Investigative Ophthalmology & Visual Science, Vol. 29, No. 4, April 1988 Copyright © Association for Research in Vision and Ophthalmology

The Retinal Effects of Vapor Exposure Shimon Gabay.t Israel Kremer,* Isaac Ben-Sira,* Dan Sagie,t Dov Weinberger,* and Gideon Erezf

The is a pulsed which emits energy in two wavelengths simultaneously: 510.6 nm () and 578.2 nm (yellow). Each pulse has a duration of 15 nsec, maximal energy of 3 mJ and a peak power of more than 100 kW. It is a variably high repetition rate laser, in the range between 1 kHz and more than 20 kHz. We studied its interaction with the rabbit , while using two different repetition rates, 4 kHz and 18 kHz. The histological analysis of the lesion produced by 4 kHz repetition rate showed undesired retinal effects, similar to those caused by other pulsed . On the other hand, the histological examination of the lesion produced by the 18 kHz repetition rate showed a desired coagulation effect, limited to the outer retinal layers, and comparable to a lesion produced by a (CW) laser. Invest Ophthalmol Vis Sci 29:528-533, 1988

Lasers in ophthalmology are used mainly for reti- above 20 kHz, delivering two wavelengths: 510.6 nm nal coagulation, iridectomy and lens capsulotomy. and 578.2 nm.10 The pulse peak power can be in- Argon laser is now widely in use as a retinal photo- creased by decreasing the repetition rate, while the coagulator,12 as it is well-transmitted by the ocular average power has an optimal level at intermediate media3 and well-absorbed by the retinal pigment epi- repetition rate. As irradiance is a function of the laser thelium (RPE) and hemoglobin.4 Since coagulation is focal spot size, the mechanical effects of the CVL can the final result of a thermal effect,5 a continuous wave be increased by decreasing the focal spot size to 50 (CW) laser, such as the argon laser, is a suitable light j^m. Therefore, it is hoped that the CVL will be suit- source for coagulation. Capsulotomy and membrane able for iridectomy as well as capsulotomy, a fact cutting can be done only by a pulsed laser character- which will have to be proven in future experiments. ized by a high peak power, inducing mechanical ef- By increasing the repetition rate and the spot size, the fects.6"8 pulse characteristics are changed toward CW mode, The question is whether one laser system can per- which is suitable for retinal coagulation without any form all the above mentioned ocular treatments suc- undesired mechanical damage. cessfully. A CW laser which is effective for retinal Retinal coagulation is caused by a thermal effect,5 coagulation cannot be suitable for capsulotomy, which depends on the temperature rise in the tissue. where there is a need for very high irradiance. On the The temperature rise depends not only on the energy other hand, a pulsed laser which fits the capsulotomy absorbed by the RPE but also on the exposure time. requirements may cause undesired retinal damage by The CVL pulse width is only 15 nsec; therefore, in mechanical effects. The answer should therefore be a such a short time, the volume of absorbing tissue can variably high repetition rate pulsed laser, which in- be regarded as an isolated volume from which the herently has the characteristics of a pulsed laser, but heat diffusion is negligible. Hence, exposure of the can at the same time act as a quasi-CW laser, de- retina to the CVL beam can instantaneously elevate pending on the pulse energy, the time interval be- the tissue temperature to a higher degree than can a tween the pulses and the reaction time constant. Such CW laser, while delivering the same energy.1011 a laser system is the copper vapor laser (CVL). This Moreover, since the retina was exposed to a burst of laser is a high repetition rate pulsed laser9 which can several hundreds of pulses, the net result depends on operate at a repetition rate range between 1 kHz and the combination between the single pulse energy and the total exposing energy."12 This total energy can be equally delivered either by a high pulse energy com- bined with a low repetition rate or by a low pulse From the *Department of Ophthalmology and Laboratory of Ophthalmic Pathology, Beilinson Medical Center, Petah Tiqva, Tel energy combined with a high repetition rate. Aviv University Sackler School of Medicine, and the fNudear As the CVL is inherently a pulsed laser with a vari- Research Center, Beer Sheva, Israel. ably high peak power, we decided to study its retinal Presented at the annual ARVO Meeting, Sarasota, Florida, May coagulation ability first in order to look for any un- 4-8, 1987. desired mechanical side effects. In this work, we ex- Submitted for publication: June 15, 1987; accepted October 27, 1987. amined the interaction between the copper vapor Reprint requests: I. Kremer, MD, Department of Ophthalmol- laser beam and the rabbit retina, in order to learn the ogy, Beilinson Medical Center, Petah Tiqva 49 100, Israel. safest range of peak power, pulse repetition rate and

528

Downloaded from iovs.arvojournals.org on 09/26/2021 No. 4 RETINAL EFFECTS OF COPPER VAPOR LASER / Goboy er ol. 529

Fig. 1. A diagram illustrating the system components which control the laser power, exposure time and beam spot size.

total exposure energy which will be suitable for reti- exposure time and beam spot size. The laser beam is nal coagulation. delivered from the laser cavity (1-3) towards the rab- bit eye, passing through an attenuator (4), which Materials and Methods controls the output power, then meets a narrowing telescope (5, 6) and falls upon a shutter (7). The shut- Copper Vapor Laser ter reflects more than 99% of the light to a power The copper vapor laser used in our experiments is meter (8), which sends a signal to a microprocessor an air-cooled laser which was designed and con- that presents the laser power measured at the power structed by Laser-On (affiliate of Rotem Industries, meter (or the calibrated power at the cornea). The Nuclear Research Center, Beer Sheva, Israel). This is remaining power is used as an aiming beam, which a self-terminating pulsed laser, the emission of which passes through dual polarizing sheets to control the is simultaneously composed of two wavelengths, aiming beam intensity. Opening the shutter for the 510.6 nm (green) and 578.2 nm (yellow). This laser desired exposure time enables full power delivery to emits up to 8 W average power at 4 kHz repetition the retina. The beam then passes through an articu- rate, at a ratio of 2:1 between green and yellow, re- lated arm (9) connected to a Haag Streit (Bern, Swit- spectively. The laser pulse width is about 15 nsec and zerland) Universal Slit Lamp, model BM 900(12). A the beam diameter is 30 mm. An unstable resonator movable focussing lens (10) is placed in such a way was constructed to decrease the beam divergence in that in its backward position the laser beam focusing order to achieve a small focused spot size. The pulse plane coincides with the slit lamp illumination focus- repetition rate.can be varied between 2 kHz and 20 ing plane. In this position, the laser spot on retina is kHz. The two laser wavelengths are well-transmitted minimal. Increasing the laser spot on the retina can by the ocular media and well-absorbed by the RPE be achieved by shifting the lens toward the slit lamp. and hemoglobin, but not absorbed by the xanto- The beam is then reflected by a prism (11) along the phyll.13 Therefore, this laser can be safely used at the slit lamp axis, then passes through the slit lamp out- macular region. Moreover, the yellow wavelength put lens (13) and finally is reflected again by a mirror which is suggested to be the ideal wavelength for reti- (14) towards the retina. The stereo biomicroscope 1415 nal photocoagulation gives this laser an advantage (15) is used to fix the slit lamp focal plane on the over other lasers currently in use. Finally, a compact retina. The filter (16) is positioned in front of the copper vapor laser of about 8 W average power has biomicroscope, prior to full power laser exposure, in low installation requirements of about 2.5 kW elec- order to prevent damage to the physician's eyes by a trical power (single phase) and 2 liter per minute tap scattered laser light. water cooling or even air cooling. The latter charac- All the optical components of the system are teristics make it convenient for installation even in achromates, in order to focus the two wavelengths small eye clinics. (green and yellow) to one small focal point on the retina. The laser control system consists of a micro- Delivery System processor which automatically controls the laser operation, governs the treatment parameters and The diagram illustrated in Figure 1 presents all the takes care of the safety requirements during the ex- system components which govern the laser power, posure process.

Downloaded from iovs.arvojournals.org on 09/26/2021 530 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / April 1988 Vol. 29

^.v^V • • • • y * *~ %

^w.^.-1

Fig. 2. Complete disruption of photoreceptor outer segments, Fig. 4. Exudative choroidal detachment (arrow) accompanied by severe vacuolar degeneration of RPE cells and a small choroidal subretinal and intraretinal exudation. Severe disruption of the pho- hemorrhage (arrow). Methylene , original magnification X400. toreceptors with separation between them and the rest of the retina. Laser parameters: 0.1 W, 0.1 sec, 4 kHz. On the left, big vacuoles are seen between the photoreceptor outer segments and Bruch's membrane, while the RPE cells are missing. Methylene blue, original magnification X400. Laser parameters: 0.1 W, 0.1 sec, 4 kHz. Methods All our experiments were performed on pigmented experimental program. The rabbits were sacrificed rabbits which were anesthetized by chlorpromazine, and their eyes enucleated 10 to 12 hr following the 5 mg/kg, together with ketalar, 50 mg/kg, given intra- last laser exposure. The enucleated eyes were slit muscularly. A series of 10 to 15 lesions were made in along the limbus and fixed in 2.5% buffered glutaral- three different areas in the retina of rabbit eye, sur- dehyde for 10 min. Thereafter, gradual removal of rounded by four intense demarcation spots. Each the cornea and iris was performed, accompanied by treatment retinal area was devoted to different laser refixation in the same fixative.Finally , the whole cili- parameters. The location of each treatment area was ary body and lens were removed and the rest of the marked on a corresponding diagram of the rabbit eye eye was fixated in the buffered glutaraldehyde for an- fundus. The laser beam was focused to a spot diame- other 24 hr. The three treated areas were subse- ter of 200 tirci on the retina, which remained constant quently identified and cut from each eye. Each piece during all the experiments. The other laser parame- of tissue was serially sectioned, Epon-embedded fol- ters, such as pulse energy (or peak power), pulse repe- lowing osmic acid (1%) fixation and then cut thin tition rate and exposure time were changed to fit the (less than 1 ^m) as for transmission electron micros- copy and finally stained by methylene blue stain. Two major groups of experiments were performed, one in repetition rate of 4 kHz, where the peak power range was 400-4000 W, and the other in 18 kHz where the peak power range was 50-300 W. These experiments adhered to the ARVO Resolu- tion on the Use of Animals in Research.

Results The histological analysis of the lesion performed by 4 kHz repetition rate with pulse energy greater than 20 jitJ, showed undesired side effects, such as exuda- tive choroidal, pigment epithelial and retinal detach- ment, disruption of the photoreceptor outer segments and RPE cells together with vacuolar changes in the Fig. 3, Complete disruption of photoreceptor outer segments and latter, which can be related to mechanical (nonther- separation between the latter and their nuclei (arrow head). mal) effects. Mild choroidal hemorrhage was also Disruption of some RPE cells with RPE detachment (arrowhead). Methylene blue, original magnification X400. Laser parameters: noted in the area of the burns. The severity of the 0.1 W, 0.1 sec, 4 kHz. tissue damage decreased gradually as the pulse energy

Downloaded from iovs.arvojournals.org on 09/26/2021 No. 4 RETINAL EFFECTS OF COPPER VAPOR LASER / Goboy er ol. 531

Fig. 6. Partial coagulation of the photoreceptor outer segments (arrow) with vacuolar degeneration of the inner segments, thermal damage to the photoreceptor nuclei (karyopyknosis). Disruption and displacement of the RPE cells are also noted. Methylene blue, original magnification X400. Laser parameters: 0.05 W, 0.1 sec, 4 kHz.

At this point, we concluded that 400 W peak power is the upper limit of "nondisruption range," while using the same spot size of 200 ^m. As no coagulation was seen, it was realized that the total exposing energy (in a relevant exposure time) had to be increased. This was achieved by increasing the repetition rate to 18 kHz, following which the peak power was de- Fig. 5. Severe damage to the photoreceptor nuclei (karyorrhexis creased by 80%, as compared with the 4 kHz mode, and karyopyknosis) with outer segments disruption. The RPE cells while the average power remained almost the same. show vacuolization and displacement from Bruch's membrane. Methylene blue, original magnification X400. Laser parameters: The histological finding of the lesions formed by. 18 0.1 W, 0.1 sec, 4 kHz. kHz repetition rate are presented in Figures 7 to 10. A nice coagulation effect was found when the laser pulse energy was changed between 2 jJ and 5 jxi (40 was lowered. Figures 2 to 5 present the histological mW and 100 mW average power), while the maximal appearance of the typical lesion caused by 4 kHz rep- exposure time was limited to 100 msec. In Figure 7, it etition rate, while the representative parameters are 25 jxJ pulse energy (1600 W peak power or irradiance of 5 MW/cm2) and total exposure energy of 10 mJ. The striking finding is severe disruption of the photo- receptor outer segments and the RPE cells in the center of the lesion (Figs, 2, 3), and the accumulation of intraretinal and subretinal fluid (Fig. 4). Figure 2 also shows evidence of mild choroidal hemorrhage. In addition, severe vacuolar degeneration, swelling and displacement of the RPE cells are also seen to- gether with piknosis and kariorrhexis of the photore- ceptor nuclei (Fig. 5). Extensive RPE and choroidal detachment were also noted (Figs. 3, 4). In order to reduce the retinal disruption and the other mechani- cal effects, we gradually decreased the pulse peak power by neutral density filters down to 200 W. While decreasing the peak power down to 400 W, it Fig. 7. The coagulation effect (arrow) is limited to the photore- ceptor outer segments and RPE cells only. No changes are seen in was found that the lower the peak power the less the the photoreceptor inner segments nor in their nuclei. Methylene disruption effect (see Fig. 6). Below 400 W we have blue, original magnification X400. Laser parameters: 70 mW. 0.1 not seenany sign of disruption nor coagulation. sec. 18 kHz.

Downloaded from iovs.arvojournals.org on 09/26/2021 532 INVESTIGATIVE OPHTHALMOLOGY G VI5UAL SCIENCE / April 1988 Vol. 29

Fig. 8. Coagulation of the photoreceptor outer segments and Fig. 10. Disruptive changes are seen in the RPE together with RPE cells. A thermal damage is also noticed in the photoreceptor displacement of RPE cells from Bruch's membrane. Additionally, a nuclei with some internal displacement of this layer and the other thermal damage is noticed in the photoreceptor outer segments and inner retinal layers (arrow). Methylene blue, original magnification nuclei (arrow). Methylene blue, original magnification X400. Laser X400. Laser parameters; 100 mW, 0.1 sec, 18 kHz. parameters: 150 mW, 0.1 sec, 18 kHz.

can be seen that the photocoagulation effect is local- found either by direct visualization during the exper- ized only in the photoreceptor outer segments and iment or histologically. RPE, while the photoreceptor nuclei and the inner segments seem to be intact. The laser parameters in Discussion this experiment were 3.5 /J per pulse (70 mW aver- Several studies have been published on the effect of age power) and 100 msec exposure time. At 6 /J per 16 19 pulse (100 mW average power) and 100 msec expo- pulsed laser exposure on the retina " that present sure, the photocoagulation effect was noticed also in the same histological results as we have found in our experiments done with 4 kHz pulse repetition rate. the outer nuclear layer (Fig. 8). Above 6 jid per pulse 16 (120 m W average power), a mild thermal damage was Geeraets found that Q-switch YAG laser caused dis- placement of the RPE layer and bulging of the photo- found in the inner retinal layers in addition to mild 17 disruption of the RPE, as can be seen in Figures 9 and receptor outer nuclear layer. Frisch reported on the 10, respectively. In the latter experiment the pulse findings of severe irregularities in the RPE layer asso- energy was 7.5 ^J (average power of 150 m W) and the ciated with vacuolization of the photoreceptors and at times subretinal hemorrhages, following exposure exposure time was 100 msec. It should be noted that 18 below 2 fiJ per pulse (40 mW average power), in a to Q-switch . Marshall found disruption relevant exposure time, no sign of coagulation was and vacuolization of the RPE cells in addition to pho- toreceptor inner and outer segment damage while using the same laser. Birngruber'9 is of the opinion that any pulsed laser which has a pulse width of less than 1 jxsec causes a mechanical effect including •V' disruption and displacement of several retinal layers. These results are in contrast to the thermal effect of the CW argon laser, causing coagulation of the outer retinal cells, which remain adherent to each other in their original layers.20 It should be stressed that all the above-mentioned experiments16"'9 were performed while using a single pulse YAG or ruby Q-switched laser. Mosier et al12 are the only researchers to report on experiments done with high repetition pulse of a fre- quency-doubled YAG laser (532 nm) exposure on the retina which can be compared with our experimental Fig. 9, A mild thermal damage is noted in the ganglion cell and 12 nerve fiber layers. Internal displacement of the outer and inner parameters. These authors point out that the use of nuclear layers is seen (arrow). Methylene blue, original magnifica- a focal spot size between 50 and 100 /urn aggravates tion X400. Laser parameters: 150 mW, 0.1 sec, 18 kHz. the mechanical effects on the retina due to irradiance

Downloaded from iovs.arvojournals.org on 09/26/2021 No. 4 RETINAL EFFECTS OF COPPER VAPOR LASER / Goboy er ol. 533

problems. They therefore recommend the use of a References larger spot size (200 nm or more), as was done in our 1. L'Esperance FA: An ophthalmic argon laser photocoagulation: CVL experiments. Their histological findings, follow- System, design, construction and laboratory investigations. ing retinal exposure to a single pulse or to a burst Trans Am Ophthalmol Soc 66:827, 1968. which consists of a first high peak power pulse, were 2. Tallef M, Ritter EJ, Rockwell RJ, Branch MS, and Lotspeich disruption and tearing of the retina and subretinal C: of the retina using the argon laser. Am J Ophthalmol 67:666, 1969. and vitreal hemorrhages with dispersion of the mela- 3. Wiesinger H, Schmidt FJ, Williams RC, Tiller CO. Ruffin RS, nin granules between the red blood cells. The latter Guerry D, and Ham WT: The transmission of the light through findings are similar to ours at 4 kHz pulse repetition the ocular media of the rabbit eye. Am J Ophthalmol 42:907, rate, with the exception that we have not found vit- 1956. real and subretinal hemorrhages but only choroidal 4. Geeraets WJ and Berry ER: Ocular spectral characteristics as hemorrhages associated with exudative detachment related to hazard from lasers and other light sources. Am J 12 Ophthalmol 66:15, 1968. of the RPE, choroid and retina. Mosier et al report 5. Birngruber R: Thermal modelling in biological tissues. In that the lesion caused by a uniform pulse train of 10 Lasers in Biology and Medicine, Hillenkamp F, Pratesi R, and kHz repetition rate of the frequency-doubled YAG Sacchi CA, editors. New York, Plenum Press, 1981, pp. 77-97. laser showed a definite zone of coagulation extending 6. Mainster MA, Sliny DH, Blecher CD, and Buzney SM: Laser photodisruptors: Damage mechanism, instrument design and from the RPE through the photoreceptors. The pri- safety. Ophthalmology 90:973, 1983. mary site of damage was the RPE, where he found 7. Aron-Rosa D: Use of Neodymium-YAG laser to open the gross disruption of the cells with melanin pigment posterior capsule after lens implant surgery: A preliminary re- dispersion. The RPE cells that remained intact were port. Journal of the American Intra-Ocular Implant Society highly vacuolated. The lamellar membranes of the 6:352, 1980. photoreceptor outer segments were fused together or 8. Fankhauser F, Kwasniewska S, Van Der Zypen E: Vitreolysis with the Q-switched laser. Arch Ophthalmol 103:1166, 1985. disintegrated. No damage was noticed in Bruch's 9. Walter WT, Solimene N, Piltch M, and Gould G: Efficient membrane or choroid, and there was some cellular pulsed gas discharge lasers. IEEE Journal of Quantum Elec- disintegration near the retinal surface. The latter find- tronics 2:474, 1966. ings are very similar to our histological results de- 10. Isaev AA and Lemerman CY: Investigation of copper vaor pulsed laser at elevated powers. Soviet Journal of Quantum scribed for retinal exposure to 18 kHz repetition rate, Electronics 7:799, 1977. with the exception that no vacuolization was noticed 11. Birngruber R, Hillenkamp F, and Gabel VP: Theoretical in- in the RPE cells, and that with the pulse energy of 3.6 vestigation of laser thermal retinal injury. Health Physics JUJ (70 mW average power) and exposure time of 0.1 48:781, 1985. second, only pure coagulation effect limited to the 12. Mosier MA, Champion J, Liaw LH, and Berns MW: Retinal effects of the frequency-doubled (532 nM) YAG laser: Histo- outermost retinal layers was seen, without any signs pathological comparison with argon laser. Lasers Surg Med of cellular disruption. This desired coagulation dam- 5:377, 1985. age is very similar to that described by Apple et al20 13. L'Esperance FA: Ophthalmic Lasers: Photocoagulation, Pho- using a CW argon laser with 200 mW power and toradiation and Surgery. St. Louis, C.V. Mosby Co., 1983, exposure time of 0.2 second. It should be added that chpt. 4, pp. 85-112. 14. Editorial: Search for the ideal laser. Br J Ophthalmol 63:655, other authors have described a wide range of retinal 1979. lesions produced by the CW argon laser, such as se- 15. Tremp CL, Mainster MA, and PomerantzeffO: Macular pho- vere damage to all retinal layers associated with sub- tocoagulation: Optimal wavelength selection. Ophthalmology retinal exudation and hemorrhage,2 vacuolization of 89:721, 1982. the RPE cells and photoreceptors18 and even thermal 16. Geeraets WJ: Retinal injury by ruby and Nd laser. Acta Oph- 21 thalmol 45:846, 1967. damage to the inner retinal layers. 17. Frisch GD, Beatrice ES, and Holsen RC: Comparative study of It is our conclusion that the high repetition rate argon and ruby retinal damage thresholds. Invest Ophthalmol copper vapor laser is suitable for retinal photocoagu- 10:911, 1971. lation in the range of 40 to 120 mW average power 18. Marshall J, Hamilton AM, and Bird AC: Histopathology of ruby and argon laser in monkey and human retina. Br J Oph- with 200 ixm spot size and in relevant exposure time thalmol 59:610, 1975. (around 0.1 sec). It is hoped that future modification 19. Bringruber R, Gabel VP, and Hillenkamp F: Experimental of this device will make it useful both as a retinal studies of laser thermal retinal injury. Health Phys 44:519, coagulator as well as a photodisruptor laser for the 1983. treatment of anterior segment pathology. 20. Apple DJ, Wyhinny GJ, Goldberg MF, Bizzell JW, and Bro- derson JP. Experimental argon laser photocoagulation. Arch Key words: copper vapor laser, high repetition rate pulsed Ophthalmol 94:309, 1976. laser, variable peak power, photodisruption, quasi-CW 21. L'Esperance FA: The ocular histopathologic effect of krypton laser and argon laser radiation. Am J Ophthalmol 68:263, 1969.

Downloaded from iovs.arvojournals.org on 09/26/2021