. This confirms previous results obtained with
`picosecond pulses [12].
`We focus our further analysis of shock wave
`dynamics on water since the parameters of this
`material are well known. Following the analysis
`of ref. [17], we determine the shock wave pressure
`generated by the femtosecond laser pulses. First,
`we fit the theoretical function
`
`
`()—A+ A2+C
`var—2
`4
`r2
`
`1
`()
`
`Fig. 5. The shock wave pressure as a function of the shock
`wave radius in water generated by laser pulses with fluences
`of!) J/cmz. (lPa = 1N/m2 = 10‘5 bar)
`
`which describes the velocity of the shock wave as
`a function of radius to the measured values of
`shock wave velocities. Here r is the radius of the
`
`shock wave, vs(r) is the shock wave velocity, A =
`1,488 m/s (velocity of sound in water), and C is a
`fitting parameter. The value C is obtained to be
`2.7 X 10"5 m4/s2 at the best fit, which is shown as
`a solid line in Figure 4. The shock wave pressure
`is determined from the expression
`
`no 1
`P(r) = C E 13
`
`(2)
`
`where P(r) represents the shock wave pressure as
`a function of the shock wave radius, C is the fit-
`ting parameter obtained from the velocity curve,
`po = 998 kg/m3 is the density of water [18], and B
`= 2.07 [19]. The above values of A and B are valid
`for pressure values up to 2 X 109 Pa (20 kbar). A
`higher order approximation for calculating P(r) is
`given in ref. [20]. The shock wave pressure as a
`function of the shock wave radius is displayed by
`the solid curve in Figure 5. The pressure at the
`closest observable point to the optical breakdown
`was ~8 X 108 Pa (8 kbar). The shock wave pres-
`sure approaches zero within a radius of ~20 am
`away from the optical breakdown.
`The increase of the delay time to the submi-
`crosecond and microsecond range made it possible
`to observe the temporal evolution of the cavita-
`tion bubbles. Figure 6a,b displays examples of
`cavitation bubbles in cornea and water, respec-
`tively. The diameter of the cavitation bubbles
`generated by the femtosecond optical breakdown
`as a function of time is shown in Figures 6 and 7
`
`Alcon Research, Ltd.
`Exhibit 1022 - Page 5
`LENSX0054454
`
`Alcon Research, Ltd.
`Exhibit 1022 - Page 5
`
`

`

`28
`
`Juhasz et a1.
`
`
`
`Fig. 6. Cavitation bubbles photographed 1 us after the optical breakdown in bovine cornea
`(a) and 6 us after the optical breakdown in water (b). The laser fluences are 16 J/cm2 and
`9 J/cm2, respectively. Bar = 20 pm.
`
`in bovine cornea and water, respectively. Laser
`fluences of 15 J/cm2 in cornea and 9 J/cm2 in wa-
`
`ter are applied. Measurements are taken in steps
`of 10 us for the first 100 ns, and in steps of 100 ns
`afterward. For each delay, time measurements
`from 8—10 independent events are averaged. The
`error bars represent the standard deviations. As
`can be seen in Figures 6 and 7, the cavitation
`bubbles reach their maximum size within 650 ns
`
`in corneal tissue and within 2.7 as in water. The
`maximum bubble radius is ~23 pm in the corneal
`tissue and ~27 pm in water. In cornea a small
`bubble remains in the tissue after one oscillation
`
`that possess a period time of ~5 us. The small
`bubble disappears within 15—30 seconds. The in-
`creased size of error bars in the case of water is
`due to the increased fluctuations of the bubble
`
`size observed during the collapsing phase.
`We now turn to the comparison of the mea-
`sured values of the bubble radius in water to the-
`
`ory. Rayleigh [21] has derived the maximum di-
`ameter of the cavitation bubbles, Rmax, as
`
`
`T'—T
`"r
`1
`r
`
`50
`
`’E‘
`3/40
`m
`530
`E
`D 20
`Lu
`3
`9 10
`ED
`
`1’
`l
`9’
`
`§
`
`:5
`
`i
`
`§} a
`
`§
`
`
`l
`l
`I
`l
`l
`0
`10‘210‘1
`10°
`10‘
`1o2
`105
`
`104
`
`TIME DELAY (#5)
`
`Fig. 7. The diameter of the cavitation bubbles in bovine cor-
`nea as a function of time shown in a semilogarithmic plot.
`The laser fluence is 15 J/cmz.
`
`Tc
`Rmax = —— (3)
`0.915 L
`p0 _ pv
`
`Alcon Research, Ltd.
`Exhibit 1022 - Page 6
`LENSX0054455
`
`Alcon Research, Ltd.
`Exhibit 1022 - Page 6
`
`

`

`Femtosecond Laser Pulses in Corneal Tissue
`
`29
`
`(um) 8
`BUBBLEDIAMETER
`
`
`N O
`
`30
`
`nil—4.4.1.;
`012345
`
`TIME
`
`D
`
`ELAY (us)
`
`Fig. 8. The diameter of the cavitation bubble in water as a
`function of time. The laser fluence is 9 Jlcmz.
`
`where p0 denotes the density of water (po = 998
`kg/ma), po is the pressure in the water at the bub-
`ble wall, pv is the vapor pressure, and Tc denotes
`the collapse time for the bubble. The values of p0
`and pv are determined in ref. [18] to be 101.5 kPa
`and 2.6 kPa, respectively. The above equation as-
`sumes a spherical bubble, in which the time of
`collapse is equal to the time of growth [22]. We
`measured the time of growth to be ~ T0 = 2.7 as.
`Using this value to calculate the maximum of
`cavitation bubble radius with Rayleigh’s equa-
`tion, we obtain Rumx = 29 am. This is within
`reasonable agreement with the measured value of
`Rm z 27 um.
`
`DISCUSSION
`
`It is important to compare our data to previ-
`ously obtained results on the dynamics of shock
`waves and cavitation bubbles generated with pi-
`cosecond laser pulses. In order to compare the re-
`sults on shock waves, we plot the pressure of
`shock waves generated by femtosecond and pico-
`second optical breakdown as a function of the
`shock wave radius in Figure 5 with solid and
`dashed curves, respectively. The picosecond data
`are copied from Figure 5 of ref. [12]. Both the
`femtosecond and the picosecond data were ob-
`tained With laser fluences approximately ten
`times above the threshold. Figure 5 indicates a
`considerably faster decay of the shock waves gen-
`erated by femtosecond optical breakdown. To
`make this comparison more quantitative, we as-
`sume isotropic propagation of the shock wave and
`arbitrarily define the range of shock wave at a
`distance when its pressure decreases below 3 X
`
`107 Pa. (At low pressure values the accuracy of
`our method, to determine the shock wave pres—
`sure, is limited [17].) We estimate the volume of
`tissue effected by the shock wave by calculating
`the volume of a sphere with a radius equal to the
`range of the shock wave. We obtain V = 4 x 10‘5
`mm3 for femtosecond and V = 4 X 10’2 mm3 for
`picosecond optical breakdown. This indicates that
`the volume of the tissue effected by the shock
`wave is ~1,000 times less if femtosecond, rather
`than picosecond, laser pulses are used during sur-
`gery. Obviously, an even more drastic reduction
`of the shock wave is observed if the current re-
`
`sults are compared to traditional nanosecond pho-
`todisruptive lasers, e.g., a Nd:YAG laser.
`Similarly to the range of the shock wave, the
`size of the cavitation bubbles is also considerably
`reduced if picosecond photodisruptive lasers are
`replaced with femtosecond ones. In corneal tissue
`the diameter of the cavitation bubbles, measured
`1 ms after the impact of the laser, is ~14 um (see
`Fig. 7) contrary to the ~80 mm diameter of the
`bubbles generated by picosecond pulses [12]. In
`both cases fluences exceeding the threshold by
`~10 times are used in the experiments. Similar
`reduction in the maximum diameter of the cavi-
`tation bubbles is observed in corneal tissue as
`
`well as in water. A faster development and col-
`lapse of the bubbles generated with femtosecond
`pulses is observed, which is consistent with their
`smaller diameter. Furthermore, full disappear-
`ance of the bubbles in corneal tissue occurs Within
`
`15-30 seconds, contrary to 10—30 minutes ob-
`served in the case of picosecond laser-induced cav-
`itation bubbles [12].
`Side effects, especially cavitation bubbles,
`play an important role in intrastromal refractive
`surgery.
`In a procedure called intrastromal
`photorefractive keratectomy (ISPRK), also fre-
`quently referred to as “intrastromal ablation,” a
`series of ultrashort laser pulses are used to de-
`stroy tissue inside the corneal stroma in order to
`alter the refractive power of the eye. The proce-
`dure shows initial success to correct myopia in
`animal experiments [23]. As is suggested by pre—
`vious results [8—11,13], tissue removal is due to
`the effect of the laser plasma. It follows that the
`most efficient tissue removal can be achieved by
`placing the approximately spherical shaped mi-
`cro—plasmas adjacent to each other. If the cavita-
`tion bubble generated by the first laser pulse is
`larger than the actual plasma size, it may inter-
`act with the surgical effect of the next pulse.
`Moreover if the impact of the pulse occurs inside
`
`Alcon Research, Ltd.
`Exhibit 1022 - Page 7
`LENSX0054456
`
`Alcon Research, Ltd.
`Exhibit 1022 - Page 7
`
`

`

`30
`
`Juhasz et a].
`
`an existing cavitation, it does not remove tissue
`but increases the size of the cavitation through
`heat transfer to the gas. In order to avoid reheat-
`ing the gas inside the cavitation, the next pulse
`must impact the tissue outside the bubble. There-
`fore, in the optimum case, the ratio of the cavita-
`tion diameter at the moment of the arrival of the
`next pulse, Dc, and the plasma diameter, DP,
`should approach unity. (Dc is measured after the
`transient oscillations at the commonly used 1 kHz
`or lower repetition rates.) Using the data ob-
`tained with femtosecond laser pulses and the pi-
`cosecond data previously published in ref. [12] al-
`lows us to determine Dc/Dp for both cases. In the
`case of femtosecond optical breakdown, we obtain
`Dc/Dpz3; for picosecond optical breakdown, Dc/
`Dpz4. Laser fluences approximately ten times
`above the threshold were used in both cases. Al-
`
`though in the high laser fluence range, this result
`is in good agreement with the empirical relation
`that DCOCEl/3 [12], where E is the energy of the
`laser pulse, further investigations are necessary
`for laser fluences close to the threshold where
`
`DcocE [12], for which a larger difference is ex-
`pected between the Dc/Dp ratios. These results
`suggest that the efficiency of ISPRK may be in-
`creased if femtosecond laser pulses are used.
`
`CONCLUSIONS
`
`The dynamics of shock waves and cavitation
`bubbles generated by femtosecond optical break-
`down have been investigated in corneal tissue
`and water with time-resolved flash photography.
`The experimental results have shown that the use
`of femtosecond laser pulses for photodisruption
`results in a strong reduction in the magnitude of
`acoustic side effects such as shock waves and cav-
`
`itation bubbles. Therefore the surgical effect of
`femtosecond laser pulses is more localized than
`that of picosecond (or nanosecond) pulses. This in—
`dicates the potential of femtosecond laser technol-
`ogy for applications in high precision intraocular
`microsurgery.
`
`ACKNOWLEDGMENTS
`
`The authors thank Intelligent Surgical La-
`sers for providing the CCD camera and a Nd:YAG
`regenerative amplifier. This research was sup-
`ported through US—Hungarian Science
`and
`Technology Fund 06/90, and funds from NSF
`DMR931175 and ARO DAAHO 93-G-0188. G. A.
`
`K. received support from the UCI Honor Student
`Program.
`
`REFERENCES
`
`1. Steinert RF, Puliafito CA. The Nd:YAG laser in ophthal-
`mology. Philadelphia: W.B. Saunders, 1985.
`2. Krasnov MM. Laser puncture of anterior chamber angle
`in glaucoma. Am J Ophthalmol 1973; 75:674—678.
`3. Frankhauser F, Roussel P, Steffen J, van der Zypen E,
`Chernakova A. Clinical studies on the efficiency of high
`power laser radiation upon some structures of the ante—
`rior segment of the eye. Int Ophthalmol 1980; 3:129—139.
`4. Barness PA, Rieckhoff KE. Laser induced underwater
`sparks. Appl Phys Lett 1968; 13:282—284.
`5. B. Zysset B, Fujimoto JG, Deutsch TF. Time resolved
`measurements of picosecond optical breakdown. Appl
`Phys B 1989; 48:139—147.
`6. Fujimoto JG, Lin WZ, Ippen EP, Puliafito CA, Steinert
`RF. Time-resolved studies of Nd:YAG laser-induced
`breakdown. Invest Ophthalmol Vis Sci 1985; 26:1771—
`1777.
`
`7. Vogel A, Hentscbel W, Holzfuss J, Lauterborn W. Cavi-
`tation bubble dynamics and acoustic transient genera-
`tion in ocular surgery with pulsed neodymiumzYAG 1a-
`ser. Ophthalmol. 1986; 93:1259—1269.
`8. Zysset B, Fujimoto JG, Puliafito CA, Birngruber R,
`Deutsch TF. Picosecond optical breakdown: Tissue effects
`and reduction of collateral damage. Lasers Surg Med
`1989; 9:193—204.
`9. Vogel A, Schweiger P, Freiser A, Asyo MN, Birngruber
`R. Intraocular Nd:YAG laser surgery: Light-tissue inter-
`actions, damage range, and reduction of collateral effects.
`IEEE J Quant Electron 1990; 26:2240—2260.
`10. Vogel A, Busch S, Jungnickel K, Birngruber R. Mecha-
`nism of intraocular photodisruption with picosecond and
`nanosecond laser pulses. Lasers Surg Med 1994; 15:32—
`43.
`
`11. Vogel A, Capon MR, Asyo-Vogel MN, Birngruber R. In-
`traocular photodisruption with picosecond and nanosec-
`ond laser pulses: Tissue effects in cornea, lens, and ret-
`ina. Invest Ophthalmol Vis Sci 1994; 35:3022—3044.
`12. Juhasz T, Hu, XH, Turi, L, Bor Z. Dynamics of shock
`waves and cavitations generated by picosecond laser
`pulses in corneal tissue and water. Lasers Surg Med
`1994; 15:91—98.
`13. Niemz M.H, Hoppeler T, Juhasz T, Bille JF. Intrastromal
`ablations for refractive corneal surgery using picosecond
`infrared laser pulses. Lasers Light Ophthalmol 1993;
`5:145—152.
`
`14. Stern D, Schoenline RW, Puliafito CA, Dobi ET, Birngru-
`ber R, Fujimoto JG. Corneal ablation by nanosecond, pi-
`cosecond, and femtosecond lasers. Arch Ophthalmol
`1989; 107:587—592.
`15. Juhasz T, Smith GO, Mehta SM, Harris K, Bron WE.
`Generation and kilohertz rate amplification of synchro-
`nized femtosecond and picosecond laser pulses. IEEE J
`Quant Electron 1989; 25:1704—1707.
`16. Du D, Squier J, Kurtz R, Elner V, Liu X, Gutmann G,
`Mourou G. Damage threshold as a function of pulse du-
`ration in biological tissue. In Barbara PF et al., eds. “U1—
`trafast Phenomena IX.” New York: Springer, 1995, pp
`254—255.
`
`Alcon Research, Ltd.
`Exhibit 1022 - Page 8
`LENSX0054457
`
`Alcon Research, Ltd.
`Exhibit 1022 - Page 8
`
`

`

`Femtosecond Laser Pulses in Corneal Tissue
`21.
`
`17.
`
`18.
`
`19.
`
`20.
`
`Doukas AG, Zweig AD, Fn’soli JK, Birng'ruber R,
`Deutsch TF. Non-invasive determination of shock wave
`pressure generated by optical breakdown. Appl Phys B
`1991; 237—245.
`Ward B, Emmony DC. Interferometric studies of the pres-
`sures developed in a liquid during infrared-laser-induced
`cavitation-bubble oscillation. Infrared Phys 1991; 32:
`489—515.
`Harris P, Presles NH. Reflectivity of a 5.8 kbar shock
`front in water. J Chem Phys 1981; 74:6864—6866.
`Rice MH, Walsh JM. Equation of state of water to 250
`kilobars. J. Chem Phys 1957; 26:824—830.
`
`22.
`
`23.
`
`31
`
`Rayleigh GE. On the pressure developed in a liquid dur-
`ing the collapse of a spherical cavity. Phil Mag 1917;
`34:94—98.
`Lauterborn W. Laser-induced cavitation. Acustica 1974;
`31:51—78.
`
`Habib MS, Speaker MG, Kaiser R, Juhasz T. Myopic in-
`trastromal photorefractive keratectomy with the NszLF
`picosecond laser in the cat cornea. Arch Ophthalmol.
`1995; 11314997503.
`
`Alcon Research, Ltd.
`Exhibit 1022 - Page 9
`LENSX0054458
`
`Alcon Research, Ltd.
`Exhibit 1022 - Page 9
`
`