Free Electron Laser Ablation of Urinary Calculi: An Experimental Study

,


I. INTRODUCTION
I NTERVENTIONAL treatments of urinary calculi have im- proved drastically over the past two decades.By the early 1990s, four major lithotripsy modalities have been used clinically to remove urinary or renal calculi.They include electrohydraulic lithotripsy (EHL), ultrasonic lithotripsy, extracorporeal shockwave lithotripsy (ESWL) and laser lithotripsy [1].Each of these modalities has its limitations.EHL, which uses shockwaves induced by electrical discharges, is only suitable for fragile calculi and may cause perforation of the ureter and adjacent tissue.Ultrasonic lithotripsy has been shown to be effective for the removal of both ureteral and renal calculi, but the 11.5 French ureteroscope that must be used to deliver the large ultrasonic probe (6 French) restricts its use to the lower ureter(s).ESWL has the advantage of being performed outside the body, and therefore is a noninvasive procedure.However, visualization of the calculus during the procedure is limited and renal injury may occur.Laser lithotripsy allows the use of small-diameter catheters, which provide access to both ureteral and renal calculi.This technique has also been shown to reliably remove calculi with minimal damage to healthy tissue without the use of ureteroscopy.
Laser ablation was introduced into clinical lithotripsy during the mid-1980s [2]- [5].The ablation or fragmentation of urinary calculi during that period was achieved with a flashlamp pumped pulsed-dye laser delivered through an optical fiber.Fragmentation occurred due to shockwaves generated as a result of plasma expansion and cavitation bubble collapse [6], [7].By 1996, the Holmium : YAG laser had become popular not only in lithotripsy but also in other urologic procedures such as laser resection of the prostate.The holmium laser reduced the calculus to passable size by ejecting fine particles from and producing well-demarcated craters on the calculus surface.Several investigators observed that the fragmentation mechanism of urinary calculi by the holmium laser was suggestive of a photothermal process [8], [9].
Two critical factors required for the acceptance of a laser medical procedure involving ablation are 1) maximizing ablation efficiency while 2) minimizing damage to surrounding healthy tissue.To achieve this, the physical processes of laser ablation must be understood.Unfortunately, these processes are complicated.Previous studies indicate that the physical mechanism is a function of laser parameters and physical properties of the material to be ablated [10].Some researchers have shown that laser pulse duration determines whether an ablation process is predominantly photothermal or photoacoustical [10], [11].Izatt et al. [12] and more recently Nahen et al. [13] have explored the effect of debris screening during a laser pulse upon ablation efficiency.Walsh and Cummings [14], and Jansen et al. [15] have examined the dependence of optical absorption upon temperature at 2.94 and 1.96 m, and how dynamic changes in absorption affect the outcome of the ablation process.The roles of tissue mechanical properties such as tensile strengths and viscosity on ablation have also been explored [16]- [18].
Because of the many advantages of the mid-infrared holmium laser over the flashlamp pumped pulsed-dye laser in clinical lithotripsy, researchers have pondered over the most efficient wavelengths for such a procedure.However, the role of wavelength dependent optical absorption and its effect on urinary calculi ablation has not been extensively studied.Daidoh et al. [19] have suggested that 3 and 6 m may be the optimum wavelengths for ablating calcium oxalate monohydrate (COM) and uric acid calculi because of their peak absorptions.Unfortunately, these authors did not introduce a sound theoretical or physical basis for their postulation.In a recent study of the dominant ablation mechanism in long-pulsed Ho : YAG laser lithotripsy [9], [20], we observed that the holmium laser-calculus ablation dynamics are comparable to several previous studies of hard-tissue laser ablation [12], [21].A photothermal ablation model was used in these studies to analyze the experimental results [12], [21]- [24].We hypothesize that the optimal wavelength(s) in the infrared spectrum for calculus ablation is highly dependent upon the optical absorption properties as specified by the value of the absorption coefficient (mm ).The optimal wavelength can be defined as the wavelength at which ablation efficiency, as measured in terms of mass-loss/Joule (mg/J), or ablation depth/Joule ( m/J), etc., is the highest relative to other achievable laser wavelengths.
Most urinary calculi strongly absorb infrared energy from 2 to 10 m with relatively low scattering.Acoustic velocities in the dense crystal structures of urinary calculi are similar to or higher than the speed of sound in water ( m/s) [24].Thus, for highly absorbed wavelengths that are used for ablative purposes, microsecond or longer laser pulses are not stress confined, and therefore do not cause fragmentation by means of shock waves as do submicrosecond laser pulses.When the process is not stress confined, ablation becomes dominated by photothermal mechanisms such as vaporization, melting, and chemical decomposition [20], [24].A simple model introduced by Olmes et al. [23] for determining the threshold fluence, (J/cm ), for thermal ablation can be expressed as (1) where (cm ) is the optical absorption coefficient of the calculus, (kg/m ) is the calculus density, (J/g.K) is the specific heat, is the threshold temperature for vaporization, melting, or chemical decomposition, is the ambient temperature, (J/g) is the latent heat for vaporization, melting, or chemical decomposition, and is the heat of ablation (J/cm ).Assuming that the calculus density, specific heat, and threshold temperature are constants, the heat of ablation is also a constant thermal property of a particular calculus type.The predicted dependence of the threshold fluence on the calculus optical absorption as a function of the heat of ablation is illustrated in Fig. 1 for three values of the heat of ablation, where .The photothermal ablation model used in this study has been derived by previous researchers for ablation studies in dentine and beef shank bone [23], [25]- [27].During the period of no heat conduction, the ablation depth ( m) is (2) where is the intrapulse debris attenuation coefficient (cm ), and is the incident laser fluence (J/cm ).Detailed derivation of (2) is discussed elsewhere [28], [29].Ostertag et al. and Hibst et al. [25], [26] have verified this model in their studies of dentine ablation depth as a function of incident laser fluence.Izatt et al. [30] has attempted to apply this ablation model to interstitial water vaporization and expansion for extracting the optical properties of beef bone shank.
If multiple laser pulses are employed to achieve sufficient ablation depth for accurate measurement, and if one pulse does not affect the result of the next pulse, then the total ablation depth is (3) where is the laser fluence of individual pulses for a total of pulses.Equation (3) assumes the laser pulses do not overlap in time, and that successive pulses are sufficiently far apart in time that debris screening from a previous pulse does not affect the next pulse.Pulse-to-pulse variation in material optical and thermal properties would affect the accuracy of (3) where we assume constant parameters.We have extended the theoretical model to evaluate the dependence of the ablation depth on the optical properties of the urinary calculi and ejected particles in the infrared wavelengths using the free electron laser.By understanding the ablation dynamics, and calculus optical properties it may be possible to determine the optimum wavelength for fragmentation.
The objectives of this study are: 1) to investigate the mechanism(s) involved in urinary calculus ablation; and 2) to determine the threshold fluences for ablation , and the corresponding ablation depths , and to examine their correlation with optical absorption.The most common human urinary calculi are used in this in vitro study in which their optical absorption properties are known, for the purpose of investigating the physical dynamics involved in the ablation process as well as addressing their relevance for future clinical studies.Experimental observations and deviations from (1) and ( 3) are used to determine the contribution of other physical mechanisms and dynamic properties involved in the ablation processes.

A. Calculus Specimen
Four types of human urinary calculi were used in this experiment.These calculi were composed of, respectively, 95% uric acid, 95% cystine, 95% calcium oxalate monohydrate (COM), and 90% magnesium ammonium phosphate hexahydrate (MAPH) and were obtained from a calculus analysis laboratory (Louis C. Herring Company, Orlando, FL).The calculi were cut with a diamond saw, producing samples 2-3 mm in thickness with flat (approximately parallel) surfaces.After cutting, the calculi were dehydrated in a lyophilizer and stored in sample container with room air until the time of use.

B. Laser Source and Delivery
The pulsed infrared Free Electron Laser (FEL) at Vanderbilt University, which is tunable from 2.1 to 9.8 m, was used for this experiment.The FEL macropulses ( -s FWHM) consisted of a sequence of 1-ps micropulses with a 350-ps interpulse interval.Hence, one macropulse had approximately 8550 to 14 250 micropulses.The FEL had a Gaussian beam profile, exhibiting laser emission with the TEM mode.These were the characteristic pulse structures and properties of the FEL at Vanderbilt University.Although it was not known if individual picosecond micropulses might contribute to fragmentation, the short macropulse duration ensured thermal confinement in highly absorbing specimens demonstrated in all previous and current clinical laser lithotripsy modalities.
Single pulses were used for threshold measurements, whereas a repetition rate of 5 Hz was used for ablation rate measurements.Wavelengths used in this experiment were 2.12, 2.50, 2.94, 3.13, 3.21, 3.62, 4.00, 4.57, 4.74, 4.77, 5.00, 5.60, 5.99, 6.04, 6.17, 6.20, 6.26, 6.29, 6.31, and 6.45 m.The 2.12 and 2.94 m wavelengths correspond to the wavelengths for the Ho : YAG laser and Er : YAG laser, respectively, and served as references for comparison to results from commercial lasers.Wavelengths were chosen to correspond to the peaks, inclinations, and valleys within the infrared absorption spectra of individual calculus types referenced from Dao and Daudon [31] and Daidoh et al. [19].The absorption spectra from these references are in relative optical absorbance units (a.u.) and not absolute values of absorption coefficients (cm ).
The experimental setup for the measurement of threshold fluence and ablation depth is illustrated in Fig. 2. The FEL beam was delivered via a vacuum beam transport system, and reached our setup through a BaF window.A footswitch-controlled mechanical shutter blocked the laser beam.When the shutter was opened, the laser beam was delivered to the urinary calculus via a 2-in diameter, 200-mm focal length CaF lens.The FEL beam is linearly polarized and the pulse energy was adjusted by rotating a double Brewster plate polarizer located within the vacuum chamber.A fast infrared (IR) photodetector (pyroelectric, Molectron P3-01, Portland, OR) was used to monitor the temporal profile of the laser pulse on a digital oscilloscope (Tektronix TDS 640A, Beaverton, OR).Two energy probes (Molectron J-8LP, Portland, OR) were used prior to the experiment to measure the energy ratio (correction factor output energy/reference energy).During each experiment, only the reference energy was recorded and was scaled afterwards according to the correction factor.For each wavelength, the spot diameter of the incident laser beam was measured at the focal point using a standard knife-edge technique [32].The FEL spatial beam profile was approximately Gaussian [33].The FEL spot diameter used was 210 10 m.The sample calculus was mounted on a manual precision -translation stage with its top surface aligned orthogonally to the optical axis and at the focal plane.Irradiation sites corresponded to a 7 7 grid (i.e., maximum data points, , grid spacing 1 mm).Post-experimental calculation of the incident laser fluence was performed by dividing the corrected incident laser energy by the measured spot area based on the diameter.

C. Ablation Threshold Fluence
The threshold fluence for ablation was defined as the incident laser fluence required to produce mass ejection from the urinary calculus.A single-shot laser pulse was gated (by opening the shutter) to each point on the 7 7 grid.A doubleblind study was conducted in which an operator altered the pulse energy level around the anticipated threshold.The observer was alerted before the operator gated a pulse to the calculus and each pulse was scored as a 1 (mass ejection) or a 0 (no mass ejection).Visual detection of plume formation was aided with forward illumination and a dark background as depicted in Fig. 2. No visual magnification was used.The double-blind study prevented biased decisions.Post-experimentally, results were an-Fig.3. A probit curve for uric acid calculus irradiated at 3.13 m.The probit curve determines the probability of ablation given a range of incidence laser fluences (J/cm ).The fiducial limits, which express the confidence interval, were set at 63% (1=e) for all of our calculations.The threshold fluence is defined as the value at which the probability of ablation is 50% (ED ), estimated to be H = 0:173 J/cm in this case.We set the lower and upper bounds of experimental errors at 37% (ED ) and 63% (ED ) probability for ablation, respectively, for all data points shown in Fig. 4. represents the raw threshold data points (n = 45) recorded during the experiment.ED: estimated dose.alyzed using Probit analysis [34] (Fig. 3).The threshold fluence, , was specified as the laser fluence at which ablation occurred with a 50% probability, ED .ED and ED were used as the upper and lower limits of the experimental error, respectively.This probabilistic approach was taken to account for localized variations in the optical, thermal, and mechanical properties of a specimen that might have affected the threshold fluence values for ablation.

D. Ablation Depth
For each calculus type, two cut samples from the same parent calculus were used.At each wavelength, ten ablation craters were made ( ).Each crater was produced by a fixed fluence of 5 J/cm per pulse for a total of 10 pulses (or a total fluence of 50 J/cm ) gated to the calculus surface at a 5-Hz repetition rate.The slow repetition rate was used to minimize interpulse debris screening which was determined using fast-flash photography [9], by estimating the period from laser onset to when ejected particles have left the optical axis of the laser beam (typically 10 ms).Post-experimentally, the ablation depths were measured under a 112.5 surgical microscope (Nikon SMZ-U, Melville, NY) by measuring the displacement in the focal plane between the crater surface and the crater floor using a precision translation stage (MX700 Series Micromanipulators, Siskiyou Design Instruments Inc., Grants Pass, OR).The spatial resolution of this measurement was limited by the depth of focus ( 20 m) of the microscope lens.
A CCD camera (HV-C20, Hitachi, Tokyo, Japan) and an optical coherence tomography (OCT) system were used to acquire images for qualitative post-ablation analyzes.The CCD images were used to compare the surface effects on the calculi produced by ablation at different wavelengths.OCT images (spatial reso-

B. Ablation Depth
Pump-probe images (Fig. 5) show that an ablation plume is visible as early as 1 s after the onset of the 3-5 s long laser pulse, indicating that plume attenuation might affect the ablation process during the laser pulse duration.In the same figure, it can be seen that by 2.5 ms after the laser onset, most ejected mass was removed from the optical axis away from the beam path.At 10 ms after the onset no remaining debris was visible in the beam path (image not shown).Since the experiment was performed at 5 Hz, the interval between each subsequent pulse was 200 ms-20 times longer than the maximum lifetime of the plume.Hence, it is reasonable to assume that no interpulse debris screening occurred at a pulse repetition rate of 5 Hz.
In general, the ablation depth caused by ten pulses with a fluence (per pulse) of 5 J/cm increases with absorbance or optical absorption (Fig. 6).The ablation depth as a function of laser wavelength in comparison to the absorption spectra of individual calculus types is presented in Fig. 6  Each ablation crater was irradiated with a total fluence of 50 J/cm (ten pulses at 5 J/cm ) at 5 Hz.Each vertical cross-sectional ablation crater images was recorded using an OCT system having a longitudinal spatial resolution of 23 m.(i)-(iii), illustrate the effects of infrared laser ablation as a function of increasing optical absorption (in absorbance unit).Wavelengths used were 2.12 m (i), 6.20 m (ii), 5.99 m (iii) for the uric acid calculus; 2.12 m (i), 6.04 m (ii), 6.17 m (iii) for the COM calculus; and 2.12 m (i), 3.13 m (ii), 6.31 m (iii) for the cystine calculus.Each ablation crater was irradiated with a total fluence of 50 J/cm (10 pulses at 5 J/cm ) at 5 Hz.Laser spot size was 210 6 10 m.Images were recorded using a surgical microscope (112.52) and a color CCD camera.
slowly as a function of rising optical absorption in this sequence of OCT images.In all OCT images recorded, no internal fracture (limited to fracture sizes 23 m) is seen surrounding the craters, although debris seems to have remained within some craters.Fig. 8 shows the surface of craters recorded with the surgical microscope (112.

IV. DISCUSSION
In order to identify the optimal wavelengths and elucidate the ablation mechanism, ablation threshold values and ablation rates for the most common types of human urinary calculi were measured over a broad range of the infrared spectrum.In this study, the measured threshold fluences for ablation in the infrared spectrum are strongly influenced by the optical absorption of the calculi.The experimental results indicate that if calculi are irradiated at wavelengths with higher optical absorption, the laser fluences required to initiate ablation can be significantly reduced [Fig.4(a)-(d)].In general, threshold fluences are proportional to the reciprocal of the absorbance, and thus to the absolute absorption coefficients [Fig.4(e)-(h)].These results and (1) suggest that the heat of ablation, is a constant thermal property that is not wavelength dependent.With higher optical absorption, less laser fluence is required to reach the heat of ablation.The excess laser fluence beyond threshold is spent on the removal of calculus mass and progression of the ablation front.The change in ablation depth as a function of absorbance is in fair agreement with this photothermal model [( 2)].
For low optical absorptions in all calculus types, ablation depth increased as absorption increased [Fig.6(d)-(f)].However, ablation depth was bounded for higher absorption coefficients, and eventually a decrease in ablation depth occurred for even higher optical absorption in uric acid, cystine, and COM calculi [Fig.6(d)-(f)].These results are in agreement with (2), which considers attenuation of the laser beam by the ablation plume.The experimental results indicate some plume generation 1 s after the onset of the laser pulse [Fig.5(a) and (b)].When the attenuation coefficient of the ablation plume or debris, , is much smaller than the calculus absorption coefficient , but increases with , there is a predicted decrease in ablation depth as becomes large [13], [28], [29].The bounding of ablation depth at higher optical absorption is consistent with a thermal mechanism but contrary to a process dominated by a photomechanical or photoacoustical mechanism.The photomechanical or photoacoustical mechanism removes calculus by large fragment dissociation [35], whereas the photothermal mechanism removes fine particles and creates welldemarcated craters [36].The plume contains ejected fragments that absorb and scatter the incident laser beam, thus reducing the amount of energy available for ablation.By performing the ablation at a relatively low repetition rate (5 Hz), interpulse debris screening of consecutive laser pulses was negligible compared to intrapulse debris screening.At higher optical absorption, the debris limits the amount of laser energy available for ablation and ablation depth decreases.In addition, more laser energy may be coupled in acoustic energy that fractures the crater proximity but is not strong enough to cause mechanical ablation [Fig.8(a)(iii) and (c)(iii)].
We believe the ablated mass from the irradiated calculus volume was removed before significant heat conduction could occur.While no explicit data on the thermal diffusivity of urinary calculi were found in the literature, analysis of data on similar materials such as bone and teeth (enamel and dentin) [37] indicate that thermal confinement is met for absorption depths 2 m (i.e., cm ).Experimentally, this assumption is supported by the fact that no carbonization to areas surrounding the crater was observed at 5 Hz (Fig. 8).
While threshold fluence data and ablation depth measurements generally followed the predictions from the thermal ablation model, some discrepancies occurred.For example, at 3.13 and 3.64 m in uric acid calculus [Fig.6(d)], the ablation depth was much higher than expected, indicating that the actual optical absorption of the calculi used in the experiment deviated from the absorption spectra reported by Dao and Daudon [31].Striated and irregular appearance of the urinary calculi suggested the presence of impurities (i.e., unknown amorphous organic compounds or crystalline minerals) and inhomogeneities (i.e., regions containing the dominant constituents but at different densities).
Another possibility is that the mechanical integrity of the uric acid molecule may have been compromised by wavelength specific absorption of intramolecular bonds.In this case, the irradiance wavelengths of 3.13 and 3.64 m are associated with the energy band close to the OH group in the uric acid calculus [19].
Ablation crater images obtained with OCT (Fig. 7) resemble predictions from a two-dimensional ablation model proposed by Majaron and Lukac that accounted for ablation front inclination [22].These craters are well demarcated with a tapered profile from the calculus surface to the crater base.The OCT crater images support the suggestion that pulsed laser ablation using infrared lasers with a pulse duration greater than 1 s, is predominantly photothermal [9], [20].The crater shape (Fig. 7) is uncharacteristic of photomechanical or photoacoustical interactions [7], [35], which often exhibit irregular surfaces due to large fragment dissociation and fractures.
Photothermal ablation of hard tissue in the infrared spectrum is evidenced by melting and recrystalization of particles [8], chemical decomposition [38], and stress due to rapid heating [10].In a previous FEL calculus ablation study, we reported thermal damage (charred layer 20 m) around a cystine crater when the repetition rate was increased to 15 Hz for a total of 25 pulses at 2.12-m [39].When the firing rate is increased and total irradiation time extended beyond the thermal relaxation time, sufficient heat is transferred beyond the volume of direct light absorption to carbonize the sides of the crater.
The effect of individual micropulses (1 ps) in the FEL macropulse (3-5 s) on stress generation in the irradiated volume is inconclusive [10].No acoustic measurement is known to us that could temporally resolve the FEL micropulse train-induced acoustic transients.Therefore, the potential contributions of micropulse-induced photomechanical damage in this experiment is not known.Fig. 8(a)(iii) and (c)(iii) show that fractures occurred surrounding the ablation crater surface for uric acid and cystine calculi with striated formation or with porous structures when the wavelengths selected coincide with relatively high optical absorptions (absolute absorption coefficients unknown).It is possible that the FEL micropulses accumulate mechanical stresses with a sharper pressure gradient within the optical penetration depth at higher absorption coefficients.Therefore, at higher optical absorptions, the micropulses may affect the ablation process in the calculus.In fact, a theoretical study by Uhlhorn where the pressure transients due to individual FEL micropulses were calculated using a hydrodynamic code (LATIS-3D), indicate that even though the absolute pressures are small, pressure gradients as high as 800 bar/ m are expected to be generated by individual FEL micropulses [40].However, in the present study, photomechanical disruptions did not result in gross detachment of the calculi beyond the irradiated areas.No fracture was observed in COM calculi for all the laser wavelengths we used.This is not surprising given that the COM calculus is known to be the harder calculus type among typical urinary calculi [41].
Overall, results indicated that optimal wavelengths for ablation occurred at the 3-m and 6-m high absorption bands for most urinary calculi examined (Fig. 6), as previously suggested by Daidoh et al. [19] for uric acid and COM calculi.The agreement of the experimental threshold fluence and ablation depth data with the photothermal model indicates that in laser lithotripsy, unlike bone ablation recently reported by Peavy et al. using the FEL, the laser wavelengths may be chosen based upon the mean optical absorption spectrum of the calculus and not the chromophores-specific absorption spectra [42].Interestingly, Peavy et al. concluded that in bone ablation, if any chromophores (i.e., water and hydroxyapatites) other than type-I collagen were targeted, the mechanical strength of the collagen resisted the removal of tissue even though destruction of the other constituents occurred.
From the standpoint of clinical application, the markedly lower threshold fluences and higher ablation depths around the 3-m absorption peak compared to other wavelengths implicated the potential of the Er : YAG laser, which produces radiation at 2.94 m.It can be seen from Fig. 4 that the threshold fluences at 2.94 m are 5.6 times lower for cystine calculus, 15.7 times lower for COM calculus, and 20.8 times lower for uric acid calculus as compared to those at 2.12 m.Meanwhile, the ablation depths at 2.94 m are 1.2 times higher for uric acid calculus, 2.9 times higher for COM calculus, and 1.7 times higher for cystine as compared to those at 2.12 m, the wavelength of the Ho : YAG laser, as shown in Fig. 6.If the Er : YAG laser is operated at comparable pulse durations (i.e., 200-300 s) to that of the Ho : YAG laser in clinical laser lithotripsy, it may be a reasonable assumption that similar trends or results in the threshold fluences and ablation depths are expected.
Since clinical laser lithotripsy is performed in a saline environment, the dynamics of ablation under water for these wavelengths must be considered [43].An in vitro study in water shows that the Er : YAG laser lithotripsy is more efficient than the Ho : YAG laser lithotripsy for the COM calculus [44], [45], supported by the fact that water absorption and typical absorptions for common human calculi are higher at 2.94 m than they are at 2.12 m.However, the clinical use of infrared laser radiation at 6 m is at the present time not ready for widespread use in a clinical setting.Clearly, the use of the FEL for medical applications, due to cost, space, and technical issues, is limited to few unique facilities.No commercially available or affordable 6-m laser for clinical applications is known to the authors.However, recent technological advances indicate that novel solid-state devices such as an OPO may become available in the near future that are amenable to clinical ablation purposes.Moreover, light delivery in this spectral band can now be accomplished by hollow waveguides [33], [46]- [49] or fiber optic materials such as chalcogenide [50] or silver halide [51], [52], thus allowing delivery of pulsed laser radiation in a noninvasive catheter-based fashion.Although no clinical endoscopic use of such fibers is known in surgical urology, research activities are emerging in this area.For instance, an in vitro investigation of Er : YAG laser lithotripsy using sapphire fibers has been performed, yielding very promising results that warranted further in vivo studies [45], [53], [54].

V. CONCLUSION
We have demonstrated in this study that the ablation process is highly dependent upon optical absorption.The threshold fluence for ablation of urinary calculi is inversely proportional to the absorption coefficient.The ablation mechanism exhibited by the free electron laser ( -s) in the infrared is predominantly photothermal, although results indicated that other mech-anisms such as photomechanical disruptions might be involved.Given a priori knowledge of laser-tissue interaction processes and physical properties such as the absorption coefficients, the optimal wavelengths may be selected for infrared laser ablation or fragmentation of urinary calculi.
Threshold fluences and ablation depths are a function of the wavelength dependent absorption properties of the calculi.These observations suggest that the 3-m and 6-m absorption bands are optimal for ablation or fragmentation of urinary calculus.Index Terms-Absorption, fluence, fragmentation, infrared, kidney stone, lithotripsy, radiant exposure, thermal, urology.

Fig. 1 .
Fig. 1.Prediction of the relationship between ablation threshold fluence and absorbance for some hypothetical tissues with different heat of ablation, W .In this case, W > W > W .

Fig. 4 .
Ablation threshold fluences (J/cm ) versus wavelength (m) for (a) uric acid, (b) COM, (c) MAPH, and (d) cystine urinary calculi.The absorption spectra of individual calculus types in unit absorbance, referenced from Dao and Daudon [31], are superimposed for comparison.In (e)-(h), the threshold fluences are plotted as a function of absorbance unit for the corresponding calculus types.An experimental fit indicates that the measured threshold fluences are in good agreement with (1).A maximum of 49 data points were recorded with varying laser fluences around the threshold, each with a single pulse.Error bars indicate the ED and ED values obtained from probit analysis for each data set.

Fig. 5 .
Fig.5.Dynamics of ablation on a COM calculus.Within 1 s after laser onset, (a) debris was seen ejected from the surface at laser impact.At the end of the laser pulse (4 s), (b) the ablation plume continues to expand.It is possible that debris screening may have affected the amount of light absorption within the calculus.By 2.5 ms, (h) most ejected particles were seen as out-of-focus debris, implying a much lower density of particles along the FEL beam path.

Fig. 6 .
Fig. 6.Ablation depths (m) versus wavelength (m) for (a) uric acid, (b) COM, and (c) cystine urinary calculi.The absorption spectra of individual calculus types in unit absorbance, referenced from Dao and Daudon [31], are superimposed for comparison.In (d)-(f), the ablation depths are plotted as a function of absorbance unit for the three types of calculi.An experimental fit indicates that the measured depths are in fair agreement with(3).A total of ten data points (n = 10) were collected at each wavelength for each calculus types, with a constant laser fluence of 5 J/cm for a total of 10 pulses at 5 Hz.

Fig. 7 .
Fig. 7. Effects on ablation depth as a function of increasing optical absorption, respectively, at (a) 2.12 m, (b) 2.94 m, and (c) 5.99 m, on an uric acid calculus.

Fig. 8 .
Fig. 8. Examples of surface images of ablation crater of (a) uric acid, (b) COM, and (c) cystine calculi.For each calculus type, the images from top to bottom, 5 ) and a color CCD camera.The figure shows craters ablated at wavelengths with low absorption to high absorption (in relative absorbance unit) from top to bottom, for various urinary calculi.With a 5-Hz pulse repetition rate, no carbonization indicative of thermal damage around the craters is seen.Surface fractures or adjacent mechanical damage are observed surrounding the craters of uric acid [Fig.8(a)(iii)], cystine [(Fig.8(c)(iii)], and MAPH (not shown) calculi at wavelengths with higher relative absorbance.No fracture is apparent surrounding the craters of COM calculi [Fig.8(b)(i-iii)] at all wavelengths tested.