Lasers in Surgery and Medicine 10:322-327 (1990)

Protection of Fiber Function by Para-Axial Fluid Flow in Interstitial Laser

Therapy of Malignant Tumors Kambiz Dowlatshahi, MD, Julee D. Bangert, BS, Michael F. Haklin,

Charles K. Rhodes, PhD, Ronald S. Weinstein, MD, and Steven G. Economou, MD

Department of General Surgery (K. D., J. D.B., M. F. H., S. G. E.) and the Department of Pathology (R. S. W.), Rush-Presbyterian-St. Luke’s Medical Center, Chicago 60612, and

MCR Technology Corporation (C. K. R.), Chicago, Illinois 6061 0

In the past, interstitial laser therapy frequently has failed be- cause of the damage to the bare fiber tip due to intense heat generated at the point of contact. Using a rat mammary tumor model, we describe a method of placing a 600 micron fiber inside a gauge 19 needle cannula after its insertion into the tumor. With this device continuous wave NdYAG laser is delivered to the target tumor while 0.9% saline flows para-axially into the tumor. Significant coagulation necrosis was induced with 500 joules at 5 watts, 100 seconds and 1 cc per minute of saline while the needle- fiber is pulled out of the tumor by 10 mm. The mean transmission loss after 500 joules was 2% in ten experiments. The tumor edema due to 1.5 ml of saline was transient. We conclude that successful hyperthermic coagulation necrosis by NdYAG laser can be achieved with minimal transmission loss by employing the above technique.

Key words: fluid pool, mammary tumor, Nd:YAG laser fiber tip integrity

INTRODUCTION

Non-contact treatment of malignant tumors by laser, predominantly Nd:YAG laser, has been practiced since the late 1970’s and has been ex- tensively reported [ 1-41. Patients with advanced esophageal, bronchial, colorectal, and bladder tu- mors have been successfully palliated with this technique. Many investigators have also tested the possibility of treating more deeply located tu- mors by inserting the laser probe into the lesion. This interstitial method of coagulation by hyper- thermia, however, has been hampered by damage to the tip of the fiberoptic probe and loss of laser energy transmission. The damage is caused by ad- hesion of the heated tissue to the quartz fiber, with charring of this tissue and the absorption of intense energy at the interface resulting in the melting of quartz fiber [5,61. This problem has been partially overcome by: a) reducing the laser power output from the 20-40 watt range to 0.5- 2.0 watts and simultaneously prolonging the ex-

posure time [7,81; b) coating the probe tip with sapphire, which has a higher melting point than quartz and also diffuses the light [9]; or c) insert- ing the fiber in a plastic sheath with circulating coolant fluid, within the sheath and around the fiber tip [10,11].

In this report we describe a simple method of protecting the fiber tip integrity and preserving the laser energy transmission for power output of up to 10 watts by para-axial fluid flow at room temperature.

Address reprint requests to Kambiz Dowlatshahi, M.D., De- partment of General Surgery, Rush-Presbyterian-St. Luke’s Medical Center, 1653 W. Congress Parkway, Chicago, IL 60612. Accepted for publication April 3, 1990.

0 1990 Wiley-Liss, Inc.

Protection of Fiber Function

skin I tumor

323 fluid pump

1

metal cannula : / Iyer fiber j

fluid pool

tumor : - . . - - - - - - - - - - -__________________________---

Fig. 1. Schematic drawing of the laser fiber placed inside a needle-cannula and a “Y”-shaped connector. The thermocouple is glued to the cannula and their tips are 3 mm apart.

MATERIALS AND METHODS

Laser A continuous wave Nd:YAG laser (Trime-

dyne, Calif.) and a 600 micron-diameter quartz fiber with a divergence angle of 8” were employed. The fiber was stripped of its terminal 5 cm of plas- tic coating and this stripped part was inserted into a 5 cm-long, extra-thin gauge 19 needle through a “Y”-shaped connector (Advanced Car- diovascular System). The internal diameter of this needle cannula was large enough to allow for easy placement of the fiber as well as to permit para-axial flow of a cooling fluid such as normal saline (Fig. 1). The position of the quartz fiber in the cannula without its stylet was adjusted so that its tip was flush with the tip of the cannula before tightening the proximal screw of the “Y” connector on the fiber. The second arm of the ‘‘Y” connector was attached to a saline drip whose flow rate was controlled by a Harvard pump. A second identical 5 cm needle was glued to a gauge 22 thermocouple (Yellow Spring Co., Ohio) with its tip positioned 3 mm lateral to and 3 mm in front of the laser fiber tip. Attention was focused on 5 watt power output of the laser at which 10 experiments were performed. Five experiments

were conducted at 2.5, 7.5, and 10 watt power set- tings. The control group consisted of five experi- ments whereby the needles were inserted into the tumor and withdrawn by 10 mm while a saline flow of 1 cc per minute was maintained.

Tumor Virgin, female Sprague-Dawley rats 40 days

of age were housed in groups of two on Ab-Sorb- Dri bedding in a windowless room, illuminated for 14 hours each day, and maintained at 22°C. All animals were allowed free access to food. At 50 days of age, the animals received a single in- travenous injection of 40 mg N-methyl-N-ni- trosurea per kilogram body weight in a volume of 0.5 ml as described previously [121. The appear- ance and progression of tumors in the mammary glands were monitored twice weekly by palpation. Generally, tumors measuring 2 cm in diameter were chosen for investigation (Fig. 2).

TECHNIQUE

The animals were anesthetized with intra- muscular injection of ketamine, 44 mg/kilogram body weight. The hair over the tumor and the surrounding skin was shaved and the tumor di-

324 Dowlatshahi et al.

Fig. 3. Nd:YAG laser-treated Sprague-Dawley rat mammary tumor; 2 mm stepwise sections.

Fig. 2. Typical female Sprague-Dawley rat mammary tumor 2 cm in diameter.

mensions were measured with a caliper. The an- imal was taped to a board and the tumor was im- mobilized with a non-crushing clamp without arresting its blood circulation. Two 1 mm stab incisions were made 3 mm apart at suitable points near the longitudinal axis of the tumor to facilitate the entry of the two glued needles. The needles then were inserted into the tumor, tra- versing its length until the tip of the thermocou- ple needle could be palpated subcutaneously on the opposite pole of the tumor. Care was taken to place the cannula carrying the laser fiber cen- trally and the thermocouple to lie paracentrally. The trocar of the laser needle was then removed. The laser fiber already fixed within the “Y” con- nector was detached from the first cannula and inserted into the second cannula already posi- tioned in the tumor. A few drops of saline were allowed to flow in order to displace the air in the cannula hub before tightening its Luer lock. Fi- nally, the cannula-thermocouple assembly was attached to a modified Sears lathe to regulate the withdrawal of the tumor while the laser energy was being delivered.

At the onset of the experiment, approxi- mately 0.2 ml of saline was allowed to flow in order to wash out any tissue debris or blood clot in the needle cannula as well as to establish a “fluid pool” in front of the fiber tip, thus preventing con- tact between the quartz fiber and the tumor tis- sue. This small fluid pool in front of the laser probe and its maintainence by a constant in-flow rate proportionate to the laser power output are crucial for the preservation of the tip integrity and for subsequent successful transmission of the laser energy. The starting temperature was re-

corded and 500 joules of laser energy was given at 5 watts in 100 seconds while a saline flow of 1 ml per minute was maintained. The withdrawal speed of the lathe was adjusted so that the needle- fiber complex was pulled back by 10 mm during this period. The temperature was recorded contin- uously. At the end of the experiment, the fiber transmission was tested by a power meter (Trim- edyne Co., Calif.). The tip integrity was also in- spected under an operating microscope for loss of shine, dark spots, or fracture.

Forty-eight hours later, the tumor was ex- cised under general anesthesia. The needle entry point was labeled with a suture; the tumor vol- ume was determined by fluid displacement in a marked cylinder and was fixed in 10% formalin. After 24 hours, the tumor was step sectioned at 2 mm intervals (Fig. 3) and the area of necrosis/ coagulation (Fig. 4) was assessed in hematoxylin- and eosin-stained sections (Fig. 5a,b).

RESULTS

Table 1 summarizes the findings. After con- ducting many experiments testing the fluid flow vs. the laser power, it was found that the optimum rate of saline flow to minimize the probe tip dam- age and to prevent excessive edema of the tumor is approximately 1 cc per minute at the 5 watt laser power setting. The comparable rates for 2.5, 7.5, and 10 watts were 0.5, 1.5, and 2 cc per minute, respectively. The laser probe transmis- sion loss remained less than 3% for all these laser power settings and 500 joules. The fiber tip was damaged whenever: a) the saline flow was im- peded or, b) the rate of fiber-needle complex with- drawal was stopped or inadvertently slowed down. In either case the accompanying thermo- couple would warn the impending damage by showing temperatures approaching 50” C.

Protection of Fiber Function 325

Fig. 4. Cross section view of one block from Figure 3 showing the NdYAG laser lesion in its center (arrows).

Tumor Temperature At 5 watt laser power setting the tumor tem-

perature recorded 3 mm away from the laser tip rose from an average base recording of 31°C to a peak of 41°C with a mean value of 8°C for 10 ex- periments. Higher temperatures were recorded for higher laser power settings, despite the in- crease in the rate of saline flow and the probe withdrawal speed resulting in shorter irradiation time. Similarly, with a lower power setting of 2.5 watts, the mean temperature rise was 15°C. A direct relationship was observed between the rise in the tumor temperature and the increase in the laser power.

Lesion Size A typical histological appearance of the co-

agulated tumor lesion is shown in Figure 4. Exact measurements of the lesion were difficult both by naked eye and under microscope because in most experiments, islands of coagulation necrosis were found outside the main area of laser necrosis. The diameter of the lesions varied between 5 and 8 mm and the calculated volume of the entire lesion for 500 joules was 200-250 cubic mm. In the con- trol group, the volume of necrotic tissue, either due to needle trauma or spontaneous necrosis, was estimated to be less than 10 cubic mm.

DISCUSSION

Treatment of solid tumors deeply located in various organs by Nd:YAG laser applied through

a needle is an attractive concept. This idea follows recent advances in precise localization and needle diagnosis of such lesions under radiographic, sonographic, and computerized tomography [131. The mechanism of the target injury is focal hy- perthermia and coagulative necrosis. The princi- pal type of laser employed is neodymenium: yttrium-aluminum-garnet (Nd:YAG) laser. In- sertion of a bare-tip quartz fiber directly into the lesion, either alone or through a needle, for deliv- ery of Nd:YAG laser energy, has failed in the past because the generated heat at the tissue-fiber in- terface damages the fiber tip before any effective coagulation takes place [5,61. Previous investiga- tors have partially succeeded in overcoming this obstacle by various methods.

Matthewson et al. employed Nd:YAG laser at low power (0.5-2.0 watts) and long exposures (up to 40 minutes) and delivered the energy through a 400 micron-diameter optical fiber into the rat liver [14]. Necrotic lesions up to 16 mm in diameter were generated, with temperatures reaching 100°C near the fiber point. These inves- tigators noted tissue charring around the fiber tip at power settings between 0.75 watt and 2.0 watts, with corresponding fall in the light trans- mission from 90% to 27%. We noted similar char- ring of the terminal 1-2 mm of the fiber and dam- age to its tip when no fluid flowed by the fiber into the tissue. Carbonization of the tumor tissue in contact with the fiber was also observed on sub- sequent histologic examination of the tissue. This phenomenon was related to the loss of power transmission as measured on power meter at the completion of each experiment.

Daikuzono et al. introduced a synthetic sap- phire probe giving a wider angle of irradiation with diffusion of low power (<5 W) laser radiation for interstitial application [8,91. The temperature of the tissue surrounding the laser probe was con- tinuously monitored by thermocouples inserted into the tissue using a n on-off computer control system to achieve a stable range of 42-44°C. The local hyperthermia caused tissue coagulation ne- crosis of 1-2 cm diameter after 20-40 minutes of treatment. They recommended multiple probe in- sertions into tumors for effective treatment within a reasonable period of time.

Gatenby et al. inserted a 400 micron fiber with its terminal 2.25 cm cylindrical diffuser in- side a 5-F clear Teflon sheath catheter into mouse tumors 181. In their experiment, they used he- matoporphyrin derivative as tissue sensitizer. The type of laser employed was pumped argon-

326 Dowlatshahi et al.

Fig. 5. a: Histopathology of laser-induced necrosis in a Spra- gue-Dawley mammary tumor. Viable tumor (VT) left, coag- ulation necrosis (CN) right. Note the sharp demarcation be-

tween the two zones. Magnification x 250. b: Close-up view of 5a. Viable tumor (VT) left, coagulation necrosis (CN) right. Magnification x 500.

TABLE 1. Comparison of Rate of Saline Flow to Amount of Laser Energv ~~ ~

Saline Power Pulse Laser Temp. Transmission volume setting duration energy rise loss (cc/min) (watts) (seconds) (joules) (" C) (%)

0.5 2.5 200 500 1.5 2 0.0" 2.5 200 500 6.0 7 1.0 5.0 100 500 8 2.2 0.0" 5.0 100 500 11 7 1.5 7.5 75 500 10 1 2.0 10.0 50 500 13 1.5

"Control groups (mean of 5 observations) with no para-axial flow.

dye with wave length >600 nm and maximum power output of 1.2 watts. An average of 900 joules was given with a mean temperature rise in the tumor of 9.O"C. There was some tumor re- sponse to hyperthermia alone although the main effect was observed in conjunction with the use of hematoporphyrin derivative. Nonetheless, from the technical point of view the Teflon sheath ap- peared to prevent blood clot adherence to the fiber and charring of the probe.

Godlewski et al. employed a water-cooled 600 micron-diameter probe and Nd:YAG laser in- serted into the piglet liver to induce hyperthermic lesions [ill. High-power settings (80 watts) raised the tissue temperature to over 400"C, causing tis- sue vaporization and damage to the protective sheath. The size of the liver lesions ranged be- tween 1 and 2 cm in diameter. The external di- ameter of the probe was 5 mm and it was intro- duced through a 6 mm trocar into the pig's liver.

The method described in this report is both simple and effective. The fluid prevents the tissue from coming into direct contact with the fiber tip

to cause a "frying pan" effect; thus allowing trans- mission of the laser energy into the target tissue. Our approach is somewhat analogous to that in the urinary bladder, where the organ is filled with water before the laser treatment of the blad- der tumor. In our experience, the damage to the fiber tip was minimal provided an adequate fluid flow was maintained during the laser beam de- livery and the temperature in the immediate vi- cinity of the tip (3 mm) was kept below 50°C. The optimal rate of saline flow corresponding to the laser power setting is shown in Table 1. Minimal transient edema of the tumor occurred with each experiment.

We observed that needle withdrawal during the laser delivery was advantageous in two ways. First, a cooler zone of the tumor was constantly entered. Thus, a smaller volume of fluid was needed to cool the fiber. Second, successive re- gions of tumor were lased, producing a coagulated column of necrotic tissue instead of a sphere of coagulation which would have been the case if the fiber had been kept stationary.

Protection of Fiber Function 327 The power settings of up to 10 watts enable

the operator to coagulate significant volumes of tumor tissue: for example, 1 cc within a short time (less than 2 minutes). This is an important prac- tical consideration when one plans to treat meta- static tumors measuring 2-5 cm in diameter in solid organs such as the liver. It may be possible to treat such tumors by using a single needle and reinserting it at several points of the tumor to coagulate its entire volume under ultrasound con- trol. Real-time sonographic and thermographic monitoring are essential for satisfactory ablation of all parts of the tumor. This procedure may be performed either percutaneously or under direct vision at laparotomy. An additional advantage of this technique is the use of the same fine diagnos- tic needle (cannula + trocar) for introduction of the fiberoptic probe and laser energy into the tar- get tumor. A gauge 19 ultrathin needle cannula of 10-15 cm length may be employed in conjunction with 600 micron-diameter quartz fiber. Thus, deeply positioned tumors may be reached with minimal trauma. Previous investigators em- ployed trocars or plastic sheaths with external di- ameters ranging between 2 and 6 mm, rendering the target entry traumatic and the procedure haz- ardous.

REFERENCES

Fleischer D, Sivak MV: Palliative Nd:YAG laser therapy as palliative treatment for advanced adenocarcinoma of the gastric cardia. Gastroenterology 87:815-820, 1984. Swain CP, Bown SG, Edwards DAW, Kirkham JS, Salmon PR, Clark CG: Laser recanalization of obstruct- ing foregut cancer. Br J Surg 71:112-115, 1984. Bown SG, Barr H, Matthewson K, Hawes R, Swain CP, Clark CG, Boulos PB: Endoscopic treatment of inoperable colorectal cancers with the Nd:YAG laser. Br J Surg 73: 949-952, 1986.

4.

5.

6.

7.

8.

9.

LO.

11.

12.

13.

14.

Hetzl MR, Nixon C, Edmondstone WM, Mitchell DM, Millard FJC, Nanson EM, Woodcock AA, Bridges CE, Humberstone AM: Laser therapy in 100 tracheobron- chial tumors. Thorax 40:341-345, 1985. Bown SG: Tumor therapy with the Nd:YAG laser. In Jo- effe SN, Muckerheide MC, Goldman L (eds): “Neody- mium:YAG Laser in Medicine and Surgery.” New York: Elsevier Science Publishers, Inc., 1983, pp 51-59. Svaasand LO: Optical dosimetry for direct and intersti- tial photoradiation therapy of malignant tumors. In Doiron DR, Gomer CJ (eds): Porphyrin Localization and Treatment of Tumors. New York: Liss, 1984, pp 91-114. Daikuzono N, Suzuki S , Hisao T, Hiroshi T, Masaru 0, Joffe SN: Laserthermia: a new computer-controlled con- tact Nd:YAG system for interstitial local hyperthermia. Lasers Surg Med 8:254-258, 1988. Gatenby RA, Hammond ND, Brown DQ: Tumor therapy with hematoporphyrin derivative and lasers via a percu- taneous fiberoptic technique: preclinical experiments. Radiology 163:167-171, 1987. Daikuzono N, Joffe SN: Artificial sapphire probe for con- tact photo-coagulation and tissue vaporization with the Nd:YAG laser. Med Instrum 19:173-178, 1985. Gatenby RA, Haatz WH, Engstrom PF, Rosenblum JR, Hammond ND, Kessler HB, Moldofsky PJ, Clair MR, Un- ger E, Broda GJ: CT guided laser therapy in resistant human tumors: phase I clinical trial. Radiology 163:172- 175, 1987. Godlewski G, Rouy S, Pignodel C, Ould-Said H, Eledjam JJ, Bourgeois JM, Sambuc P: Deep localized neodymium (ND):YAG laser photocoagulation in liver using a new water cooled and echoguided handpiece. Lasers Surg Med

McCormick DL, Adamowski CB, Fiks A, Moon RC: Life- time dose-response relationships for mammary tumor in- duction by a single administration of N-methyl-N-nitro- surea. Cancer Res 41:1690-1694, 1981. Hashimoto D, Takami M, Idezuki Y: In depth radiation therapy by YAG laser for malignant tumors in the liver under ultrasonic imaging (abstr). Gastroenterology 88: 1663,1985. Matthewson K, Coleridge-Smith P, O’Sullivan JP, North- field TC, Bown SG: Biological effect of intrahepatic neodymium: yttrium-aluminum-garnet laser photocoagu- lation in rats. Gastroenterology 93:550-557, 1987.

8:501-509, 1988.