Ultrasound Biomicroscopic Imaging of the Effects of YAG Laser Cycloablation in Postmortem Eyes and Living Patients Charles]. Pavlin, MD,l Peter Macken, MD,l Graham E. Trope, MD, l Godfrey Heathcote, MD,2 Michael Sherar, PhD,3 Kasia Harasiewicz, PEng,4 F. Stuart Foster, PhD4 Purpose: The authors performed a series of experiments designed to determine if early effects of Y AG laser cycloablation could be detected by ultrasound biomicroscopy in postmortem eyes and living patients. They also designed an apparatus that allowed simultaneous ultrasound biomicroscopic imaging of YAG laser cycloablation. Methods: Treated and untreated regions of postmortem eyes treated with Y AG cycloablation were imaged and compared. Treatment was placed at varying distances from the limbus in postmortem eyes and the resulting effects imaged. Histologic examinations were performed after imaging. Six living patients had ultrasound biomicroscopy before and after YAG cycloablation. An apparatus combining contact YAG laser and ultrasound biomicroscopy was used in postmortem eyes. Results: Early treatment effects imaged included ciliary epithelial disruption, ciliary epithelial separation, and bubble formation. Ultrasound biomicroscopic findings varied with the distance of treatment from the limbus and were maximal below the treatment site. Results of histologic examination showed close correlation to the ultrasound biomicroscopic images. Similar findings to those found in postmortem eyes were found in living patients after treatment. The apparatus combining contact Y AG and ultrasound biomicroscopy allowed realtime imaging of effects of YAG laser cycloablation. Conclusions: The ability of ultrasound biomicroscopy to detect changes associated with cyclodestructive procedures potentially could provide us with a method of improving treatment preciSion and correlating treatment effect with clinical response. Ophthalmology 1995;102:334-341
Originally received: July 8, 1994. Revision accepted: October 5, 1994. I Department of Ophthalmology, The Toronto Hospital, Toronto. 2 Department of Pathology, St. Joseph's Health Centre, London, Ontario. 3 Department of Medical Physics, Princess Margaret Hospital, Toronto. 4 Department ofimaging Research, Sunnybrook Health Science Centre, Toronto. Supported in part by the National Cancer Institute of Canada, Toronto, and the Ontario Cancer Research and Treatment Foundation, Toronto. Drs. Pavlin, Sherar, and Foster have a small proprietary interest in ultrasound biomicroscopic instrumentation.
Several methods of applying energy to the ciliary body have been used to produce decreased aqueous secretion in refractory glaucoma. These methods include cryotherapy,l therapeutic ultrasound,2 and various forms oflaser energy, including ruby,3 diode,4 krypton,S and neodymium:YAG (Nd:YAG). Neodymium:YAG laser energy can Presented in part at the AR VO Annual Meeting, Sarasota, May 1994. Reprint requests to Charles J. Pavlin, MD, Ocular Oncology Clinic, Princess Margaret Hospital, 500 Sherbourne St, Toronto, Ontario, CanadaM4X IK9.
Pavlin et al . Ultrasound Biomicroscopy of YAG Laser Cycloablation be applied to the eye by contact and noncontact methods and commonly is used for accomplishing ciliary body destruction. 6- IO This procedure has complications that include inflammation and hypotony. Improved methods of titrating the amount of laser energy applied and more precise determination of the site of application theoretically could improve the therapeutic effect/complication ratio. I I Results of histologic studies in cadaver eyes, animal eyes, and eyes enucleated after treatment have shown various treatment effects, including ciliary epithelial disruption, ciliary epithelial separation, and regions of ciliary body coagulation. 12 - 19 These effects are difficult to study in living human patients because of problems in visualizing the ciliary body. The technique of high-frequency ultrasound biomicroscopy, developed at our institution, allows subsurface imaging of the ciliary body at much higher resolution than has been possible previously.20-22 Using this instrument, we performed a series of experiments in eye bank eyes and a series of examinations in living patients designed to determine if the effects ofYAG laser cyclodestruction can be detected by ultrasound biomicroscopy in eye bank eyes, the nature of the ultrasound changes, whether the observed effects vary according to the site of laser application, and whether the same ultrasound biomicroscopic findings can be detected in the clinical setting. We performed further experiments designed to assess the feasibility of imaging treatment effects in the ciliary body region simultaneously with energy application. We constructed an apparatus combining ultrasound biomicroscopy with a contact Y AG laser probe and used this apparatus to perform ciliary destructive procedures on cadaver eyes.
Materials and Methods Two ultrasound biomicroscopic instruments were used in these experiments: the original instrument designed in our laboratories and the commercial version of our instrumentation developed by Zeiss-Humphrey (San Leandro, CA). The former instrument was used with a 62MHz transducer, and the latter instrument was used with a 50-MHz transducer. Details regarding the concepts and methods of ultrasound biomicroscopy have been described previously.2o-22 Ultrasound biomicroscopy uses high-frequency transducers in a B-scan mode to produce images below the surface of the eye at microscopic resolution. Penetration is confined to 4 to 5 mm, which allows imaging of the entire ciliary body and adjacent structures. Eye bank eyes were examined in a water bath. Living patients were examined using an eyecup of our own design and methyl cellulose as a couplant.
In Vitro Experiments Two eye bank eyes unsuitable for transplantation were used for initial experiments designed to determine whether a treatment effect was detectable by ultrasound biomicroscopy. The laser used in this study was the Lasag Mi-
croruptor II Neodymium:YAG laser (Thun, Switzerland). Each eye received 20 burns of 3.7 J, offset 9, and pulse duration of 20 mseconds over 180°. Burns were placed 1.5 mm posterior to the limbus. One half of the circumference of the globe was left untreated for comparison. Eyes then were examined by ultrasound biomicroscopy in a water bath approximately I hour after treatment. Treated zones were imaged and compared with untreated zones. Images were stored on a laser disk. The eyes then were fixed by immersion in neutral-buffered formalin for at least 48 hours. The posterior half of the globe was removed, and the ciliary body photographed. Sections from treated and untreated regions were processed routinely for light microscopy. An eye bank eye had markers consisting of 7-0 silk sutures placed at various distances from the limbus. These markers were placed at the limbus and I, 2, and 3 mm from the limbus. Adjacent to each of these sites, a series of approximately five contiguous laser burns (3.7 J, offset 9, and pulse duration of 20 mseconds) were applied at the same distance from the limbus and covering approximately 2 clock hours. A region of approximately 3 clock hours was left untreated. The eye was examined in a water bath by ultrasound biomicroscopy approximately I hour after treatment. Images were taken through the sutures to determine the relation of underlying structures to surface anatomy. Images then were taken through the treated regions adjacent to the sutures at each of the four sites and through the untreated region. After fixation , sections from each region were processed for light microscopy.
In Vivo Experiments Six patients were examined by ultrasound biomicroscopy after being treated with YAG laser cyclodestruction for varied types of refractory glaucoma. Three patients had neovascular glaucoma: one had aphakic glaucoma secondary to a penetrating eye injury and two had uncontrolled pseudophakic glaucoma. In each case, 40 burns of 3.7 J, offset 9, and 20 mseconds pulse duration were applied 1.5 mm posterior to the limbus over 360°. Ultrasound biomicroscopic examination was performed using an eyecup and methyl cellulose couplant approximately 0.5 to 1.5 hours after treatment.
Combining Laser and Ultrasound Biomicroscopy We constructed an apparatus that combined the ultrasound biomicroscope with a contact Y AG laser (Surgical Laser Technologies Inc, Malvern, PA). We used the original version of the ultrasound biomicroscope constructed in our laboratories with a 62-MHz transducer. Because of the nature of the process of ultrasound biomicroscopy (i.e., a moving transducer in a B-scan mode), it is difficult to produce a laser beam and image plane that are coaxial. We constructed a holder attached to the ultrasound biomicroscope that held the probe at a fixed angle to the plane of the image (Fig I). We attempted to make this
Volume 102, Number 2, February 1995 3B. Ultrasound biomicroscopy of the treated region is shown in Figure 3C. The effects observed consisted of separation of the ciliary epithelium and gas bubble formation. Gas bubbles appear as a region of high reflectivity with reverberation echoes behind them. This is a typical ultrasound appearance of gas bubbles and is unlikely to be produced by any other feature. In this case, the gas
Scanning Mechanism UBM Image Plane Perpendicular to page
~. uttrasound Beam
Figure L Diagrammatic representation of a device combining contact YAG laser and ultrasound biomicroscopy. The laser probe is fixed to the ultrasound biomicroscope at an angle of approximately 65 0 , with the laser beam passing through the focal plane of the transducer at the mid point of its excursion.
angle as small as possible to be consistent with mechanical constraints. The angle of the probe to the ultrasound instrument was designed with the goal of placing the line of delivery of laser energy through the focal plane of the scan at mid range of transducer excursion. The angle the probe made to the ultrasound instrument was approximately 65 0 • Figure 2A shows the apparatus and Figure 2B shows a closeup view of a cadaver eye in a water bath with the apparatus in place. Eye bank eyes unsuitable for transplantation were used. The eyes were fixed in the bottom of a water bath by suturing the optic nerve to a weight. The apparatus was applied to the eye so that the probe contacted the sclera approximately I mm from the limbus. Continuous ultrasound images were obtained before and during treatment. The laser was turned on using various energy levels from 5 to 8 J at a O.5-second duration. We used higher energy levels, because our goal was to determine our ability to observe effects, not to image subtle findings, at low energy. Resulting effects were observed in realtime on the screen, and representative images were stored on a laser disk.
In Vitro Experiments Figure 3A shows the internal appearance of the treated and untreated zones in a cadaver eye. The treated region shows areas of gray discoloration at the posterior aspect of the pars plicata. Ultrasound biomicroscopic appearance of the untreated ciliary body region is shown in Figure
Figure 2. A, photograph of the experimental apparatus used to apply laser energy to a postmortem eye in a water bath while simultaneously imaging the region of application by ultrasound biomicroscopy. B, closeup view shows the relation of the laser probe and transducer to the ocular surface.
Pavlin et al . Ultrasound Biomicroscopy of YAG Laser Cycloablation
Figure 3. A, photograph of the ciliary body region of a dissected postmortem eye after laser cycloablation. The treated area (right side of the photograph between the arrows) shows gray areas just behind the ciliary processes. B, ultrasound biomicroscopic image through the untreated region shows a normal appearance of the ciliary body (arrow). C, ultrasound biomicroscopic image of the treated region shows separation of the ciliary epithelium just behind the ciliary process (white arrow). A bubble is present in the cyst created (black arrow). The bubble shows high reflectivity, reverberation echoes, and shadowing of structures behind it. D, histologic section through the treated region shows ciliary epithelial separation and disruption (arrow) with some subepithelial coagulative changes (hematoxylin-eosin; original magnification, X20).
bubble is within the region of ciliary separation. A histologic section through the treated region (Fig 3D) shows separation of the ciliary pigmented and nonpigmented epithelium and a region of coagulative change in the subepithelial stroma. Gas bubbles would not survive the processing required for histologic examination. Figures 4 to 7 show a series of images in the cadaver eye that was treated at the limbus and at I-mm intervals posterior to the limbus. The second image in each group shows the histologic findings in the treated region. An open arrow in each ultrasound biomicroscopic illustration shows the surface treatment site as determined by imaging the suture placed to mark each treatment zone. With the treatment site at the limbus, ultrasound biomicroscopy of the treated region shows loss of iris pigment epithelium in the peripheral iris (Fig 4A). This is
noted as a disruption in the highly reflective layer normally found on the posterior iris surface. The histologic section (Fig 4B) shows loss of peripheral iris pigment epithelium. With the treatment site I mm posterior to the limbus, ultrasound biomicroscopy shows some disruption of ciliary epithelium on the anterior aspect of the ciliary process as well as loss of peripheral iris pigment epithelium (Fig SA). The histologic section (Fig SB) shows epithelial disruption in the ciliary process and peripheral iris region as well as some areas of coagulative change in the stroma of the ciliary process. With the treatment site 2 mm posterior to the limbus, ultrasound biomicroscopy (Fig 6A) shows separation of the ciliary epithelium and gas bubble formation at the posterior aspect of the ciliary process. Histologic findings (Fig 6B) show ciliary epithelial separation at the posterior
Volume 102, Number 2, February 1995 aphakic patient with advanced glaucoma before treatment. Figure 8B shows the ciliary body region after treatment. Ultrasound biomicroscopy shows separation of the ciliary epithelium and bubble formation. The ultrasound biomicroscopic appearance of a cyst with a highly reflective bubble with reverberation echoes behind it is similar to the findings in eye bank eyes. The sixth patient did not show detectable changes on ultrasound biomicroscopy and this may have been due to an inability to detect changes in some patients or to obscuration of changes by the extensive disorganization of the anterior segment present. Exudate also was present over the ciliary body in this patient.
Combined Laser and Ultrasound Biomicroscopy The immediate effects of laser application were visible in realtime using our apparatus that combined contact laser
Figure 4. A, treated zone centered at the limbus as determined by imaging the sutures placed at various distances from the limbus (open arrow). Ultrasound biomicroscopic image shows loss of peripheral iris epithelium (arrow). B, histologic section through this zone shows loss of iris epithelium from the peripheral iris (arrow) (hematoxylin-eosin; original magnification, X20).
aspect of the ciliary process and some coagulative changes in the subepithelial stroma. With the treatment site 3 mm posterior to the limbus, ultrasound biomicroscopy (Fig 7A) shows separation of the ciliary epithelium in the pars plana region with gas bubble formation. Histologic findings (Fig 7b) show ciliary epithelial separation in the pars plana with coagulative changes in the subepithelial stroma.
In Vivo Experiments Five of six patients showed changes on ultrasound biomicroscopy after YAG laser cyclophotocoagulation. These five patients showed evidence of ciliary epithelial disruption and separation. Three patients showed bubble formation. Figure 8A shows the ciliary body region in an
Figure 5. A, treated zone centered 1 mm from the limbus as determined by imaging the sutures placed at various distances from the limbus (open arrow). Ultrasound biomicroscopic image shows some disruption of the ciliary epithelium on the anterior ciliary process (arrow) and loss of peripheral iris epithelium. B, histologic section through this zone shows disruption of anterior ciliary epithelium (arrow) and loss of peripheral iris epithelium. Some subepithelial coagulative changes are present (hematoxylin-eosin; original magnification, X20).
Pavlin et al . Ultrasound Biomicroscopy of YAG Laser Cycloablation segment abnormalities. Visual monitoring allows titration of power levels and precise localization of treatment position. This opportunity is denied us when treatment is applied to structures not accessible to direct observation. Ciliary body destructive procedures are examples of conditions in which energy is applied without direct visualization of the intended target of treatment. The development of ultrasound biomicroscopy in our laboratories has allowed imaging of the ciliary body at microscopic resolution in the intact eye. We designed a series of experiments in eye bank eyes and examined a series ofliving patients to determine the ability of ultrasound biomi-
Figure 6. A, treated zone centered 2 mm from the limbus as determined by imaging the sutures placed at various distances from the limbus (open arrow). Ultrasound biomicroscopic image shows some disruption and separation of the ciliary epithelium just behind the ciliary process (arrow) with a gas bubble on the surface. B, histologic section through this zone shows separation of ciliary epithelium just behind the ciliary process (arrow). Some subepithelial coagulative changes arepresent (hematoxylineosin; original magnification, X20).
and ultrasound biomicroscopy. Effects were more apparent as the laser energy was increased. Figures 9A and 9B show a region of the ciliary body before treatment and the same area during application oflaser energy. Observed effects included vibration of structures, disruption of ciliary epithelium, disruption of peripheral iris epithelium, and bubble formation. Bubble formation is indicated by small regions of high reflectivity with shadowing behind them.
Discussion Visual monitoring of laser treatment effects is extremely useful when treating retinal disease and various anterior
Figure 7. A, treated zone centered 3 mm from the limbus as determined by imaging the sutures placed at various distances from the limbus (open arrow). Ultrasound biomicroscopic image shows some separation of the ciliary epithelium over the pars plana (arrow) with a g as bubble in the resulting cyst. B, histologic section through this zone shows separation of ciliary epithelium over the pars plana (arrow). Some subepithelial coagulative changes are present (hematoxylin-eosin; original magnification, X20).
Volume 102, Number 2, February 1995 mostly confined to iris epithelium with treatment at the limbus to ciliary epithelial disruption and separation at progressively posterior locations as the distance of the treatment from the limbus increased. These findings show that ultrasound biomicroscopy has the capability of localizing the site of treatment and could play a role in improving the precision by which such treatment is applied. These results do not establish the relative effectiveness of different treatment positions in intraocular pressure reduction. The findings in clinical cases show that in the majority of patients, effects similar to those observed experimentally are detectable by ultrasound biomicroscopy. Ciliary epithelial disruption, epithelial separation, and bubble formation all were observed. Patients were examined within 1.5 hours after treatment. It is likely that the observed
Figure 8. A, ultrasound biomicroscopy of the ciliary body region in a patient with intractable glaucoma secondary to neovascularization. The ciliary body (arrow) is somewhat thinner than normal. B, ultrasound biomicroscopy of the ciliary body region after YAG laser cyclodestruction. Epithelial separation and disruption (arrow) with gas bubble formation is noted.
croscopy to image changes associated with YAG laser cyclodestructive treatment. Changes observed on ultrasound biomicroscopy in eye bank eyes consisted of ciliary epithelial disruption, ciliary epithelial detachment, and bubble formation. Epithelial disruption and detachment were confirmed by histologic examination. Gas bubble formation is likely a transient phenomenon. Gas bubble formation, though transient, may be a helpful ultrasound indicator of the site of treatment at the time of, or immediately after laser application. Coagulative effects noted in the ciliary body stroma on histology could not be detected by ultrasound biomicroscopy. Treatment at various distances from the limbus produced ultrasound biomicroscopically detectable effects that were maximal in the region immediately subjacent to the treatment site. These effects varied from those
Figure 9. A, ultrasound biomicroscopy of the ciliary body region in a cadaver eye before treatment with the combined contact YAG and ultrasound biomicroscopic apparatus. B, ultrasound biomicroscopy during treatment shows disruption of ciliary processes and peripheral iris with gas bubble formation (arrow).
Pavlin et al . Ultrasound Biomicroscopy of YAG Laser Cycloablation ultrasound biomicroscopic signs diminish with time. This is certainly true of gas bubbles. We did not attempt to do serial examinations over time in these patients. We have attempted to establish the principle that realtime ultrasound biomicroscopic monitoring oflaser effects is possible in regions not visually accessible. Using an apparatus that combined ultrasound biomicroscopy with contact Y AG laser, we were able to image various structural changes in the ciliary body as they occurred. We did not attempt to precisely quantitate the degree of observed effects with the level of energy application. The inability to have laser energy coaxially aligned with the ultrasound biomicroscopic plane requires a controlled angle at which these two energy sources cross. This limits the ease of monitoring and makes the apparatus more complex. Engineering improvements could minimize these problems. The ability of ultrasound biomicroscopy to detect changes associated with cyclodestructive procedures could potentially provide us with a method of improving treatment precision, and correlating treatment effect with clinical response. Simultaneous imaging of treatment application is feasible and could result in an apparatus by which the laser energy could be "aimed" at the desired location, and the amount of treatment titrated. It should be emphasized that we have not correlated ultrasound findings with treatment effect in this article, and further research will be required to determine if such a correlation can be made.
References I. Bellows AR. Cyclocryotherapy for glaucoma. Int Ophthalmol Clin 1981 ;21(1):99-111. 2. Burgess SEP, Silverman RH, Coleman DJ, et al. Treatment of glaucoma with high-intensity focused ultrasound. Ophthalmology 1986;93:831-8. 3. Beckman H, Kinoshita A, Rota AN, Sugar HS. Transscleral ruby laser irradiation of the ciliary body in the treatment of intractable glaucoma. Trans Am Acad Ophthalmol Otolaryngol 1972;76:423-36. 4. Hennis HL, Stewart We. Semiconductor diode laser transscleral cyclophotocoagulation in patients with glaucoma. Am J Ophthalmol 1992;113:81-5. 5. Immonen IJR, Puska P, Raitta e. Transscleral contact krypton laser cyclophotocoagulation for treatment of glaucoma. Ophthalmology 1994; 101 :876-82. 6. Beckman H, Sugar HS. Neodymium laser cyclocoagulation. Arch Ophthalmol 1973;90:27-8.
7. Fankhauser F, van der Zypen E, Kwasniewska S, et al. Transscleral cyclophotocoagulation using a neodymium YAG laser. Ophthalmic Surg 1986; 17:94-100. 8. Brancato R, Giovanni L, Trabucchi G, Pietroni C. Contact transscleral cyclophotocoagulation with Nd:YAG laser in uncontrolled glaucoma. Ophthalmic Surg 1989;20:547-51. 9. Trope GE, Ma S. Mid-term effects of neodymium:YAG transscleral cyclocoagulation in glaucoma. Ophthalmology 1990;97:73-5. 10. Schuman JS, Bellows AR, Shingleton BJ, et at. Contact transscleral Nd:YAG laser cyclophotocoagulation. Midterm results. Ophthalmology 1992;99: 1089-95. II. Allingham RR, de Kater AW, Bellows AR, Hsu J. Probe placement and power levels in contact transscleral neodymium:YAG cyclophotocoagulation. Arch Ophthalmol 1990;108:738-42. 12. Devenyi RG, Trope GE, Hunter WHo Neodymium-YAG transscleral cyclocoagulation in rabbit eyes. Br J Ophthalmol 1987;71:441-4. 13. Shields SM, Stevens, JL, Kass MA, Smith ME. Histopathologic findings after Nd:YAG transscleral cyclophotocoagulation [letter]. Am J Ophthalmol 1988; 106: 100-1. 14. Schubert HD. Noncontact and contact pars plana transscleral neodymium:YAG laser cyclophotocoagulation in postmortem eyes. Ophthalmology 1989;96:1471-5. 15. Simmons RB, Blasini M, Shields MB, Erickson PJ. Comparison of transscleral neodymium:YAG cyclophotocoagulation with and without a contact lens in human autopsy eyes. Am J Ophthalmol 1990; 109: 174-9. 16. Blasini M, Simmons R, Shields MB. Early tissue response to transscleral neodymium:YAG cyclophotocoagulation. Invest Ophthalmol Vis Sci 1990;31:1114-18. 17. Brancato R, Leoni G, Trabucchi G, Cappellini A. Histopathology of continuous wave neodymium:yttrium aluminum garnet and diode laser contact transscleral lesions in rabbit ciliary body. Acomparative study. Invest Ophthalmol Vis Sci 1991;32:1586-92. 18. Coleman AL, Jampel HD, Javitt JC, et al. Transscleral cyc1ophotocoagulation of human autopsy and monkey eyes. Ophthalmic Surg 1991 ;22:638-43. 19. Simmons RB, Prum BE Jr, Shields SR, et al. Videographic and histological comparison of Nd:YAG and diode laser contact transscleral cyclophotocoagulation. Am J Ophthalmol 1994;117:337-41. 20. Pavlin CJ, Sherar MD, Foster FS. Subsurface ultrasound microscopic imaging of the intact eye. Ophthalmology 1990;97:244-50. 21. Pavlin CJ, Harasiewicz K, Sherar MD, Foster FS. Clinical use of ultrasound biomicroscopy. Ophthalmology 1991;98: 287-95. 22. Pavlin CJ, Harasiewicz K, Foster FS. Ultrasound biomicroscopy of anterior segment structures in normal and glaucomatous eyes. Am J Ophthalmol 1992;113:381-9.