- Email: [email protected]
Potential Artifacts in Scanning Electron Microscopy of the Trabecular Meshwork in Glaucoma
POTENTIAL ARTIFACTS IN SCANNING ELECTRON MICROSCOPY OF THE TRABECULAR MESHWORK IN GLAUCOMA M. MAGLIO,
B.A., C. AND
McMAHON, M.D., D. HOSKINS, M.D., ALVARADO, M.D.
J.
San Francisco, California
The scanning electron microscope is a useful tool for studying the trabecular meshwork surface and abnormalities that may contribute to the malfunction of this tissue in glaucoma. 1-3 Trabeculectomy specimens are one important source of pathologic tissue, but these specimens require special care in order to prevent contamination of the tissue surface during collection and processing for scanning electron microscopy. While studying the trabecular meshwork from patients with primary openangle glaucoma, we became familiar with the scanning electron microscopic appearance of some potential problems or artifacts caused by methods of collecting and preparing trabecular tissues. In an attempt to identify the source of contamination, we studied the effects of decompression of the globe, bleeding, and other complications introduced by trabeculectomy, and compared the appearance of washed and unwashed trabeculectomy specimens. We used whole enucleated eyes or eyes obtained at autopsy to show that the meshwork in these eyes is not coated by the materials found in the unwashed trabeculectomy specimens. We examined glaucomatous trabecular tissues from whole eyes obtained at enucleation or autopsy to determine whether From the Department of Ophthalmology, University of California School of Medicine, San Francisco, California. This study was supported by grants EY-02068 and EY-02162A from the National Eye Institute (Dr. Alvarado). Reprint requests to J. Alvarado, M.D., U-490, University of California, School of Medicine, San Francisco, CA 94143.
the meshwork of such patients is obstructed by a blocking substance as reported by others. I We studied the scanning electron microscopic appearance of the substances that coat and apparently obstruct the trabecular interstices in trabeculectomy specimens. We interpreted these findings as artifacts caused by specimen collection or processing. These data do not confirm the reported finding that an "amorphous blocking substance" is found in the trabecular meshwork in primary open-angle glaucoma. I We offer an alternative explanation. MATERIAL AND METHODS
This study included 17 specimens of two types: surgical specimens obtained at trabeculectomy and whole eyes obtained at autopsy or enucleation. There were ten trabeculectomy specimens. The assistant surgeon washed five of them during collection by continuously irrigating the tissues with a balanced saline solution; the other five were collected without washing. For comparison, we studied seven whole eyes, including five eyes without glaucoma obtained at autopsy and two glaucomatous eyes. One glaucomatous eye was enucleated because of a posterior choroidal melanoma in a patient with open-angle glaucoma and exfoliation. Another globe, obtained at autopsy, had primary open-angle glaucoma. Originally it had been fixed in formalin. We submerged the seven whole eyes in fixative for one to two hours and dissected pie-shaped pieces that included the trabecular meshwork. Immediately after
AMERICAN JOURNAL OF OPHTHALMOLOGY 90:645-653, 1980
645
646
AMERICAN JOURNAL OF OPHTHALMOLOGY
excision, we immersed all of the specimens in a 2% glutaraldehyde and 1% paraformaldehyde solution at pH 7.4, buffered with 0.067 M of sodium cacodylate containing 5mM of calcium chloride. All the specimens were kept in the fixative solution overnight at 4°C; the next day they were transferred to a freshly prepared batch of the same fixative. We kept all the tissues in this fixative for two to four days and then rinsed them in a cold solution of 0.5 M of sucrose at pH 7.4, buffered with 0.1 M of sodium cacodylate. They were dehydrated in a series of alcohols increasing in concentration up to 100% ethyl alcohol; the specimens then were critical-point dried through carbon dioxide. We mounted the specimens on metal stubs in the usual manner, using a conducting silver adhesive, and coated them with approximately 15 nm of gold in a Hummer II sputter-coater. All the specimens were viewed with a scanning electron microscope at 20 kV and photographed with Polaroid 55 PIN film.
NOVEMBER, 1980 RESULTS
Studies using trabeculectomy specimens are limited by the difficulty of obtaining appropriate controls, since ideally these consist of trabeculectomy specimens from non glaucomatous patients. In an attempt to circumvent this problem, we compared washed to unwashed trabeculectomy specimens in order to study the presence of a tightly adhering blocking material in the meshwork. Additionally, we compared the scanning electron microscopic findings in nonglaucomatous and glaucomatous whole eyes (obtained at autopsy or enucleation) in regard to the presence of this same blocking substance. Finally, we examined the scanning electron microscopic appearance of probable artifacts introduced after tissue collection and specimen preparation. The trabecular meshwork of washed specimens was free of substances that might block the flow of aqueous through the uveal and corneoscleral meshwork (Fig. 1). The unwashed trabeculectomy
Fig. 1 (Maglio and associates). Washed trabeculectomy specimen. Left, Schwalbe's line (arrow). Uveoscleral and comeoscleral meshworks contain no amorphous material obstructing the interstices (x 2(0). Right, A higher-magnification view, showing meshwork interstices (x 1,500).
Fig. 2 (Maglio and associates). Unwashed trabeculectomy specimens. Top left, Amorphous substance within the trabecular spaces (x 5,(00). Top right, Fibrillar material forming a reticular pattern over the surface of the meshwork (x 1,(00). Bottom left, Numerous red blood cells covering a space in the meshwork (x 3,600).
647
648
AMERICAN JOURNAL OF OPHTHALMOLOGY
specimens had a number of substances lodged in the trabecular interstices (Fig. 2). One of these materials was amorphous (Fig. 2, top left) and closely resembled some of the materials shown by other investigators. 1 Another material was fibrillar in nature and looked like fibrin (Fig. 2, top right). These materials were observed in most unwashed specimens even when there had been no apparent bleeding during the trabeculectomy. When noticeable bleeding had occurred, they were present in abundance and were usually accompanied by red blood cells (Fig. 2, bottom left). Thus, these materials may have come from blood vessels and were readily removed from the trabeculectomy specimens by washing.
NOVEMBER, 1980
Figure 3 shows the appearance of the trabecular meshwork in one normal eye obtained at autopsy and Figure 4 shows that of an enucleated specimen from a patient with open-angle glaucoma and exfoliation. Neither specimen shows the presence of the materials observed in the unwashed trabeculectomy specimens. The only substance found in the trabecular interstices in the enucleated specimen consists of the exfoliation material (Fig. 4, top right and bottom left), which is formed by numerous fibrils gathered into a ball resembling a tumbleweed. Although we suspect that patients with the exfoliation syndrome have primary openangle glaucoma made worse by this obstructive exfoliation material, our comparison might be questioned by others
Fig. 3 (Maglio and associates). Normal autopsy specimen from the eye of a 28-year-old individual. Left, Endothelial surface of the cornea (a), uveoscleral meshwork (b), and iris surface (c) (x 1,500). Right, Side view of same specimen shows trifurcated Schlemm's canal (a), the sheets of the corneoscleral meshwork (b), and the uveoscleral meshwork (c). Note the open trabecular pores (arrows) (x 440).
Fig. 4 (Maglio and associates). Autopsy specimen from a patient with open-angle glaucoma and the exfoliation syndrome. Top left, Trabecular spaces are open and the only substance found consists of the fibrillar exfoliative material (arrows) (x 1,5(0). Top right, Higher-power view of exfoliative material (x 3,000). Bottom left, Trabecular cells with clear intercellular spaces; the nodular surface of these cells is presumably related to the presence of ingested melanin granules (x 1,100). Bottom right, Uveoscleral meshwork; there is a loosely adherent material of unknown nature but the meshwork has clear interstices (x 1,100).
649
650
AMERICAN JOURNAL OF OPHTHALMOLOGY
who consider this condition secondary open-angle glaucoma. Therefore, we examined by scanning electron microscopy another eye with primary open-angle glaucoma obtained at autopsy from the University of California Eye Pathology Laboratory, San Francisco. Figure 5 shows that there was no material blocking the meshwork of this specimen. Although we accounted for the presence of most of the blocking materials by the formation of plasma and other proteins during trabeculectomy, we still had not demonstrated the presence of another substance with a smooth appearance and many cracks on its surface found by other investigators. 1 Such cracks might have been the result of incomplete carbon dioxide drying or of the removal of
NOVEMBER, 1980
some volatile substance in the vacuum chamber of the scanning electron microscope. Acetone or butyl acetate are volatile substances present in the silver paint adhesive used to fasten the tissue specimens to metal stubs before viewing. When we liberally applied this adhesive to trabecular tissues similar in size to the typical trabeculectomy specimen (2 X 4 mm), the adhesive seeped into the hollow trabecular interstices, completely blocking them. Figure 6 shows the smooth appearance of this adhesive and the many cracks on its surface. With the aid of a dissecting microscope, small amounts of this adhesive can be applied, thus preventing contamination of the small pieces of tissue obtained at trabeculectomy.
Fig. 5 (Maglio and associates). Eye obtained at autopsy from a patient with primary open-angle glaucoma. Left, Despite formalin fixation, the usual three-dimensional architecture of the meshwork is evident (x 48). Right, Detail of uveoscleral meshwork showing that the trabecular interstices are not covered with amorphous material (x 1(0).
VOL. 90, NO. 5
TRABECULAR NETWORK MICROSCOPY
651
Fig. 6 (Maglio and associates). Contamination of the meshwork with the silver paint adhesive. Left, Surface of the trabecular cells appears to be coated with the adhesive; particles shown at arrows represent large deposits of silver. The cracks along the surface (double arrows) are produced by outgassing in the vacuum chamber (x 7,8(0). Right, A specimen contaminated by a light application of the adhesive. The surface, through which nodular structures in the trabecular cells are visible, is smooth. Note the minute cracks along a uveal band (x 2,000).
652
AMERICAN JOURNAL OF OPHTHALMOLOGY DISCUSSION
The appearance of the trabecular meshwork surface as shown by the scanning electron microscope varies considerably depending on the conditions at specimen collection and on the source of the tissue specimens. The surfaces of specimens obtained at trabeculectomy are often covered by an amorphous substance, a fibrillar material, and, occasionally, red blood cells. These materials can be easily removed in most cases by washing the specimens during collection. When whole eyes obtained at autopsy or enucleation are used as the source of tissue samples, the trabecular meshwork surface is free of these substances and washing is not necessary. We found this to be so in normal as well as glaucomatous specimens. These findings seem to support Svedbergh's" suggestion that abnormal substances can be released from the circulation during trabeculectomy and may be precipitated onto tissue surfaces. A recent series of studies has shown that there is a marked alteration of the various blood barriers with decompression of the globe in primate and human eyes.v" Haviola! was the first to show that plasma proteins reflux through Schlemm's canal and the trabecular meshwork into the anterior chamber after paracentesis. Okisaka" demonstrated that leakage of proteins into the posterior chamber also occurs with decompression of the globe, and that this is related to a breakdown of the blood-aqueous barrier localized to the anterior portion of the pars plicata. Bartels and associates" studied the sequence of events in the breakdown of the 'blood-aqueous barrier with paracentesis. They reported that protein leaks first into the posterior chamber and a few minutes later into the anterior chamber from Schlemrn's canal and the meshwork. Another important finding was the determination that the degree of
NOVEMBER, 1980
protein leakage is influenced more by the rapidity with which the intraocular pressure is decreased than the level to which it decreases.f To our knowledge, there have been no studies comparing the rate of pressure decrease at trabeculectomy with pressure decrease at paracentesis and other surgical procedures. Tso and Shih," however, reported that there was a breakdown of the blood-retina and blood-pigment epithelium barriers after trabeculectomy in monkeys. Possibly, most procedures in which the anterior chamber is entered produce a rapid decompression and much protein leakage. This is supported by the report of Zimmerman and associates'" who measured a significant degree of protein leakage into the anterior chamber in humans undergoing most anterior segment surgical procedures. We believe that plasma proteins, as well as other substances released by bleeding, constitute the major contaminants observed in our trabeculectomy specimens. The appearance of the trabecular meshwork after paracentesis and enucleation in monkeys has been studied by scanning electron microscopy." This report agrees with our findings, and both studies show that the meshwork of enucleated eyes is free of materials coating or blocking the trabecular meshwork. This is consistent with the idea that at trabeculectomy the intraocular pressure decreases much more rapidly than it does in enucleated eyes, and that this difference is an important factor in determining the degree of protein leakage. 8 We needed data from both enucleated eyes and trabeculectomy specimens in order to examine the provocative finding of an amorphous blocking substance in the trabecular meshwork associated with primary open-angle glaucoma.' Enucleated and autopsy specimens from patients with primary-open angle glaucoma are difficult to obtain. Nevertheless, we had
VOL. 90, NO. 5
TRABECULAR NETWORK MICROSCOPY
an opportunity to study eyes from three patients with primary open-angle glaucoma who had never undergone ocular surgery. We performed scanning electron microscopy in the two specimens reported here; the third specimen was studied only by transmission electron microscopy (I. Wood, unpublished data). Our scanning electron microscopy studies found no substances deposited over the trabecular tissues. A review of the transmission electromicrographs from the third specimen gave similar findings. The data from the trabeculectomy specimens and from eyes of patients with primary open-angle glaucoma suggest that the blocking material is released at the time of the trabeculectomy procedure as a result of the sudden decompression of the globe. We observed a blocking substance similar to that previously reported' only when the trabecular tissues were contaminated with the adhesive used in the preparation of the tissue samples for viewing by scanning electron microscope. Whether this accounts for the presence of a blocking material in the previous report! is not certain. We believe, however, that special care must be taken in the analysis of all trabeculectomy specimens because contamination often produces misleading effects. SUMMARY
We examined the trabecular meshwork surface by scanning electron microscopy to investigate possible artifacts appearing during collection and preparation of trabeculectomy specimens. As controls, trabecular tissue samples from normal and glaucomatous patients were dissected from whole eyes obtained at enucleation or at autopsy. Our results suggest that the major contaminants are released during decompression of the globe at trabeculectomy; these consist of an amorphous
653
substance, a fibrillar material, and, occasionally, red blood cells. Small tissue samples may also be contaminated by the conducting adhesive used to fasten them to metal stubs. This silver paint may coat the surface of the meshwork and appear to be a smooth substance blocking the outflow of aqueous humor. We were unable to confirm the finding reported by others that an amorphous blocking material is present in the trabecular interstices in patients with open-angle glaucoma. REFERENCES 1. Chaudhry, H. A., Dueker, D. K., Simmons, R J., Bellows, A. R, and Grant, W. M.: Scanning electron microscopy of trabeculectomy specimens in open-angle glaucoma. Am. J. Ophthalmol. 88:78, 1979. . 2. Sampaolesi, R, and Argento, C.: Scanning electron microscopy of the trabecular meshwork in normal and glaucomatous eyes. Invest. Ophthalmol. 16:302, 1977. 3. Bill, A., and Svedbergh, B.: Scanning electron microscopic studies of the trabecular meshwork and the canal of Schlemm. An attempt to localize the main resistance to outflow of aqueous humor in man. Acta Ophthalmol. 50:295, 1972. 4. Svedbergh, B.: Pretrabecular membrane in open-angle glaucoma, correspondence. Am. J. Ophthalmol. 88:1105, 1979. 5. Raviola, G.: Effects of paracentesis on the blood-aqueous barrier. An electron microscope study on Macaca mulatta using horseradish peroxidase as a tracer. Invest. Ophthalmol. 13:828, 1974. 6. Okisaka, S.: Effects of paracentesis on the blood-aqueous barrier. A light and electron microscopic study on eynomologus monkey. Invest. Ophthalmol. 15:824, 1976. 7. Bartels, S. P., Pederson, J. E., Gaasterland, D. E., and Armaly, M. F.: Sites of breakdown of the blood-aqueous barrier after paracentesis of the rhesus monkey eye. Invest. Ophthalmol. Vis. Sci. 18:1050, 1979. 8. Pederson, J. E., MacLellan, H. M., and Gaasterland, D. E.: The rate of reflux fluid movement into the eye from Schlemm's canal during hypotony in the rhesus monkey. Invest. Ophthalmol. Vis. Sci. 17:377, 1978. 9. Tso, M. O. M., and Shih, C. Y.: Experimental macular edema after lens extraction. Invest. Ophthalmol. Vis. Sci. 16:381, 1977. 10. Zimmerman, T. J., Gravenstein, N., Sugar, A., and Kaufman, H. E.: Aspirin stabilization of the blood-aqueous barrier in the human eye. Am. J. Ophthalmol. 79:817, 1975.
B.A., C. AND
McMAHON, M.D., D. HOSKINS, M.D., ALVARADO, M.D.
J.
San Francisco, California
The scanning electron microscope is a useful tool for studying the trabecular meshwork surface and abnormalities that may contribute to the malfunction of this tissue in glaucoma. 1-3 Trabeculectomy specimens are one important source of pathologic tissue, but these specimens require special care in order to prevent contamination of the tissue surface during collection and processing for scanning electron microscopy. While studying the trabecular meshwork from patients with primary openangle glaucoma, we became familiar with the scanning electron microscopic appearance of some potential problems or artifacts caused by methods of collecting and preparing trabecular tissues. In an attempt to identify the source of contamination, we studied the effects of decompression of the globe, bleeding, and other complications introduced by trabeculectomy, and compared the appearance of washed and unwashed trabeculectomy specimens. We used whole enucleated eyes or eyes obtained at autopsy to show that the meshwork in these eyes is not coated by the materials found in the unwashed trabeculectomy specimens. We examined glaucomatous trabecular tissues from whole eyes obtained at enucleation or autopsy to determine whether From the Department of Ophthalmology, University of California School of Medicine, San Francisco, California. This study was supported by grants EY-02068 and EY-02162A from the National Eye Institute (Dr. Alvarado). Reprint requests to J. Alvarado, M.D., U-490, University of California, School of Medicine, San Francisco, CA 94143.
the meshwork of such patients is obstructed by a blocking substance as reported by others. I We studied the scanning electron microscopic appearance of the substances that coat and apparently obstruct the trabecular interstices in trabeculectomy specimens. We interpreted these findings as artifacts caused by specimen collection or processing. These data do not confirm the reported finding that an "amorphous blocking substance" is found in the trabecular meshwork in primary open-angle glaucoma. I We offer an alternative explanation. MATERIAL AND METHODS
This study included 17 specimens of two types: surgical specimens obtained at trabeculectomy and whole eyes obtained at autopsy or enucleation. There were ten trabeculectomy specimens. The assistant surgeon washed five of them during collection by continuously irrigating the tissues with a balanced saline solution; the other five were collected without washing. For comparison, we studied seven whole eyes, including five eyes without glaucoma obtained at autopsy and two glaucomatous eyes. One glaucomatous eye was enucleated because of a posterior choroidal melanoma in a patient with open-angle glaucoma and exfoliation. Another globe, obtained at autopsy, had primary open-angle glaucoma. Originally it had been fixed in formalin. We submerged the seven whole eyes in fixative for one to two hours and dissected pie-shaped pieces that included the trabecular meshwork. Immediately after
AMERICAN JOURNAL OF OPHTHALMOLOGY 90:645-653, 1980
645
646
AMERICAN JOURNAL OF OPHTHALMOLOGY
excision, we immersed all of the specimens in a 2% glutaraldehyde and 1% paraformaldehyde solution at pH 7.4, buffered with 0.067 M of sodium cacodylate containing 5mM of calcium chloride. All the specimens were kept in the fixative solution overnight at 4°C; the next day they were transferred to a freshly prepared batch of the same fixative. We kept all the tissues in this fixative for two to four days and then rinsed them in a cold solution of 0.5 M of sucrose at pH 7.4, buffered with 0.1 M of sodium cacodylate. They were dehydrated in a series of alcohols increasing in concentration up to 100% ethyl alcohol; the specimens then were critical-point dried through carbon dioxide. We mounted the specimens on metal stubs in the usual manner, using a conducting silver adhesive, and coated them with approximately 15 nm of gold in a Hummer II sputter-coater. All the specimens were viewed with a scanning electron microscope at 20 kV and photographed with Polaroid 55 PIN film.
NOVEMBER, 1980 RESULTS
Studies using trabeculectomy specimens are limited by the difficulty of obtaining appropriate controls, since ideally these consist of trabeculectomy specimens from non glaucomatous patients. In an attempt to circumvent this problem, we compared washed to unwashed trabeculectomy specimens in order to study the presence of a tightly adhering blocking material in the meshwork. Additionally, we compared the scanning electron microscopic findings in nonglaucomatous and glaucomatous whole eyes (obtained at autopsy or enucleation) in regard to the presence of this same blocking substance. Finally, we examined the scanning electron microscopic appearance of probable artifacts introduced after tissue collection and specimen preparation. The trabecular meshwork of washed specimens was free of substances that might block the flow of aqueous through the uveal and corneoscleral meshwork (Fig. 1). The unwashed trabeculectomy
Fig. 1 (Maglio and associates). Washed trabeculectomy specimen. Left, Schwalbe's line (arrow). Uveoscleral and comeoscleral meshworks contain no amorphous material obstructing the interstices (x 2(0). Right, A higher-magnification view, showing meshwork interstices (x 1,500).
Fig. 2 (Maglio and associates). Unwashed trabeculectomy specimens. Top left, Amorphous substance within the trabecular spaces (x 5,(00). Top right, Fibrillar material forming a reticular pattern over the surface of the meshwork (x 1,(00). Bottom left, Numerous red blood cells covering a space in the meshwork (x 3,600).
647
648
AMERICAN JOURNAL OF OPHTHALMOLOGY
specimens had a number of substances lodged in the trabecular interstices (Fig. 2). One of these materials was amorphous (Fig. 2, top left) and closely resembled some of the materials shown by other investigators. 1 Another material was fibrillar in nature and looked like fibrin (Fig. 2, top right). These materials were observed in most unwashed specimens even when there had been no apparent bleeding during the trabeculectomy. When noticeable bleeding had occurred, they were present in abundance and were usually accompanied by red blood cells (Fig. 2, bottom left). Thus, these materials may have come from blood vessels and were readily removed from the trabeculectomy specimens by washing.
NOVEMBER, 1980
Figure 3 shows the appearance of the trabecular meshwork in one normal eye obtained at autopsy and Figure 4 shows that of an enucleated specimen from a patient with open-angle glaucoma and exfoliation. Neither specimen shows the presence of the materials observed in the unwashed trabeculectomy specimens. The only substance found in the trabecular interstices in the enucleated specimen consists of the exfoliation material (Fig. 4, top right and bottom left), which is formed by numerous fibrils gathered into a ball resembling a tumbleweed. Although we suspect that patients with the exfoliation syndrome have primary openangle glaucoma made worse by this obstructive exfoliation material, our comparison might be questioned by others
Fig. 3 (Maglio and associates). Normal autopsy specimen from the eye of a 28-year-old individual. Left, Endothelial surface of the cornea (a), uveoscleral meshwork (b), and iris surface (c) (x 1,500). Right, Side view of same specimen shows trifurcated Schlemm's canal (a), the sheets of the corneoscleral meshwork (b), and the uveoscleral meshwork (c). Note the open trabecular pores (arrows) (x 440).
Fig. 4 (Maglio and associates). Autopsy specimen from a patient with open-angle glaucoma and the exfoliation syndrome. Top left, Trabecular spaces are open and the only substance found consists of the fibrillar exfoliative material (arrows) (x 1,5(0). Top right, Higher-power view of exfoliative material (x 3,000). Bottom left, Trabecular cells with clear intercellular spaces; the nodular surface of these cells is presumably related to the presence of ingested melanin granules (x 1,100). Bottom right, Uveoscleral meshwork; there is a loosely adherent material of unknown nature but the meshwork has clear interstices (x 1,100).
649
650
AMERICAN JOURNAL OF OPHTHALMOLOGY
who consider this condition secondary open-angle glaucoma. Therefore, we examined by scanning electron microscopy another eye with primary open-angle glaucoma obtained at autopsy from the University of California Eye Pathology Laboratory, San Francisco. Figure 5 shows that there was no material blocking the meshwork of this specimen. Although we accounted for the presence of most of the blocking materials by the formation of plasma and other proteins during trabeculectomy, we still had not demonstrated the presence of another substance with a smooth appearance and many cracks on its surface found by other investigators. 1 Such cracks might have been the result of incomplete carbon dioxide drying or of the removal of
NOVEMBER, 1980
some volatile substance in the vacuum chamber of the scanning electron microscope. Acetone or butyl acetate are volatile substances present in the silver paint adhesive used to fasten the tissue specimens to metal stubs before viewing. When we liberally applied this adhesive to trabecular tissues similar in size to the typical trabeculectomy specimen (2 X 4 mm), the adhesive seeped into the hollow trabecular interstices, completely blocking them. Figure 6 shows the smooth appearance of this adhesive and the many cracks on its surface. With the aid of a dissecting microscope, small amounts of this adhesive can be applied, thus preventing contamination of the small pieces of tissue obtained at trabeculectomy.
Fig. 5 (Maglio and associates). Eye obtained at autopsy from a patient with primary open-angle glaucoma. Left, Despite formalin fixation, the usual three-dimensional architecture of the meshwork is evident (x 48). Right, Detail of uveoscleral meshwork showing that the trabecular interstices are not covered with amorphous material (x 1(0).
VOL. 90, NO. 5
TRABECULAR NETWORK MICROSCOPY
651
Fig. 6 (Maglio and associates). Contamination of the meshwork with the silver paint adhesive. Left, Surface of the trabecular cells appears to be coated with the adhesive; particles shown at arrows represent large deposits of silver. The cracks along the surface (double arrows) are produced by outgassing in the vacuum chamber (x 7,8(0). Right, A specimen contaminated by a light application of the adhesive. The surface, through which nodular structures in the trabecular cells are visible, is smooth. Note the minute cracks along a uveal band (x 2,000).
652
AMERICAN JOURNAL OF OPHTHALMOLOGY DISCUSSION
The appearance of the trabecular meshwork surface as shown by the scanning electron microscope varies considerably depending on the conditions at specimen collection and on the source of the tissue specimens. The surfaces of specimens obtained at trabeculectomy are often covered by an amorphous substance, a fibrillar material, and, occasionally, red blood cells. These materials can be easily removed in most cases by washing the specimens during collection. When whole eyes obtained at autopsy or enucleation are used as the source of tissue samples, the trabecular meshwork surface is free of these substances and washing is not necessary. We found this to be so in normal as well as glaucomatous specimens. These findings seem to support Svedbergh's" suggestion that abnormal substances can be released from the circulation during trabeculectomy and may be precipitated onto tissue surfaces. A recent series of studies has shown that there is a marked alteration of the various blood barriers with decompression of the globe in primate and human eyes.v" Haviola! was the first to show that plasma proteins reflux through Schlemm's canal and the trabecular meshwork into the anterior chamber after paracentesis. Okisaka" demonstrated that leakage of proteins into the posterior chamber also occurs with decompression of the globe, and that this is related to a breakdown of the blood-aqueous barrier localized to the anterior portion of the pars plicata. Bartels and associates" studied the sequence of events in the breakdown of the 'blood-aqueous barrier with paracentesis. They reported that protein leaks first into the posterior chamber and a few minutes later into the anterior chamber from Schlemrn's canal and the meshwork. Another important finding was the determination that the degree of
NOVEMBER, 1980
protein leakage is influenced more by the rapidity with which the intraocular pressure is decreased than the level to which it decreases.f To our knowledge, there have been no studies comparing the rate of pressure decrease at trabeculectomy with pressure decrease at paracentesis and other surgical procedures. Tso and Shih," however, reported that there was a breakdown of the blood-retina and blood-pigment epithelium barriers after trabeculectomy in monkeys. Possibly, most procedures in which the anterior chamber is entered produce a rapid decompression and much protein leakage. This is supported by the report of Zimmerman and associates'" who measured a significant degree of protein leakage into the anterior chamber in humans undergoing most anterior segment surgical procedures. We believe that plasma proteins, as well as other substances released by bleeding, constitute the major contaminants observed in our trabeculectomy specimens. The appearance of the trabecular meshwork after paracentesis and enucleation in monkeys has been studied by scanning electron microscopy." This report agrees with our findings, and both studies show that the meshwork of enucleated eyes is free of materials coating or blocking the trabecular meshwork. This is consistent with the idea that at trabeculectomy the intraocular pressure decreases much more rapidly than it does in enucleated eyes, and that this difference is an important factor in determining the degree of protein leakage. 8 We needed data from both enucleated eyes and trabeculectomy specimens in order to examine the provocative finding of an amorphous blocking substance in the trabecular meshwork associated with primary open-angle glaucoma.' Enucleated and autopsy specimens from patients with primary-open angle glaucoma are difficult to obtain. Nevertheless, we had
VOL. 90, NO. 5
TRABECULAR NETWORK MICROSCOPY
an opportunity to study eyes from three patients with primary open-angle glaucoma who had never undergone ocular surgery. We performed scanning electron microscopy in the two specimens reported here; the third specimen was studied only by transmission electron microscopy (I. Wood, unpublished data). Our scanning electron microscopy studies found no substances deposited over the trabecular tissues. A review of the transmission electromicrographs from the third specimen gave similar findings. The data from the trabeculectomy specimens and from eyes of patients with primary open-angle glaucoma suggest that the blocking material is released at the time of the trabeculectomy procedure as a result of the sudden decompression of the globe. We observed a blocking substance similar to that previously reported' only when the trabecular tissues were contaminated with the adhesive used in the preparation of the tissue samples for viewing by scanning electron microscope. Whether this accounts for the presence of a blocking material in the previous report! is not certain. We believe, however, that special care must be taken in the analysis of all trabeculectomy specimens because contamination often produces misleading effects. SUMMARY
We examined the trabecular meshwork surface by scanning electron microscopy to investigate possible artifacts appearing during collection and preparation of trabeculectomy specimens. As controls, trabecular tissue samples from normal and glaucomatous patients were dissected from whole eyes obtained at enucleation or at autopsy. Our results suggest that the major contaminants are released during decompression of the globe at trabeculectomy; these consist of an amorphous
653
substance, a fibrillar material, and, occasionally, red blood cells. Small tissue samples may also be contaminated by the conducting adhesive used to fasten them to metal stubs. This silver paint may coat the surface of the meshwork and appear to be a smooth substance blocking the outflow of aqueous humor. We were unable to confirm the finding reported by others that an amorphous blocking material is present in the trabecular interstices in patients with open-angle glaucoma. REFERENCES 1. Chaudhry, H. A., Dueker, D. K., Simmons, R J., Bellows, A. R, and Grant, W. M.: Scanning electron microscopy of trabeculectomy specimens in open-angle glaucoma. Am. J. Ophthalmol. 88:78, 1979. . 2. Sampaolesi, R, and Argento, C.: Scanning electron microscopy of the trabecular meshwork in normal and glaucomatous eyes. Invest. Ophthalmol. 16:302, 1977. 3. Bill, A., and Svedbergh, B.: Scanning electron microscopic studies of the trabecular meshwork and the canal of Schlemm. An attempt to localize the main resistance to outflow of aqueous humor in man. Acta Ophthalmol. 50:295, 1972. 4. Svedbergh, B.: Pretrabecular membrane in open-angle glaucoma, correspondence. Am. J. Ophthalmol. 88:1105, 1979. 5. Raviola, G.: Effects of paracentesis on the blood-aqueous barrier. An electron microscope study on Macaca mulatta using horseradish peroxidase as a tracer. Invest. Ophthalmol. 13:828, 1974. 6. Okisaka, S.: Effects of paracentesis on the blood-aqueous barrier. A light and electron microscopic study on eynomologus monkey. Invest. Ophthalmol. 15:824, 1976. 7. Bartels, S. P., Pederson, J. E., Gaasterland, D. E., and Armaly, M. F.: Sites of breakdown of the blood-aqueous barrier after paracentesis of the rhesus monkey eye. Invest. Ophthalmol. Vis. Sci. 18:1050, 1979. 8. Pederson, J. E., MacLellan, H. M., and Gaasterland, D. E.: The rate of reflux fluid movement into the eye from Schlemm's canal during hypotony in the rhesus monkey. Invest. Ophthalmol. Vis. Sci. 17:377, 1978. 9. Tso, M. O. M., and Shih, C. Y.: Experimental macular edema after lens extraction. Invest. Ophthalmol. Vis. Sci. 16:381, 1977. 10. Zimmerman, T. J., Gravenstein, N., Sugar, A., and Kaufman, H. E.: Aspirin stabilization of the blood-aqueous barrier in the human eye. Am. J. Ophthalmol. 79:817, 1975.