Holmium:YAG Laser Lithotripsy: ADominant Photothermal Ablative

Mechanism With ChemicalDecomposition of Urinary Calculi

Kin Foong Chan, MSE,1* George J. Vassar, MD,2 T. Joshua Pfefer, PhD,3

Joel M. H. Teichman, MD,2 Randolph D. Glickman, PhD,4

Susan T. Weintraub, PhD,5 and Ashley J. Welch, PhD1,2

1Department of Electrical and Computer Engineering, University of Texas at Austin,Austin, Texas 78712

2Division of Urology, University of Texas Health Science Center,San Antonio, Texas 78284

3Biomedical Engineering Program, University of Texas at Austin, Austin, Texas 787124Department of Ophthalmology, University of Texas Health Science Center,

San Antonio, Texas 782845Department of Biochemistry, University of Texas Health Science Center,

San Antonio, Texas 78284

Background and Objective: Evidence is presented that the frag-mentation process of long-pulse Holmium:YAG (Ho:YAG) litho-tripsy is governed by photothermal decomposition of the calculirather than photomechanical or photoacoustical mechanismsas is widely thought. The clinical Ho:YAG laser lithotriptor (2.12mm, 250 ms) operates in the free-running mode, producing pulsedurations much longer than the time required for a sound waveto propagate beyond the optical penetration depth of this wave-length in water. Hence, it is unlikely that shock waves are pro-duced during bubble formation. In addition, the vapor bubbleinduced by this laser is not spherical. Thus the magnitude of thepressure wave produced at cavitation collapse does not contrib-ute significantly to lithotripsy.Study Design/Materials and Methods: A fast-flash photographysetup was used to capture the dynamics of urinary calculusfragmentation at various delay times following the onset of theHo:YAG laser pulse. These images were concurrently correlatedwith pressure measurements obtained with a piezoelectricpolyvinylidene-fluoride needle-hydrophone. Stone mass-loss

Contract grant sponsor: Office of Naval Research Free Elec-tron Laser Biomedical Science Program; contract grant num-ber: N00014-91-J-1564 (UT Austin); contract grant sponsor:The Albert W. and Clemmie A. Caster Foundation (UT Aus-tin); contract grant sponsor: The Research to Prevent Blind-ness (UTHSCSA); contract grant sponsor: The Air Force Of-fice of Scientific Research; contract grant number: F49620-98-1-0210 (UTHSCSA); contract grant sponsor: The AirForce Office of Scientific Research through MURI fromDDR&E; contract grant number: F49620-98-1-0480 (UT Aus-tin).

*Correspondence to: Kin Foong Chan, Department of Electri-cal and Computer Engineering, University of Texas at Austin,ENS 610, Austin, TX 78712.E-mail: kfchan@ccwf.cc.utexas.edu

Accepted 2 February 1999

Lasers in Surgery and Medicine 25:22–37 (1999)

© 1999 Wiley-Liss, Inc.

measurements for ablation of urinary calculi (1) in air (dehy-drated and hydrated) and in water, and (2) at pre-cooled and atroom temperatures were compared. Chemical and compositionanalyses were performed on the ablation products of severaltypes of Ho:YAG laser irradiated urinary calculi, including cal-cium oxalate monohydrate (COM), calcium hydrogen phosphatedihydrate (CHPD), magnesium ammonium phosphate hexahy-drate (MAPH), cystine, and uric acid calculi.Results: When the optical fiber was placed perpendicularly incontact with the surface of the target, fast-flash photographyprovided visual evidence that ablation occurred approximately50 ms after the initiation of the Ho:YAG laser pulse (250–350 msduration; 375–400 mJ per pulse), long before the collapse of thecavitation bubble. The measured peak acoustical pressure uponcavitation collapse was negligible (< 2 bars), indicating thatphotomechanical forces were not responsible for the observedfragmentation process. When the fiber was placed in parallel tothe calculus surface, the pressure peaks occurring at the col-lapse of the cavitation were on the order of 20 bars, but no frag-mentation occurred. Regardless of fiber orientation, no shockwaves were recorded at the beginning of bubble formation. Ab-lation of COM calculi (a total of 150 J; 0.5 J per pulse at an 8-Hzrepetition rate) revealed different Ho:YAG efficiencies for de-hydrated calculus, hydrated calculus, and submerged calculus.COM and cystine calculi, pre-cooled at −80°C and then placed inwater, yielded lower mass-loss during ablation (20 J, 1.0 J perpulse) compared to the mass-loss of calculi at room tempera-ture. Chemical analyses of the ablated calculi revealed productsresulting from thermal decomposition. Calcium carbonate wasfound in samples composed of COM calculi; calcium pyrophos-phate was found in CHPD samples; free sulfur and cysteinewere discovered in samples composed of cystine samples; andcyanide was found in samples of uric acid calculi.Conclusion: These experimental results provide convincing evi-dence that long-pulse Ho:YAG laser lithotripsy causes chemicaldecomposition of urinary calculi as a consequence of a domi-nant photothermal mechanism. Lasers Surg. Med. 25:22–37,1999. © 1999 Wiley-Liss, Inc.

Key words: fast-flash photography; kidney stones; laser lithotripsy; long-pulseHo:YAG; mass-loss; photoacoustical pressure waves; photothermalmechanism; thermal breakdown

INTRODUCTION

Long-pulse Holmium:YAG (Ho:YAG) laser(2.12 mm) has been used extensively in urology forlaser lithotripsy [1–12]. The Ho:YAG laser is anattractive alternative to other conventional laserlithotriptors such as the Q-switched Nd:YAG,pulsed-dye and ruby lasers because it offers highpower (∼ kW, up to several Joules at 250-ms du-ration) pulses that are transmittable via a lowOH− optical fiber. As a solid state laser, it is lessexpensive and easier to maintain than pulsed-dyesystems. Whereas it has been reported that theQ-switched Nd:YAG laser is not efficient in frag-menting calcium oxalate monohydrate (COM) cal-culi [13] and that the pulsed-dye laser is not ca-pable of fragmenting cystine calculi [3,4], the

long-pulse Ho:YAG laser has been effective for alltypes of urinary calculi [1,14]. Ho:YAG lithotripsyhas been so widely accepted that attempts havebeen made to design a smart laser lithotriptor fortissue selectivity and feedback control of energy[15].

However, the dominant mechanism by whichcalculi are fragmented by the Ho:YAG laser hasnot been well defined. The fragmentation processof calculi produced by nanosecond [16–18] Q-switched Nd:YAG laser (1.06 mm) and microsec-ond [17–19] pulsed-dye lasers (596 nm) is photo-mechanical and is described as laser-inducedshockwave lithotripsy (LISL). The long-pulseHo:YAG laser is not known to produce shockwaves due to plasma expansion at the onset oflaser irradiation [20–21]. In addition, the pres-

Holmium:YAG Laser Lithotripsy 23

sure waves generated at cavitation collapse arespeculated to have little contribution to fragmen-tation because of the weak pressure waves due tomultiple collapses [20]. This is because theHo:YAG laser generates elongated cavitation as aresult of continuous deposition of laser energy atthe base of the cavitation bubble due to the longpulse duration [22,23].

A fairly recent attempt by Dushinski et al.[7] to define the mechanism(s) of fragmentationby long-pulse Ho:YAG laser suggested a possiblephotothermal effect on urinary calculi. The au-thors reasoned that since experimental resultsshowed no correlation between cavitation size (orfiber diameter) and lithotripsy efficiency, photo-mechanical effects (shock wave generated at cavi-tation collapse) could not be the dominant mecha-nism and therefore photothermal effects were as-sumed. However, these results did not delineatethe photomechanical effect in fragmentation. Inaddition, the authors argued that the “glowinghot” stone fragments under water provided evi-dence of a photothermal effect. Schafer et al. [21]made the similar conclusion with their “dim whit-ish glow” and their measured emission spectrumthat resulted in an unexpectedly high tempera-ture (5,000 K). Although we support photother-mal ablation as the probable damage mechanism,such a high blackbody temperature is unlikely.

In addition, Schafer et al. did not performchemical analyses on the ablated calculi and itspost-ablation water contents. In eliminating pho-tomechanical effects as a damage mechanism, theauthors also failed to provide sufficient evidenceof acoustic transient, during which the deliveryfiber was placed in parallel or perpendicularly incontact with the calculus surface. Instead, theyreplaced water with a low absorption liquid (ace-tonitrile; ma 4 0.88 cm−1 at 2.12 mm) to demon-strate the role of cavitation. Although we agreethat vapor bubbles help disperse fragmentation,acetonitrile may not have provided the necessaryenvironment for chemical breakdown of stonecomposition through reduction and oxidation asobserved in clinical procedures. Instead, the au-thors only observed melting of the biliary stone byapplication of Ho:YAG under acetonitrile. Be-cause the laser application was made in a quasi-anerobic environment, few if any further chemicaldegradation reactions occurred. Therefore, it isnot surprising that they concluded a photother-mal mechanism without any knowledge of pos-sible chemical reactions.

On the other hand, both Schafer et al. and

Dushinski et al. reported that fiber tips could be“welded” to the calculi upon holmium irradiationin the contact mode [7,21]. Dushinski et al. fur-ther noted that sulfur gas odor was detected dur-ing fragmentation of cystine calculi [7]. Zhong etal. [24] also reported that the fragmentationmechanisms differ for pulsed-dye laser and long-pulse Ho:YAG laser. Whereas LISL fragments areinduced by shock waves, Zhong et al. observedthat the holmium laser drills into the stone sur-face, a result of the well-known “Moses effect”[22].

These valuable observations suggested adominant photothermal effect causing chemicaldecomposition of the calculus that warranted fur-ther verification. Previous experiments by Teich-man et al. revealed that Ho:YAG lithotripsy cre-ated smaller fragments compared to pulsed-dyelasers [8], suggesting a different mechanism fromLISL. Schafer et al. also observed that Ho:YAGirradiated calculus fragments appeared filamen-tary, indicating melting and re-crystallization,whereas those of LISL and mechanically crashedfragments were granular [21]. Also, radiant expo-sure and urinary-calculus compositions alteredthe efficiency of lithotripsy [9–10], leading us tobelieve that direct absorption of laser energyplayed a major role in calculus fragmentation asSchafer et al. suggested. The “Moses effect” seemsto be a convenient model, which allows channel-ing of the holmium laser, and direct irradiation ofand absorption by urinary calculi. Most recently,the discovery of cyanide as a result of laser-induced photothermal breakdown of uric acid cal-culus [11] confirmed the role of non shock waveablative mechanism in long-pulse Ho:YAG litho-tripsy.

In order to demonstrate that photothermaleffects are the dominant mechanism behindHo:YAG lithotripsy and erase the conventionalwisdom that most if not all laser lithotripsies in-volve photomechanical stresses, we investigatedthe interaction of long-pulse Ho:YAG laser with avariety of urinary calculi. Several experimentswere performed to demonstrate the dominantlyphotothermal-induced fragmentation process.First, fast-flash photography was used to captureimages at different delay times with respect to theonset of the long-pulse Ho:YAG laser. This pro-vided visualization of the lithotripsy event. Acous-tical pressure measurements were recorded witha needle-hydrophone for analysis and correlationwith images captured at different delay times.Mass-loss experiments were conducted with vari-

24 Chan et al.

ous urinary calculi in air (dehydrated and hy-drated), and in water. The effects of initial calcu-lus temperature on fragmentation were also in-vestigated with fast-flash photography and mass-loss experiments. Finally, chemical analyses ofcalculus compositions pre- and post-irradiationwere performed.

MATERIALS AND METHODS

Human urinary calculi were obtained from astone analysis laboratory (Louis C. Herring Co.,Orlando, FL). The calculi composed of either> 98% calcium oxalate monohydrate (COM),> 98% calcium hydrogen phosphate dihydrate(CHPD), > 98% magnesium ammonium phos-phate hexahydrate (MAPH), > 98% cystine, or> 98% uric acid. The experiments were performedeither with a Laser 1-2-3 from Schwartz Electro-Optics (l 4 2.12 mm, 250– 350 ms) at the Univer-sity of Texas at Austin, or with a VersaPulse Se-lect from Coherent Medical Group (l 4 2.12 mm,250 ms) at the University of Texas Health ScienceCenter at San Antonio. Except for the experi-ments comparing calculus mass-loss in air (dehy-drated and hydrated calculi) and in water, wherea 365-mm diameter low OH− fiber (Simline-365,Coherent Medical Group, Palo Alto, CA) wasused, all other experiments were conducted with a550-mm diameter low OH− fiber (Simline-550, Co-herent Medical Group).

Six experiments were conducted to elucidatethe mechanisms of long-pulse Ho:YAG lithotripsyof urinary calculi:

(1) Fast-flash photography of laser-calculus

interaction (COM and cystine; UT Aus-tin);

(2) Pressure transient measurements with aPVDF needle-hydrophone (COM andcystine; UT Austin);

(3) The effect of fiber orientation on litho-tripsy (COM and cystine; UT Austin);

(4) Mass-loss measurements of calculi in air(dehydrated and hydrated calculi) and inwater (COM; UTHSC San Antonio);

(5) The effect of initial calculus temperatureon fragmentation or mass-loss (UTHSCSan Antonio);

(6) Chemical analysis of calculus composi-tion pre- and post-irradiation (COM,CHPD, cystine, MAPH, and uric acid cal-culi; UTHSC San Antonio).

Fast-Flash Photography ofLaser-Calculus Interaction

Fast-flash photography was performed tomonitor the dynamics of laser lithotripsy on twotypes of urinary calculi: COM and cystine. In thisexperiment, the calculi were cut with a dental dia-mond band saw to provide a flat target surface. Acalculus was then placed in a water-filled glasscuvette at room temperature (∼ 23°C). Duringlithotripsy, the delivery fiber was placed at threeorientations; either perpendicular, at a 45° angle,or parallel to the calculus surface.

The setup for fast-flash photography isshown in Figure 1. A 100-MHz computer (Gate-way 2000, North Sioux City, SD) with a LabVIEWcontrolled data acquisition software package wasused to trigger a pulse generator (DG535, Stan-ford Research Systems, Sunnyvale, CA), which in

Fig. 1. Experimental setup forfast-flash photography andacoustic pressure measure-ment. A nitrogen-dye laser (l4 540 nm) was used as aflashlamp initially for obtain-ing high quality (500-ps expo-sure time) time-resolved im-ages. When no shock waveswere observed, a broadbandXenon flashlamp (5-ms expo-sure time) was used in place ofthe nitrogen-dye laser, and aPVDF-needle-hydrophone wasadded during the pressuremeasurement experimentalong with concurrent acquisi-tion of fast-flash images.

Holmium:YAG Laser Lithotripsy 25

turn controlled the delay time between the onsetof the Ho:YAG laser pulse and the nitrogen-dyelaser pulse (LN-1000 nitrogen gas laser and LN-102 dye module, Laser Photonics, Orlando, FL; tp4 500 ps; l 4 540 nm).

A long pass filter (03-FCG-113, Melles Griot,Irvine, CA) was used to prevent contamination ofour images by blocking visible and near infraredwavelengths (< 1,000 nm) emitted from theHo:YAG laser flashlamp, which consistently pro-duced bright “glow” due to scattering. The hol-mium laser beam was coupled into a 550-mm lowOH− optical fiber to be delivered to the calculus ina water bath. Two beam splitters were placed atthe outputs of the long-pulse Ho:YAG laser andthe nitrogen-dye laser to couple a small fraction(∼ 4%) of the laser energies to two photodiodes.Signals generated from the photodiodes, repre-senting the temporal beam profiles of the lasers,were monitored on a digital oscilloscope (TDS-640A 500 MHz-200 GS/s, Tektronix, Beaverton,OR). Knowing the exact relative temporal dis-placement between the two lasers allowed us todetermine the delay time of our fast-flash imagesfrom the onset of the Ho:YAG laser pulse.

The nitrogen-dye laser (500-ps exposuretime) was used as the light source for fast-flashphotography (A Xenon flashlamp was used inplace of the nitrogen-dye laser for a later experi-ment). This light source was split into two beamsfor simultaneous reflectance and transmissionimaging. A series of delay times was used to rec-ord lithotripsy events. A line filter (03-FIV-113,Melles Griot, Irvine, CA) at (540 ± 10) nm wasplaced in front of the CCD camera (GP-MF 602,Panasonic, Japan), allowing only the nitrogen-dyewavelength to be transmitted into the CCD array.This technique further enhanced our imaging ca-pability by avoiding contamination from theHo:YAG laser flashlamp. The images captured bythe CCD camera were recorded on a VCR (HR-S5200U, JVC, Japan) and displayed on a videomonitor. A time counter (TG-50, Horita, MissionViejo, CA) was used to label the images as theywere recorded. From the videocassette, individualframes of images were analyzed in freeze-framestyle on a video monitor or with a frame-grabber(Snappy, Play Incorporated, Rancho Cordova, CA)on a computer monitor.

Before each experiment, both ends of the op-tical fiber were cleaved and polished. The outputenergy from the delivery fiber was measured by apower meter (EPM-2000 powermeter, and J25 se-ries pyroelectric joulemeter, Molectron Detector,

Inc., Portland, OR). The energy employed in thisexperiment was maintained at 375 ± 5 mJ/pulsefor a pulse duration of 250 ms. During the experi-ment, nine warm-up Ho:YAG laser pulses (1.5–2.0 Hz) were delivered to a beam dump with amirror shutter to allow the laser to achieve astable energy level. The tenth and final pulse wasdelivered to the target calculus in the water bathby removing the shutter from the beam path. Thenitrogen-dye laser pulse was then triggered at apre-set delay time from the onset of the finalHo:YAG laser pulse. A sequence of laser pulsesand delay times provided images of the completedynamic lithotripsy event, extending to hundredsof microseconds after the end of the Ho:YAG laserpulse. At least five images were captured for eachdelay time for the COM calculus and the cystinecalculus at each different fiber orientation; eitherperpendicular, at a 45° angle, or parallel to thecalculus surface.

Pressure Wave Measurements

The laser-induced pressure transients weremeasured using a PVDF needle-hydrophone,which consisted of a piezoelectric polyvinylidene-fluoride foil as its active element (40-ns rise time,1.502 mV/bar, where 1 bar ≈ 105 Pa ≈ 1 atm, Imo-tec Messtechnik, Germany). Pressure measure-ments were conducted for three different configu-rations: (1) ablation in clear water, (2) ablationwith delivery fiber perpendicular to calculus sur-face in contact mode, and (3) ablation with deliv-ery fiber parallel to calculus surface in contactmode. The hydrophone was placed a few millime-ters away from the calculus surface to avoid pos-sible damage due to ejected fragments. Ho:YAGirradiation was delivered via a 550-mm low OH−

fiber. An energy of 400 mJ/pulse (2.5 Hz) at 350-ms pulse duration was used. The signal from thehydrophone was displayed and recorded on a digi-tal oscilloscope (Tektronix TDS 640A). Later, thesignals were analyzed and back-calculated to apressure transient 1-mm away from the origin ofbubble collapse. At the same time, images werecaptured using fast-flash photography at differentdelay times with respect to the onset of theHo:YAG pulse. These images provided concurrentinformation on the dynamics of cavitation bubblesor lithotripsy. The setup for this experiment wasthe same as the fast-flash photography experi-ment (Fig. 1), but with an additional needle-hydrophone in the water bath and a Xenonflashlamp (5-ms exposure time) in place of the ni-trogen-dye laser. The nitrogen-dye laser was used

26 Chan et al.

in the previous experiment to image possibleshock waves induced by the holmium laser withits short exposure time (500 ps), but was replacedwith the less tedious Xenon flashlamp when noplasma-induced shock wave was discovered byphotography and pressure measurements. Atleast five pressure measurements were recordedfor each configuration mentioned above.

Comparison of Fiber Orientation

COM calculi were sandpapered to obtain aflat surface, cleaned, and dried for a few hours. Animage was then captured with a CCD camera as acontrol. After hydrating the stones for a fewhours, the experiment was then conducted in twoconfigurations: (1) a fiber placed parallel to thecalculus surface with the side of the fiber in con-tact with the calculus, and (2) a fiber placed per-pendicular to the calculus surface with the fibertip in contact with the calculus. After ablation forapproximately 330 seconds (400 mJ/pulse at 2.5Hz), the calculi were removed from the water bathand left to dry. Images of the resulting pits orcraters were captured again with the CCD cam-era and their dimensions were estimated with acaliper.

Mass-Loss Measurements

Prior to laser irradiation, all COM calculi in-volved in mass-loss measurements were dehy-drated by placing the samples in a −80°C freezerfor 24 hours. The samples were then desiccatedfor another 24 hours in a vacuum lyophilizer toreduce humidity. Initial calculus mass was mea-sured after desiccation. The calculi were ran-domly divided into three groups for contact litho-tripsy (1) in air after submerging in water for 24hours—“hydrated calculus in air,” (2) in water af-ter submerging in water for 24 hours, and (3) inair after storage within the lyophilizer for 24hours—“dehydrated calculus in air.”

Each calculus was irradiated with 150 J hol-mium energy (0.5 J/pulse at 8 Hz) via a 365-mmlow OH− optical fiber. The fiber tip was cleavedprior to lithotripsy of each calculus. The calculiwere ablated with fiber contact using a scanningmode along the surface. This helped preserve the“parent” calculus by preventing it from being frag-mented into several large “sister” calculi, whichmight cause difficulty during mass-loss measure-ments. A total of 13 measurements for submergedcalculi, 14 measurements for “hydrated calculi inair,” and 18 measurements for “dehydrated cal-culi in air” were performed.

After lithotripsy, the calculi ablated in waterwere centrifuged at 2,000 rpm for 20 minutes. Thesupernatant was then discarded, and the pelletwas placed in a −80°C freezer for two hours. Cal-culi from all groups were desiccated once againwith a vacuum lyophilizer before the final calcu-lus mass was measured. Calculus mass-loss wasdefined as the difference between the initial cal-culus mass and the final calculus mass of the re-maining “parent” calculus.

Effects of Initial Calculus Temperatureon Fragmentation

The effects of initial calculus temperaturewere determined in terms of mass-loss. This studywas performed using COM and cystine calculi byqualitatively comparing the mass-loss of pre-cooled calculi taken from a −80°C freezer and withthat of uncooled calculi at room temperature. Theinitial mass was measured on both pre-cooled anduncooled calculi. Each calculus was then irradi-ated in contact with a 550-mm low OH− opticalfiber using single 1.0-J holmium pulses at 20separate locations with the scanning mode. Thefinal mass for each calculus was measured. Themeasurements were based upon five samples ofeach calculus types at either the pre-cooled or un-cooled temperature. Again, mass-loss was definedas the difference between the initial mass prior tolithotripsy and the final mass post-lithotripsy.

Pre- and Post-Irradiation Chemical Analysis

A set of Ho:YAG pulses with a total energy of200 J (1 J/pulse at 10 Hz) was delivered using a550-mm low OH− optical fiber to various calculustypes including COM, CHPD, MAPH, cystine, anduric acid. At least three samples of each calculustypes were used to perform this series of experi-ments. After irradiation, composition analyseswere performed separately for the parent calculi,the calculus craters, fragments $ 710 mm, andfragments < 710 mm. Composition analyses of allfragments were performed using infrared spec-troscopy and x-ray crystallography at the samecalculus laboratory where initial calculus compo-sition analyses were performed. The laboratorywas asked to look for thermal products for each ofthe calculi (Table 1).

For cystine, the parent calculus, the frag-ments, and the water in which lithotripsy wasperformed were analyzed. Electron impact ioniza-tion mass spectra were acquired on a FinniganMAT 4615 quadrupole mass spectrometer (Finni-gan, Palo Alto, CA) at an electron energy of 70 eV

Holmium:YAG Laser Lithotripsy 27

and an ion source temperature of 160°C. Sampleintroduction was by means of a direct insertionprobe that was heated ballistically until evapora-tion was complete. Electrospray ionization massspectra were obtained on a Finnigan MAT LCQion trap mass spectrometer. Samples were dis-solved in 2% triofluoroacetic acid and diluted with50% aqueous acetonitrile containing 0.5% aceticacid to reach a final concentration of 0.5 pmole/ml.Sample introduction was by direct infusion intothe mass spectrometer at a flow rate of 10 ml/min.The charge-state of each ion of interest was deter-mined through the Zoom Scan feature of the LCQ.MS/MS spectra were obtained for each analyteand compared to reference standards in order toconfirm identification.

For uric acid, a standard micro-diffusion as-say was used, in which cyanide is trapped in al-kali, and reacted with pyridine and barbituricacid to form a color complex that is measured byspectrophotometry [25]. Because of the clinicalsignificance of cyanide, three separate uric acidcalculi were irradiated and sent to three separatetoxicology laboratories for analysis.

RESULTS

Fast-Flash Photography ofLaser-Calculus Interaction

In this experiment, fast flash images on thedynamics of cavitations and laser lithotripsy werecaptured with the holmium laser set at 375 mJ/pulse for a pulse duration of 250 ms. With thedelivery fiber suspended in clear water and theurinary calculus (COM or cystine) located at adistance of ∼ 5 mm away from the fiber tip, anelongated vapor bubble was produced that did notreach the calculus. No fragmentation was visibleat any time. When the delivery fiber was ad-

vanced to within 1 mm of the calculus surface,lithotripsy occurred if and only if the expandingbubble reached the surface of the calculus. Be-cause of the long pulse duration, laser energypropagated through the vapor channel of thebubble and was directly absorbed on the calculussurface. This off-contact lithotripsy produced fewfragments with a large hemispherical bubble thatexpanded to its maximum size around 250 ms andcollapsed around 450 ms after the onset of the hol-mium laser pulse (Figure 2A–D). No fragmenta-tion was observed after the bubble collapsed.

With the delivery fiber placed in contact andperpendicular to the calculus surface, fast-flashphotography of COM and cystine calculi showedlithotripsy began after about 50 ms, as illustratedin Figure 3, when a plume formation consisting ofdust or fragments and bubbles appeared. Theplume continued to enlarge, and reached its maxi-mum expansion at approximately 200 ms and 400ms for the COM and cystine calculi, respectively,as shown in Figure 4A and 4B. No lithotripsy at-tributable to pressure wave expansion in theearly stage of bubble formation or in the laterstage of bubble collapse was observed. When thefiber was placed at a 45° angle in contact mode,the Ho:YAG laser produced a distorted bubbleand a smaller plume with less fragmentation (Fig-ure 5A–C). Again, no fragmentation was seen ateither the onset of the holmium laser or afterbubble collapse.

When the delivery fiber was placed in con-tact but parallel to the calculus surface, no frag-mentation was observed during bubble expansionor collapse. Figure 6 illustrates that an elongatedhalf-bubble was formed along the surface of aCOM calculus. It reached maximum expansion at350 ms (Fig. 6A) and collapsed between 450 and550 ms (Fig. 6B) after the initiation of the Ho:YAG

TABLE 1. Dehydration, Decomposition, or Conversion Temperatures for VariousUrinary Calculi, and Their Products [40].

CompositionTemperature

(°C) Decomposition products

Calcium oxylate monohydrate(COM) 206 Calcium carbonate

Calcium hydrogen phosphatedihydrate (CHPD) 109 Calcium phyrophosphate

Cystine 264 Free sulfur, cysteinea

Magnesium ammonium phosphatehexahydrate (MAPH) 100 Dehydrates

Uric acidTemperature

unspecified CyanideaCysteine has a boiling point lower than cystine melting point.

28 Chan et al.

laser pulse. Beyond 600 ms (Fig. 6C), no plume orfragmentation ejecting from the calculus surfacewas apparent.

Pressure Wave Measurements

A typical temporal profile of the Ho:YAG la-ser pulse (400 mJ/pulse, 350 ms) and the corre-sponding pressure transients for COM and cys-tine calculi are shown in Figures 7 and 8, respec-tively. The pressure transients were corrected to 1mm away from the center of the bubble or cavita-tion collapse. The figures also illustrate the rela-tive collapse time of cavitations created in waterthrough the fiber in three configurations: (1) inclear water, (2) with the fiber placed perpendicu-

lar to the calculus surface, and (3) with the fiberplaced parallel to the calculus surface.

According to our pressure transient data, atypical vapor bubble in clear water collapsedwithin 400–450 ms after the onset of a 400-mJlaser pulse energy (Figs. 7A and 8A). However, abubble induced with the fiber placed parallel tothe stone surface collapsed later between 450 msand 550 ms (Figs. 7C and 8C), indicating that thebubble was distorted or larger in at least one di-mension. Our fast-flash images confirmed this ob-servation. The bubble induced when placing thefiber parallel to the calculus surface was moreelongated than that in clear water. This was be-cause the calculus surface obstructed the bubble’s

Fig. 2. These images illustrate noncontact holmium laser lithotripsy: A: At 50 to 75 ms, the expanding vapor bubble (glassyappearance) reaches the calculus surface. Lithotripsy occurs upon penetration of holmium energy through the vapor channelto the exposed calculus surface. At 100 ms, B: The vapor bubble becomes more “dusty” due to thermal breakdown of calculuscompositions and fragmentation. At 250 ms, C: The maximum expansion of the vapor bubble occurs with more fragmenteddusts. After bubble collapse, shown here at 500 ms, the plume disperses and no further fragmentation is observed D: Holmiumlaser energy was set at 375 mJ/pulse for 250-ms pulse duration.

Holmium:YAG Laser Lithotripsy 29

radial expansion, so it expanded more axially.This configuration did not produce fragmentation,as proven by an image captured after the bubblehad collapse at about 600 ms, shown in Figure 6C.The pressure peak due to the collapse was lessthan 20 bars, and did not induce any photome-chanical damage to the urinary calculus.

When lithotripsy was performed with the fi-ber perpendicular to the calculus surface, a pres-sure transient indicating a small pressure peak ofabout two bars for the COM stone occurred uponbubble collapse at 300 ms (Fig. 7B). Irradiation ofthe cystine stone produced pressure fluctuationsof less than one bar, with no apparent pressurepeak due to cavitation collapse (Fig. 8B). How-ever, plume formation began approximately 50 msafter the initiation of the Ho:YAG pulse as re-ported previously, and reached its maximumaround 150–200 ms (Fig. 4A and 9A) for the COMcalculus before any pressure peak was observed,and around 400 ms for the cystine calculus. Im-ages captured on the COM stone just after thesmall pressure peak (∼ 2 bars; Fig. 9B) at 300 msshow barely visible fragmentation due to cavita-tion collapse. This illustrates, again, that the

pressure wave induced by cavitation collapse, orby a presumed photomechanical mechanism isnot responsible for significant calculus fragmen-tation by the long-pulse Ho:YAG laser.

Comparison of Fiber Orientation

When a COM calculus was ablated with thefiber parallel to its surface in water for 330 sec-onds with 400 mJ/pulse at 2.5 Hz, a small pit (∼ 1mm diameter and < 0.5 mm deep) resulted (Fig.10B). On the contrary, when a COM calculus wasablated with the fiber placed perpendicularly toits surface, a large crater was pictured with theCCD camera. The crater measured approximately

Fig. 3. For contact-mode, holmium lithotripsy begins at 50 msafter the onset of the holmium laser. This discovery does notagree with conventional knowledge of LISL, whereby litho-tripsy occurs by shock wave generation during plasma expan-sion in nanosecond Nd:YAG laser less than 1 ms after theonset of the laser, and during cavitation collapse in microsec-ond pulsed-dye laser more than 200 ms after the onset of thelaser, respectively. Pressure measurements did not revealany shock wave during bubble expansion and only a pressurewave on the order of 2 bars were detected at 1-mm distancefrom the location of cavitation collapse at 300 ms for COMcalculus. Holmium laser energy was set at 375 mJ/pulse for250-ms pulse duration.

Fig. 4. Maximum expansion of “plume” formation, a mixtureof vapor bubble and dust containing calculus fragmentationupon thermal breakdown of A: COM, and B: Cystine calculi.Holmium laser energy was set at 375 mJ/pulse for 250-mspulse duration.

30 Chan et al.

5 mm in width and about 2 mm in depth, as shownin Figure 10C.

Mass-Loss Measurements

Calculi were divided into three groups formass-loss measurements where there was no sta-

Fig. 5. Holmium lithotripsy at 45-degree angle: A: Showsbubble expansion at 50 ms. B: Shows that the maximum ex-pansion of the vapor bubble and the fragmented dust is com-paratively much less than that in contact-mode. C: Upon col-lapse at 350 ms, only small fragments continued to be ejectedfrom the ablated site. Holmium laser energy was set at 375mJ/pulse for 250-ms pulse duration.

Fig. 6. Holmium lithotripsy with parallel fiber orientationalong the urinary calculus. A: Illustrates the half-sized elon-gated vapor bubble at its maximum expansion around 350 ms.Upon cavitation collapse around 450 ms B: A pressure waveon the order of 20 bars was detected, but no fragmentationoccurred thereafter, C. Holmium laser energy was set at 375mJ/pulse for 250-ms pulse duration.

Holmium:YAG Laser Lithotripsy 31

Fig. 7. Time profile of the holmium laser pulse (top) and themeasured pressure for each figure. A: Bubble only: Ho:YAGlaser pulse begins at 0 ms. A small pressure bump character-istic of long-pulse mid-infrared laser was detected after laseronset. Bubble collapse occurred at 450 ms inducing a pressurepeak >20 bars (typically above 15 bars), rebounded, and re-collapsed at about 600 ms. B: Perpendicular ablation on COM:Ho:YAG laser pulse begins at 0 ms. Upon cavitation collapse,a small pressure peak about 2 bars was detected at around300 ms. C: Parallel ablation on COM: Ho:YAG laser pulsebegins at 0 ms. A pressure peak <20 bars at 545 ms (range: 450to 600 ms) was detected; sometimes a second pressure peakoccurred. Holmium laser energy was set at 400 mJ/pulse for350-ms pulse duration.

Fig. 8. Time profile of the holmium laser pulse (top) and themeasured pressure for each figure. A: is the same as Figure7A; shown here for ease of comparison. B: Perpendicular ab-lation on cystine: Ho:YAG laser pulse begins at 0 ms. No sig-nificant pressure peak was observed (∼1 bar). C: Parallel ab-lation on cystine: A pressure peak was detected at 475 ms (<20bars typical and occurred in the range of 450 to 500 ms); asecond pressure wave occurred at about 700 ms. Holmiumlaser energy was set at 400 mJ/pulse for 350-ms pulse dura-tion.

32 Chan et al.

tistically significant pairwise difference in the ini-tial average mass of each group. With the previ-ously mentioned laser parameters, calculus mass-loss was significantly greater for dehydratedcalculi in air than for hydrated calculi in air,and both were greater than the mass-loss for cal-culi submerged in water. Analysis indicated that,with P < 0.001, there was statistical significanceamong the results of the three experimental con-figurations (Table 2).

Effects of Initial Stone Temperatureon Fragmentation

From Table 3, it can be seen that mass-losswas significantly lower for pre-cooled COM andcystine calculi (2.2 ± 0.9 and 0.8 ± 0.4, respec-

Fig. 10. A: Shows the control image prior to holmium litho-tripsy in water. With a parallel fiber orientation in contactwith the calculus surface, B, no significant fragmentation oc-curred after delivery of 330 J of holmium energy. With a per-pendicular fiber orientation in contact-mode, C, however,lithotripsy occurred leaving behind a crater measuring 5-mmin width and 2-mm in depth.

Fig. 9. Illustrates that holmium lithotripsy occurred prior tocavitation collapse. A: Shows a maximum plume formation at150 ms after laser onset on a COM calculus. B: Shows nofurther fragmentation upon bubble collapse at about 300 ms.Holmium laser energy was set at 400 mJ/pulse for 350-mspulse duration.

Holmium:YAG Laser Lithotripsy 33

tively) compared to uncooled COM and cystine atroom temperature (5.2 ± 1.6 and 2.2 ± 1.1, respec-tively).

Pre- and Post-Irradiation Chemical Analysis

For COM, the parent calculi and fragments$710 mm showed no change in composition. Thefragments < 710 mm were a combination of cal-cium carbonate and COM. For CHPD calculi, nochemical changes were detected in the parent cal-culi and fragments $ 710 mm. CHPD crater sur-faces showed amorphous calcium phosphate (car-bonate apatite and hydroxylapatite), whereasfragments < 710 mm showed CHPD, amorphouscalcium phosphate, and calcium pyrophosphate.For MAPH calculi, no chemical change was foundin the parent calculi. MAPH crater surfaces, frag-ments $710 mm, and fragments < 710 mm showedMAPH and the presence of ammonium carbonateand magnesium carbonate.

For cystine calculi, all fragments remainedas cystine. The water in which cystine lithotripsywas performed, however, showed the presence offree sulfur. With electron impact ionization massspectrometric (EI/MS) analysis, substantial quan-tities of free sulfur and minute traces of cysteinein the aqueous medium above the laser-treatedcalculi were detected. There was no evidence ofeither free sulfur or cysteine in a washed sample

of the remaining solid material after laser irra-diation. Likewise, there was no free sulfur or cys-teine in untreated calculi. By electrospray ioniza-tion mass spectrometry (ESI/MS), it was observedthat only cystine was present in the washed laser-treated calculus residue and that the supernatantcontained predominantly cystine along with verylow levels of cysteine.

For uric acid calculi, all fragments recoveredwere uric acid. However, the water in which uricacid lithotripsy was performed indicated the pres-ence of cyanide in all specimens. This cyanide testwas repeated at three separate toxicology labora-tories with consistent results.

DISCUSSION

It is well documented that shock wave-induced fragmentation occurs during plasma ex-pansion and upon bubble collapse in Q-switchedNd:YAG and pulsed-dye laser lithotripsies [16–19], respectively. In nanosecond Q-switchedNd:YAG lithotripsy, both plasma expansion andbubble collapse generate shock waves with mag-nitudes in excess of 100 bars. These shock wavesresult in instantaneous fragmentation as theytraverse the calculus surface and mechanicallycouple their energy to the target. In microsecondpulsed-dye laser lithotripsy, fragmentation occursonly upon bubble collapse because the microsec-ond pulse duration creates a delayed plasma ex-pansion and a smaller initial shock wave.

In this study, no fragmentation was ob-served in fast-flash images during the earlystages of bubble expansion or upon bubble col-lapse. The long-pulse Ho:YAG laser did not gen-erate shock waves or pressure waves on the orderof magnitude necessary to contribute to fragmen-tation. Concurrent pressure measurements indi-cated that maximum fragmentation of the COMand cystine calculi was not a consequence of shock

TABLE 2. COM Calculus Mass-Loss (mg, Mean ± Standard Deviation) for aTotal of 150-J Ho:YAG Energy (0.5 J/pulse at 8 Hz)a

Calculi inwater

(n 4 13)

Hydratedcalculi in air

(n 4 14)

Dehydratedcalculi in air

(n 4 18) P-value

Pre-Ablationcalculus mass 141 ± 22 128 ± 48 118 ± 78 0.8

Post-Ablationcalculus mass-loss 17 ± 3 25 ± 9 40 ± 12 0.001

aPairwise comparisons were statistically significant at an a # 0.01 for calculi in watervs. hydrated calculi in air and dehydrated calculi in air; and for hydrated calculi in airvs. dehydrated calculi in air.

TABLE 3. Mass-Loss (mg, Mean ± StandardDeviation) for COM and Cystine Calculi at aPre-Cooled Temperature (> −80°C) and at RoomTemperature (∼ 23°C), in Air

Pre-cooledtemperature

(> −80°C)(n 4 5)

Roomtemperature

(∼ 23°C)(n 4 5) P-value

Calcium oxylatemonohydrate (COM) 2.2 ± 0.9 5.2 ± 1.6 0.02

Cystine 0.8 ± 0.4 2.2 ± 1.1 0.05

34 Chan et al.

wave impact (Fig. 4 and 9A). With COM calculi,only a small pressure wave (< 2 bars, Fig. 7B) wasrecorded upon bubble collapse. No pressure wavewas detected for cystine calculi except for a smallpressure transient (< 1 bar, Fig. 8B). However,when a 20-bar pressure wave (Figs. 7C and 8C)was recorded during bubble collapse for a parallelfiber along the calculus surface, no fragmentationwas observed in fast-flash images. If shock wavesproduced by plasma expansion or bubble collapseresulted in lithotripsy, the parallel fiber orienta-tion would have caused fragmentation, sinceshock waves propagate spherically in all direc-tions [16–20]. This suggests that during routineHo:YAG clinical use, urinary calculus fragmenta-tion is not shock wave-induced [21].

On the other hand, there was evidence thatlithotripsy occurred by direct absorption of laserenergy by the calculus. It was observed that litho-tripsy only began when the “Moses effect” [22–23]allowed channeling of the laser beam through thevapor bubble to the calculus surface (Fig. 2). Theabsorption coefficient, ma, of Ho:YAG laser in wa-ter was 25 cm−1 associated with 1/e of light pen-etration according to Beer’s Law; the penetrationdepth, 1/ma, was 400 mm [26]. In the absence ofstress confinement in the medium because of thelong laser pulse duration [20,23,27–29], a layer ofwater 400 mm in depth was evaporated rapidlyupon irradiation without plasma expansion orshock wave generation. The conversion of waterinto vapor created an expanding vapor bubble be-neath the delivery fiber. Since the density of va-por is 1,672 times less than that of water (at100°C, rwater 4 1,000 g/cm3 and rvapor 4 0.598g/cm3), the absorption coefficient of vapor was re-duced proportionally to about 0.015 cm−1. Hence,the vapor bubble acted as a channel allowing con-tinuous laser energy transmission to the exposedcalculus surface.

In the early phase of our research, we con-sistently observed the hot or whitish “glow” [7,21].Upon identifying this “whitish glow” as scatteredlight originating from the holmium flashlamp, weplaced a long pass filter (1,000 nm) at the outputcoupler of the laser cavity to eliminate the Xenonarc emission that was delivered through the clini-cal fiber along with the holmium energy. As canbe seen, all fast flash images (Figs. 2–6) captureddid not show the presence of a hot “glow” at thelaser pulse energies used to perform our experi-ments. Xenon arc lamps are the commonly usedholmium pumps and are known to have a black-body temperature of 5,200 K [30]. If care is not

taken, the “whitish glow” can easily be mistakenas a result of holmium-calculus interactions.

Previous studies have shown that bubble col-lapse does not necessarily result in shock wave orsignificant pressure wave formation [20,24,29].This is because long-pulse mid-infrared laserslike the Ho:YAG laser produce pear-shaped orelongated vapor bubbles that eventually termi-nate with multiple collapses instead of sphericalbubbles that collapse to a point. These multiplecollapses weaken the mechanical coupling effi-ciency of the bubble, generating only weak pres-sure waves.

Experimental results (Figs. 7C and 8C) showthat the maximum pressure wave (∼ 20 bars) wasgenerated when the delivery fiber was parallel tothe surface of the calculus. This pressure ampli-tude was significantly larger than that recordedwhen the delivery fiber was placed in contact andperpendicular to the target, as in a typical clinicalprocedure (P < 2 bars, Figs. 7B and 8B). Uponbubble collapse, however, no fragmentation wasdetected (Fig. 6C) and no effective lithotripsy wasrecorded after delivery of 330-J laser energy (Fig.10B). Only a small pit was visible in Figure 10B,most likely a result of the diverging laser beamfrom the delivery fiber (large numerical aperture)and a wallward jet [31–34] upon bubble collapse.These data rule out stone ablation through pho-tomechanical or photoacoustical mechanisms,consistent with the observation made by Beghuinet al. that vapor bubbles have “a minimal inci-dence on stone fragmentation” [29].

Chemical analyses of various urinary calculipost-lithotripsy consistently revealed thermalbreakdown components (Table 1). Unlike LISLwhere 22 mJ of nanosecond Q-switched Nd:YAGand 100 mJ of microsecond pulsed-dye laser en-ergies were enough to create fragmentation bymeans of shock wave coupling, the long-pulseHo:YAG laser has to operate above the calculusthreshold ablation in order for lithotripsy to oc-cur. Although these threshold energy levels havenot been quantified, clinical lithotripsy is typi-cally performed between 300 mJ/pulse to 2J/pulse. These high energy levels are necessary toraise the temperature of the target calculus overthe threshold breakdown temperature of itschemical composition, as opposed to the tempera-ture needed for water vaporization.

In air, all Ho:YAG energy was deposited di-rectly onto the surface of the dried urinary calcu-lus, producing the greatest mass-loss among thethree scenarios (Table 2). For hydrated urinary

Holmium:YAG Laser Lithotripsy 35

calculus, some laser energy was required to heatand possibly vaporize surface and interstitial wa-ter. For urinary calculi submerged in water, moreenergy was needed to vaporize the water layerbetween the fiber-tip and the surface of the cal-culus in addition to heating interstitial water.Therefore, it is not surprising that calculus in wa-ter yielded the least amount of mass-loss. It islikely that interstitial water expansion and explo-sive vaporization do not contribute directly tofragmentation, but may facilitate the ejection ofdecomposed stone fragments (or cysteine and freesulfur gas in the case of cystine lithotripsy) as canbe seen in the plume formation which beginsabout 50 ms after the initiation of Ho:YAG irra-diation (Fig. 3).

Izatt et al. have theorized that water withinbone (hydroxyapatite) is heated upon Ho:YAG ir-radiation [35]. As this water rapidly vaporizes, itproduces momentum necessary for calcium frag-ment ejection. This thermal mechanism on explo-sive vaporization of interstitial water has beenproposed for Ho:YAG lithotripsy [29]. However, itis well known that clinical long-pulse Ho:YAG la-ser ablation of calcified tissue occurs without sig-nificant thermal damage to surrounding tissue,i.e., thermal confinement occurs [36]. In addition,holmium lithotripsy creates sharply demarcatedcraters, with microscopic preservation of crystal-line lamella on the crater surface, implying aregular, consistent, and stereotypical laser-calculus interaction [9]. All these observations in-dicate that holmium lithotripsy is governed bythermal breakdown of calculus compositionthrough direct laser absorption, whereas intersti-tial water and vapor expansion merely facilitatesthe ejection of fragments within the volume of de-composed compounds, leaving the parent calculusor original compound intact.

Mass-loss measurement based on the initialtemperature of calculus further strengthens thehypothesis of a dominant photothermal lithotrip-sy. COM and cystine calculi produce lower mass-loss when the calculi are ablated after removingfrom a −80°C freezer, compared to calculi at roomtemperature. The study shows that more laser en-ergy must be deposited for the pre-cooled calculito overcome the thermal threshold temperaturefor chemical breakdown.

The confirmation that long-pulse Ho:YAGlithotripsy is due to photothermal effects has im-portant clinical implications. The finding that onethermal by-product of uric acid lithotripsy is cya-nide, raises safety issues [11,37–39]. Lithotripsy

of the holmium laser is most efficient in the con-tact mode with the delivery fiber placed perpen-dicular to the calculus surface for laser-calculusinteraction. LISL, however, requires a small fi-ber-calculus separation to maximize the couplingof photoacoustical energy [16–19,24]. In addition,since shock waves propagate spherically, LISLrisks collateral mucosal injury and bleeding. Theholmium laser does not generate significant shockwaves and its thermally-induced fragmentation ishighly directional as reported in the fiber orien-tation experiments above. Hence, the long-pulseHo:YAG laser is a solution to the problem of col-lateral tissue damage. Finally, the negligible me-chanical effects should minimize the problem ofstone migration during treatment.

CONCLUSION

We conclude that long-pulse Ho:YAG laserlithotripsy is primarily a photothermal mecha-nism. Direct irradiance of the urinary calculi bythe holmium laser through a laser-induced vaporchannel (the “Moses effect”) increases the tem-perature of the irradiated volume above a criticalthreshold temperature, causing the chemicalbreakdown of the calculus. Chemical breakdownweakens the mechanical integrity of the irradi-ated volume, allowing the vapor bubble and inter-stitial water or vapor expansion to facilitate theejection of fragmented breakdown products.

REFERENCES

1. Yiu MK, Liu PL, Yiu TF, Chan AYT. Clinical experiencewith holmium:YAG laser lithotripsy of ureteral calculi.Lasers Surg Med 1996;19:103–106.

2. Razvi HA, Denstedt JD, Chun SS, Sales JL. Intracorpo-real lithotripsy with the holmium:YAG laser. J Urology1996;156:912.

3. Das A, Erhard MJ, Bagley DH. Intrarenal use of the hol-mium laser. Lasers Surg Med 1996;19:103–106.

4. Adams DH. Holmium:YAG laser and pulsed dye laser: acost comparison. Lasers Surg Med 1997;21:29–31.

5. Teichman JMH, Rao RD, Rogenes VJ, Harris JM. Ure-teroscopic management of ureteral calculi: electrohy-draulic versus holmium:YAG lithotripsy. J Urology 1997;158:1357–1361.

6. Teichman JMH, Glickman RD, Harris JM, Holmi-um:YAG percutaneous nephrolithotomy: the laser inci-dent angle matters. J Urology 1998;159:690–694.

7. Dushinski JW and Lingeman JE. High-speed photo-graphic evaluation of holmium laser. J Endourology1998;12(2):177–181.

8. Teichman JMH, Vassar GJ, Bishoff JT, Bellman GC. Hol-mium:YAG lithotripsy yields smaller fragments than

36 Chan et al.

pulsed dye, lithoclast, or electrohydraulic lithotripsy. JUrology 1998;159:18–27.

9. Vassar GJ, Teichman JMH, Glickman RD. Holmi-um:YAG lithotripsy efficiency varies with energy density.J Urology 1998;160: 471–476.

10. Teichman JMH, Vassar GJ, Glickman RD. Holmium:YAG lithotripsy efficiency varies with stone composition.J Urology 1998;52:392–397.

11. Teichman JMH, Vassar GJ, Glickman RD, Beserra CM,Cina SJ, Thompson IM. Holmium:YAG lithotripsy: pho-tothermal mechanism converts uric acid calculi to cya-nide. J Urology 1998;160:320–324.

12. Teichman JMH, Rogenes VD, McIver BJ, Harris JM. Hol-mium:YAG laser cystolithotripsy of large bladder calculi.J Urology 1997;50:44.

13. Thomas S, Pensel J, Engelhardt R, Meyer W, HofstetterAG. The pulsed dye laser versus the Q-switched Nd:YAGlaser in laser-induced shock-wave lithotripsy,” LasersSurg Med 1988;8:363–370.

14. Spindel ML, Moslem A, Bhatia KS, Jassemnejad B, Bar-tels KE, Powell RC, O’Hare CM, Tytle T. Comparison ofholmium and flashlamp pumped dye lasers for use inlithotripsy of biliary calculi,” Lasers Surg Med 1992;12:482–489.

15. Goldey CL, Rosen DI, Hayes GB, Willscher MK, Roth RA.Development of a smart Holmium:YAG laser lithotriptor.Lasers Surg Med 1997;21:20–28.

16. Rink K, Delacretaz G, Salathe RP. Fragmentation pro-cess induced by nanosecond laser pulses. Appl Phys Lett1992;61(22):2644–2646.

17. Rink K, Delacretaz, Salathe RP. Influence of the pulseduration on laser induced mechanical effects. SPIE 1994;2077:181–194.

18. Rink K, Delacretaz G, Salathe RP. Fragmentation pro-cess of current laser lithotriptors. Lasers Surg Med 1995;16:134–146.

19. Rink K, Delacretaz G, Salathe RP. Fragmentation pro-cess induced by microsecond laser pulses during litho-tripsy. Appl Phys Lett 1992;61(3):258–260.

20. Jansen ED, Asshauer T, Frenz M, Motamedi M,Delacretaz G, Welch AJ. Effect of pulse duration onbubble formation and laser-induced pressure waves dur-ing holmium laser ablation. Lasers Surg Med 1996;18:278–293.

21. Schafer SA, Durville FM, Jassemnejad B, Bartels KE,Powell RC. Mechanisms of biliary stone fragmentationusing the Ho:YAG laser. IEEE Trans On Biomedical En-gineering March 1994;41(3):276–283.

22. Isner JM. Blood. In Isner JM, Clarke R, editors. Cardio-vascular Laser Therapy. New York: Raven Press; 1989. p39–62.

23. Jacques SL. Laser-tissue interactions: photochemical,photothermal, and photomechanical. Surg Clin N Am1992;72:531–558.

24. Zhong P, Tong HL, Malenbaum J, Cocks FH, PremingerGM. Transient cavitation and acoustic emission producedby different laser lithotripters. J Endourology 1998;12(4):371–378.

25. Rieders F, Cyanides, Type B procedure. In Sunshine I,editor. Methodology for analytical toxicology. Cleveland:CRC Press; 1975. p 114–115.

26. Jansen ED, van Leeuwen TG, Motamedi M, Borst C,Welch AJ. Temperature dependency of the absorption co-efficient of water for mid-infrared laser radiation. LasersSurg Med 1994;14:258–264.

27. Welch AJ, van Gemert MJC. Optical-thermal response oflaser-irradiated tissue. New York: Plenum Press; 1995. p709–763.

28. Frenz M, Pratisto H, Konz F, Jansen ED, Welch AJ, We-ber HP. Comparison of the effects of absorption coeffi-cient and pulse duration of 2.12 mm and 2.79 mm radia-tion on ablation of tissue. IEEE J Quantum Electronics1996;32(12):2025–2036.

29. Beghuin D, Delacretaz G, Schmidlin F, Rink K. Fragmen-tation process during Ho:YAG laser lithotripsy revealedby time-resolved imaging. SPIE 1998;3195:220–224.

30. Koechner W, Solid-state laser engineering. Springer Se-ries in Optical Sciences, 3rd Edition. New York: Springer-Verlag; 1992. p 282.

31. Lauterborn W, Bolle H. Experimental investigations ofcavitation-bubble collapse in the neighbourhood of a solidboundary. J Fluid Mech 1975;72(2):391–399.

32. Plesset MS, Chapman RB. Collapse of an initially spheri-cal vapour cavity in the neighbourhood of a solid bound-ary. J Fluid Mech 1971;47(2):283–290.

33. Godwin RP, Chapyak EJ. Laser mass ablation efficiencymeasurements indicate bubble-driven dynamics domi-nates laser thrombolysis. SPIE 1998;3245:4–11.

34. Rudhart M, Hirth A. Use of an absorbent in laser litho-tripsy with dye lasers: in vitro study of fragmentationefficiency and jet formation. J Urology 1994;152:1005–1008.

35. Izatt JA, Albagli D, Itzkan I, Feld M. Pulsed laser abla-tion of calcified tissue: physical mechanisms and funda-mental parameters. SPIE 1990;1202:133–140.

36. Cervanin I. A comparison of the effects of Nd:YAG andHo:YAG laser irradiation on dentine and enamel. AustDent J 1995;40:79–84.

37. Mahn WJ. Academic laboratory chemical hazards guide-book. New York: Van Nostrand Reinhold; 1991. p 187–190.

38. Ballantyne B, Mares TC. Toxicology of cyanides. Bristol:Wright and Sons; 1987.

39. Teichman JMH, Champion PC, Wollin TA, Denstedt JD.Holmium:YAG lithotripsy of uric acid calculi. J Urology1998;160:2130–2132.

40. Budavari S, O’Neil MJ, Smith A, Heckelman PE, editors.The Merck Index. Rahway, NJ: Merck & Co.; 1989.

Holmium:YAG Laser Lithotripsy 37