2.1.

Thulium Fiber Laser Parameters

A 100-W experimental TFL (TLR 110-1908, IPG Photonics, Oxford, Massachusetts) with a center wavelength of 1908 nm was used in these studies. This wavelength was chosen to closely match a high-temperature water absorption peak in tissue in the infrared spectrum.^8–10 The continuous-wave TFL was electronically modulated using a function generator (DS345, Stanford Research Systems, Sunnyvale, California) to operate in long-pulse mode, producing $500\text{-}\mu \mathrm{s}$ pulses for lithotripsy studies, similar to conventional holmium laser pulse lengths of 350 to $700\mu \mathrm{s}$. The TFL produced an approximately Gaussian, near single-mode beam profile, originating from an $18\text{-}\mu \mathrm{m}$-core thulium-doped silica fiber, with a built-in collimator providing a 5.0-mm-diameter output beam. A 25-mm-focal-length calcium fluoride lens was used to focus the TFL beam down to a $1/e^2$ spot diameter of $\sim 25\mu \mathrm{m}$ for coupling into a standard, disposable, low-hydroxyl, $100\text{-}\mu \mathrm{m}$-core silica optical fiber (AFS105/125, Thorlabs, Newton, New Jersey). All stone ablation experiments were performed with a TFL output pulse energy of 35 mJ, pulse duration of $500\mu \mathrm{s}$, and variable pulse rates of 150, 300, or 500 Hz.

2.2.

Holmium:YAG Laser Parameters

A 20-W, clinical holmium:YAG laser (Medilas H20, Dornier MedTech, Wessling, Germany) with a center wavelength of 2100 nm was used in these studies for direct comparison with the TFL. The holmium laser was used with a standard $270\text{-}\mu \mathrm{m}$-core clinical, low-OH silica optical fiber (RFID Holmium Lightguide, Dornier MedTech). The laser was operated with standard clinical parameters, including a pulse energy of 600 mJ, pulse duration of $350\mu \mathrm{s}$, and pulse rate of 6 Hz. Laser pulse energy was measured using a pyroelectric detector (ED-200, Gentec, Canada) connected to an energy/power meter (EPM1000, Molectron, Portland, Oregon).

2.3.

Urinary Stone Samples

All stone samples were composed of 60% calcium oxalate monohydrate and 40% calcium phosphate (Fig. 1). These stones were chosen because calcium oxalate stones are common and comprise about 80% of all stone compositions encountered in the clinic.¹¹ All of the stone samples originated from a single patient, had a consistent mass (40 to 100 mg) and size (4 to 5-mm diameter) and were available in large quantities. Stone samples were desiccated in an oven for 15 min and then weighed with an analytical balance (Model AB54-S, Mettler-Toledo, Columbus, Ohio) before lithotripsy experiments to determine their initial mass. Stone samples were then placed in the ureter model, immersed in a saline bath and experiments were conducted immediately upon rehydration. A total of 12 stone samples were used for each set of laser parameters.

Fig. 1

Human urinary stone samples with composition of 60% calcium oxalate monohydrate and 40% calcium phosphate used in all of these studies.

2.4.

Experimental Setup

A 6-mm-inner-diameter tube with an integrated 1.5-mm mesh sieve and microthermocouple was used as a simple in vitro ureter model in these studies (Fig. 2). Kidney stones were placed inside of the tube and rested on the sieve, with the entire ureter model submerged in a saline bath. The distal tip of a flexible digital video-ureteroscope (URF-V, Olympus, Southborough, Massachusetts) was then placed inside the tube. The optical fiber was inserted through the 3.6 Fr (1.2 mm) single working channel of the ureteroscope and positioned in contact with the urinary stone sample under magnification. Constant saline irrigation at room temperature (22°C) was provided by a saline bag elevated 100 cm above the experimental setup. Saline flow rates through the ureteroscope working channel with the $100\text{-}\mu \mathrm{m}$-core [$244\text{-}\mu \mathrm{m}$-outer diameter (OD)] TFL and $270\text{-}\mu \mathrm{m}$-core ($464\text{-}\mu \mathrm{m}$-OD) holmium laser fibers measured 22.7 and $13.5\mathrm{ml}/\min$, respectively. It should be noted that there is always continuous saline irrigation during endoscopic laser lithotripsy in the clinic, primarily to clear stone dust and maintain visibility of the stone in the surgical field during the procedure. All stone samples were free to move around inside the ureter model during the studies. An experienced, practicing urologist (P.I.) performed all of the laser lithotripsy experiments. The stone sample was irradiated with the laser until all ablated fragments were sufficiently small ($<1.5\text{-}\mathrm{mm}$ diameter) to pass through the mesh of the sieve. Total laser times and operation times were recorded for each experiment. “Laser time” was defined as the total time that the laser was on. “Operation time” was defined as the total time from laser initiation until the final stone fragments passed through the sieve, including both laser time and any additional time used to re-adjust the fiber position or clear the visual field when the laser was momentarily not in use. Videos of each stone ablation experiment were also recorded for subsequent analysis.

Fig. 2

(a) Experimental setup showing the ureter model, including 6-mm-inner-diameter tube, 1.5-mm mesh sieve, and microthermocouple. (b) Close-up view through the flexible ureteroscope of the fiber tip and stone sample during the experiment. The thermocouple was placed 3 mm from the center of the tube (along the tube wall) and 1 mm above the mesh sieve.

2.5.

Thermocouple Temperature Measurements

Temperature monitoring was performed during all experiments. An insulated, $125\text{-}\mu \mathrm{m}$-diameter, microthermocouple (Type T, Omega, Stamford, Connecticut) was positioned 1 mm above the sieve mesh and 3 mm from the center of the tube (at the tube wall), to monitor and record saline temperatures in close vicinity to the stone sample during laser ablation (Fig. 2). A digital data acquisition system (OM-USB-TC, Omega) connected to the microthermocouple and controlled by a laptop personal computer was used to record all temperature data. The peak temperature from each individual study ($n=12$) of a given data set was recorded, and then the mean and standard deviation of all of these peak temperatures was calculated and are provided in Table 1. Great care was taken to prevent directly irradiating the thermocouple with laser energy from the fiber optic tip so as to avoid both potential damage to the thermocouple and erroneous temperature values from direct absorption of laser energy by the thermocouple.

Table 1

Total laser and operating times to fragment 4- to 5-mm diameter calcium oxalate stones into fragments smaller than 1.5 mm, and peak saline temperatures measured during experiments.

3.1.

Laser and Operation Times

Mean laser and operation times during holmium:YAG and TFL lithotripsy experiments are summarized in Table 1. The laser was periodically turned off to allow re-positioning of the fiber and stone debris clearance for improved visibility during the procedure, which resulted in the reported differences between laser and operation times. The initial stone mass was similar for all data sets and measured $59\pm 4\mathrm{mg}$ for the holmium laser data set, and $55\pm 15$, $60\pm 15$, and $66\pm 10\mathrm{mg}$ for the TFL pulse rates of 150, 300, and 500 Hz, respectively. Holmium laser times measured $167\pm 41\mathrm{s}$. TFL times measured $111\pm 49$, $39\pm 11$, and $23\pm 4\mathrm{s}$, for pulse rates of 150, 300, and 500 Hz, respectively. As the TFL pulse rate was increased from 150 to 500 Hz, laser times decreased since the laser pulses were more rapidly delivered to the stone. Operation times were also recorded. Holmium laser operation time measured $207\pm 50\mathrm{s}$. TFL operation times measured $116\pm 54$, $54\pm 22$, and $60\pm 22\mathrm{s}$ for pulse rates of 150, 300, and 500 Hz.

Both stone laser times and total operation times were significantly shorter for the TFL at all pulse rates (150, 300, and 500 Hz) than for the holmium laser ($P<0.05$). However, there was no statistical difference between the 300 and 500-Hz TFL operation times ($P=0.37$). These overall findings were due to a number of factors, including TFL operation at higher power densities, higher pulse rates, higher average powers, and reduced stone retropulsion, as described below.

The two lasers produced two different stone motion effects. The TFL with a lower pulse energy and higher pulse rate produced a vibrational effect, which led to the stone oscillating in the same relative position. The holmium laser exhibited a retropulsion effect, causing the stone to recoil within the confines of the ureter model, and making it more difficult to ablate the stone in an efficient manner, which was reflected by longer holmium laser and operation times provided in Table 1. It should be noted, however, that the presence of the sieve may have distorted to some degree the normal retropulsion movement typically encountered in the clinic.

3.2.

Saline Temperatures

Mean peak saline temperatures (defined as average of $n=12$ individual peak temperatures for each data set) during holmium:YAG and TFL lithotripsy experiments were calculated to be $24\pm 1°\mathrm{C}$ for holmium, and $33\pm 3°\mathrm{C}$, $33\pm 7°\mathrm{C}$, and $39\pm 6°\mathrm{C}$ for the TFL at pulse rates of 150, 300, and 500 Hz, respectively (Table 1). Temperatures during TFL lithotripsy were significantly higher than for the holmium laser ($P<0.05$). The results also suggest that a decrease in the TFL pulse rate from 500 to 300 Hz also translated into a lower average power and lower saline temperatures ($P=0.01$), and thus provides an additional safety margin for potential future clinical studies.

Temperature history graphs showing the worst case (highest peak temperature) during holmium and TFL lithotripsy experiments are provided in Fig. 3. The variation in temperatures in all of the graphs is due to a number of factors, including the laser parameters, experimental setup, and surgical technique. Specifically, variable laser pulse rates were used translating into different cooling rates in between delivery of individual laser pulses. The ureteroscope was handheld and the distance between the fiber optic tip and thermocouple location also varied constantly, affecting temperature readings as well.

Fig. 3

Temperature versus time plots showing worst case (highest peak temperature) for holmium:YAG and thulium fiber laser (TFL) lithotripsy procedures. (a) Holmium laser, $T_p=25.5°\mathrm{C}$; $\text{laser time}=288\mathrm{s}$; $\text{total time}=320\mathrm{s}$; (b) TFL at 150 Hz; $T_p=38°\mathrm{C}$; $\text{laser time}=154\mathrm{s}$; $\text{total time}=154\mathrm{s}$; (c) TFL at 300 Hz; $T_p=48°\mathrm{C}$; $\text{laser time}=43\mathrm{s}$; $\text{total time}=47\mathrm{s}$; (d) TFL at 500 Hz; $T_p=48°\mathrm{C}$; $\text{laser time}=22\mathrm{s}$; $\text{total time}=107\mathrm{s}$. Note that the variation in temperatures measured was a function of not the only laser parameters used but also of the variable movement of the fiber tip and urinary stone sample with respect to the thermocouple position.

The two laser systems produced different saline temperature profiles near the stone sample. During holmium laser lithotripsy, a small overall elevation in the saline temperature was observed, averaging only a few degrees Celsius by the end of the procedure. The highest temperature recorded for any of the stone samples was only about 26°C [Fig. 3(a)]. This minimal temperature rise may be explained by the low pulse rate (6 Hz) and low duty cycle ($1:167$) with sufficient time for saline cooling in between holmium laser pulses.

On the contrary, during TFL stone ablation there was a rapid and substantial increase in saline temperature, presumably due to the higher duty cycles of $1:10$, $1:5$, and $1:3$ for the TFL pulse rates of 150, 300, and 500-Hz operation, respectively, and correspondingly lower cooling times in between laser pulses. The highest temperature measured for any stone sample was 48°C [Figs. 3(c) and 3(d)]. However, such temperatures were maintained for a short time, typically less than 1 s [Fig. 3(d)] and less than 4-s total [Fig. 3(c)]. Such temperatures corresponded to the time when large stone debris or chips were obstructing the sieve, reducing irrigation rates, and temporarily resulting in a thermal buildup in the saline (Fig. 4). As expected, mean peak saline temperatures decreased as the TFL pulse rate was decreased from 500 to 300 Hz, due to lower average power and duty cycle, and longer cooling times in between laser pulses.

Fig. 4

(a) Stone fragments after TFL lithotripsy. Inset figure shows the original stone at the same scale. (b) Video frame showing multiple large stone chips temporarily obstructing the sieve and saline irrigation flow at the time point in which the absolute peak temperature of 48°C was measured by the microthermocouple during TFL lithotripsy at 500 Hz [Fig. 3(d)].

It is of interest that the TFL operation at 300-Hz led to a significant decrease in saline irrigation temperatures without decreasing overall operation times, in comparison with 500 Hz. It may be that although TFL operation at 500-Hz results in shorter laser irradiation times, the operating time is not reduced due to both greater stone movement and the need to more frequently pause during the procedure and momentarily turn the laser off for improved fiber positioning and visibility.

It should be emphasized that the saline temperatures recorded during TFL lithotripsy may be of potential concern during a clinical procedure, because a prolonged and excessive temperature rise, although not observed here, could possibly cause undesirable collateral thermal damage to surrounding soft urinary tissues such as the ureter or kidney, if left unchecked. However, to put these specific results into proper perspective, it should be noted that even the highest temperatures reached (48°C) only briefly ($<4\mathrm{s}$), and are safe based on standard Arrhenius integral calculations for thermal damage to soft urological tissues using published values,^12–14 as discussed in more detail below.