- Email: [email protected]
Corneal Endothelial Changes in Primary Acute Angle-closure Glaucoma
Corneal Endothelial Changes Primary Acute Angle-closure Glaucoma
•
In
FRANCIS BIGAR, MD, RUDOLF WITMER, MD
Abstract: In 20 patients, corneal endothelium was examined by specular microscopy after acute angle-closure glaucoma, and before any surgical treatment in order to assess possible damage during pressure induced abnormal hydration of the cornea. The mean intraocular pressure was elevated to 55 mm Hg and had lasted, on an average, 47 (5-192) hours. The mean endothelial cell density in the affected eye was 1534 and in the nonaffected fellow eye 2243 cells/mm2 (mean decrease 33%, P = 0.002). The amount of cell loss correlates with the duration of the intraocular pressure increase. Thirty-five percent of these patients presented a bilateral cornea guttata. This high incidence of endothelial dystrophy was confirmed in a retrospective specular microscopic study in 20 patients with a history of unilateral acute angle-closure glaucoma. The decreased number of endothelial cells after acute angle-closure glaucoma frequently combined with bilateral cornea guttata, accounts for the corneal degeneration in these patients following a later cataract extraction. [Key words: corneal endothelium, cornea guttata, primary acute angle-closure glaucoma, specular microscopy.] Ophthalmology
89:596-599, 1982
Acute angle-closure glaucoma may lead to corneal damage, as well as to iris atrophy and changes of the lens. The corneal thickness may be increased and there may be striate keratopathy and epithelial edema. After a medically induced reduction of intraocular pressure, the corneal thickness returns to normal after a shorter or longer interval. An attack of acute glaucoma may be followed years later by a gradual From the Department of Ophthalmology, University of Zurich, Zurich, Switzerland. Presented at the Eighty-sixth Annual Meeting of the American Academy of Ophthalmology, Atlanta, Georgia, November 1-6,
1981.
Supported by the Albrecht Bruppacher Foundation of the Zurich Eye Bank. Reprint requests to Francis Bigar, MD, Department of Ophthalmology, University of Zurich, 100, Ramistrasse, CH-8091 Zurich, Switzerland.
596
onset of a corneal degeneration; this may be accelerated by a cataract extraction. To what an extent an attack of acute glaucoma damages the corneal endothelium has therefore been a question of interest.
MATERIALS AND METHODS The central corneal endothelium of both eyes of 20 patients was examined and photographed with a Leitz Biophthal contact specular microscope after successful medical treatment of their first attack of primary acute angle-closure glaucoma, as soon as the cornea had cleared up and before performing any surgery. The specular photomicrographs of each eye were analyzed with a semiautomatic image analyzer Leitz-ASM. The mean number of endothelial cells measured in the affected eye was 126 (range 54- 216) and in the nonaffected eye 171 (range 57-255). Seven patients were 0161-6420/82/0600/596/$00.70
© American
Academy of Ophthalmology
BIGAR AND WITMER • ACUTE GLAUCOMA
Table 1. Corneal Endothelial Cell Loss in 20 Patients with Primary Acute Angle-Closure Glaucoma Cell/mm 2
Mean Range Standard deviation
Affected Eye
Fellow Eye
1534 239-2701
2243 1152-3058
818
401
Difference 33%
men, 13 were women. The mean age of the patients was 74 years (range 54-90). Thirteen of the patients were hypermetropes, with a mean refractive error of +3 diopters (range + 1.0 to +6.0 diopters). The mean increase in intraocular pressure at admission was 55 mrn Hg (range 32-70). From the onset of symptoms (ocular pain, decrease in visual acuity), the pressure was normalized within an average of 47 hours (5-192). In the mean the cornea cleared up within 4.8 days (range 2-12 days) sufficiently so that specular photomicrography could be performed before any surgery. Statistical analysis was made for different variables. In a separate retrospective specular microscopic study, 20 patients with a history of unilateral acute angle-closure glaucoma were examined to check the incidence of cornea guttata. More than 12 central nonconfluent excrescences had to be present to be considered a cornea guttata.
RESULTS The mean central endothelial cell density was 1534 (range 239-2701, SD 818) per mm2 in the affected eye, and 2243 cells/mm2 (range 1152- 3058, SD 401) in the normotensive fellow eye (Table 1). The mean difference in the cell count is 709/mm 2 (range + 101 down to -2297) or 33% (range +4 down to -91) which is statistically highly significant (Kolmogorov-Smirnov
cellloss/mm 2
•
2000
• 1000
••••••
••
0 0
•
• •
• •
•• SO
150 100 200 duration oT;;ff;;ck (hrs)
Fig 1. Endothelial cell loss plotted against the duration of acute angle-closure glaucoma attack.
test, P = 0.002). The amount of cell loss for all 20 patients correlates with the duration of the intraocular pressure increase (Kendall's correlation coefficient r = 0.40, P = 0.004) (Fig 1). In six patients there was a cell count below 800/mm2 • In four of these patients, medical therapy waS not started or not successful until four to eight days after the onset of symptoms. In seven (35%) of these patients dark, nonreflective, round- to oval-shaped areas, typical for cornea grtttata, were found in both eyes within the endothelial mosaic (Fig 2). There was no difference in the amount of cell loss between eyes with and without cornea guttata (Kolmogorov-Smirnov test, P = 0.29). In the retrospective study of 20 additional patients with a previous attack of unilateral acute angle-closure glaucoma, six (30%) were found to have a bilateral cornea guttata. These corneas with low cell counts showed an even cell size distribution over the whole posterior corneal surface. All corneas were clear and had a normal thickness.
DISCUSSION The acute increase in intraocular pressure in acute angle-closure glaucoma leads to massive damage of the corneal endothelium. It has been shown that there is no significant difference of the central endothelial cell density in all age groups of a normal population between pairs of eyes. 1 ,2 This fact allows one to draw retrospective conclusions on the extent of cell damage in a unilateral condition using the unaffected and presumably normal, fellow eye as a control. Olsen found in 23 patients a cell loss of 23% without indicating the length of the glaucoma attack. 3 Setala4 reported a difference of the endothelial cell count between the affected and the fellow eye of 10% in 25 patients after an acute glaucoma of a mean duration of 2.5 days. In our present study, the mean cell loss was 33% after a similar length of time. In four of the 20 patients, the attack lasted between four and 12 days. The cell loss in this group averaged 77%, and the mean cell count was reduced to 448 cells/mm2 (Figs 3, 4). The longer the attack of acute glaucoma lasts the greater the endothelial cell loss is . In contrast, endothelial cell densities in far advanced open-angle glaucoma with fully excavated discs, reduced visual acuities, and visual fields never reached such a low level. 1 They remained as in unilateral capsular glaucoma and in glaucornato-cyclitic crises (PosnerSchlossmann) within the large range of individual variation of the normal population. 5 ,6 Finally, the reduced cell count in congenital glaucoma was in accordance with the increase of the estimated enlargement of the posterior corneal surface. 1 Therefore, it can be assumed that the corneal endothelium tolerates a chronic increase in intraocular pressure much better than an attack of angle-closure glaucoma. This endothelial damage is due most probably to hypoxia resulting from the impaired flow of aqueous humor in acute 597
OPHTHALMOLOGY. JUNE 1982 • VOLUME 89 • NUMBER 6
Fig 2. Cornea guttata in the nonaffected fellow eye of a 74-year-old patient with acute angle-closure glaucoma: round to oval dark areas correspond to the base of wart-like excrescences on Descemet's membrane. Bar = 100 /Lm.
Fig 3. Enlargement of endothelial cells after longlasting acute angle-closure glaucoma (144 hours) in a 61-year-old woman. Photomicrography was performed before iridectomy on the eighth day after the attack. Top, normal endothelium in the fellow eye 2000 cells/ mm 2 • Bottom, enlarged endothelial cells with pleomorphic forms and some curved cell borders with abnormal cell intersections, 700 cells/ mm 2 • Bar = 100 /Lm.
glaucoma. This assumption is supported by Sperling's experimental findings in human donor eyes that high pressure under normal oxygenation does not affect the endothelial cell density. 7 A coincidental, and so far unreported, finding is the frequent incidence of bilateral cornea guttata in one third of the patients with acute angle-closure glaucoma, whereas the incidence of this endothelial
as.
o.D. 2044
730
me.m celldensity/mm 2
30
dystrophy is given in the literature as between 4 and 10%.1,8-10 The attack of acute glaucoma, the peripheral iridectomy, and the age of patient contribute to cataract formation. 1o ,11 These patients then require a cataract extraction. With an already irreversibly decreased cell count, frequently associated with bilateral cornea guttata, the operation constitutes a risk to the maintenance of corneal transparency. The additional trauma of cataract extraction may lead to immediate or late endothelial decompensation and corneal edema in spite of perfect technique and a smoothly performed operation. Specular microscopy has thus brought new understanding of a possible entity in which endothelial damage due to angle-closure glaucoma together with a frequent bilateral endothelial dystrophy accounts for the corneal degeneration in these patients following a later cataract extraction.
ACKNOWLEDGMENTS eel/area
o
1000
2000
3000
,#m2
Fig 4. Histogram of patient of Figure 3. Black columns: healthy fellow eye with a normal frequency distribution curve of cell areas. Empty columns: flat frequency distribution after acute glaucoma. The histograms of both eyes show a significantly different distribution (Kolmogorov-Smirnov test P < 0.01).
598
The authors thank Miss Ruth Hiirzeler for her constant excellent technical assistance and Dr. P. Rejrnan for the statistical work-up.
REFERENCES 1. Bigar F. Specular microscopy of the corneal endothelium. Optical solutions and clinical results. Dev Ophthalmol 1982; 6:
1-94.
BIGAR AND WITMER • ACUTE GLAUCOMA
2. Sturrock GO, Sherrard ES, Rice NSC. Specular microscopy of the corneal endothelium. Br J Ophthalmol 1978; 62:809-14. 3. Olsen T. The endothelial cell damage in acute glaucoma. On the corneal thickness response to intraocular pressure. Acta Ophthalmol 1980; 58:257-66. 4. Setala K. Corneal endothelial cell density after an attack of acute glaucoma. Acta Ophthalmol 1979; 57:1004-13. 5. Vannas A, Setiilii K, Ruusuvaara P. Endothelial cells in capsular glaucoma. Acta Ophthalmol 1977; 55:951-8. 6. Setala K, Vannas A. Endothelial cells in the glaucomatocyclitic crisis. Adv Ophthalmol 1978; 36:218-24. 7. Sperling S. Human corneal endothelial cell density after an in vitro imitation of elevated intraocular pressure In preparation,
1981.
8. Goar EL. Dystrophy of the corneal endothelium (cornea guttata) with a report of a histological examination. Am. J. Ophthalmol.
1934; 17:215- 20. 9. Moeschler H. Untersuchungen tiber Pigmentierung der Hornhautruckflache bei 395 am Spaltlampenmikroskop untersuchten Augen gesunder Personen. Z Augenhei Ik 1922;
48:195-202. 10. Godel V, Regenbogen L. Cataractogenic factors in patients with primary angle-closure glaucoma after peripheral iridectomy. Am J Ophthalmol 1977; 83:180-4.
11 . Krupin T, Mitchell KB, Johnson MF, Becker B. The long-term effects of iridectomy for primary acute angle -closure glaucoma Am J Ophthalmol 1978; 86:506 - 9
599
•
In
FRANCIS BIGAR, MD, RUDOLF WITMER, MD
Abstract: In 20 patients, corneal endothelium was examined by specular microscopy after acute angle-closure glaucoma, and before any surgical treatment in order to assess possible damage during pressure induced abnormal hydration of the cornea. The mean intraocular pressure was elevated to 55 mm Hg and had lasted, on an average, 47 (5-192) hours. The mean endothelial cell density in the affected eye was 1534 and in the nonaffected fellow eye 2243 cells/mm2 (mean decrease 33%, P = 0.002). The amount of cell loss correlates with the duration of the intraocular pressure increase. Thirty-five percent of these patients presented a bilateral cornea guttata. This high incidence of endothelial dystrophy was confirmed in a retrospective specular microscopic study in 20 patients with a history of unilateral acute angle-closure glaucoma. The decreased number of endothelial cells after acute angle-closure glaucoma frequently combined with bilateral cornea guttata, accounts for the corneal degeneration in these patients following a later cataract extraction. [Key words: corneal endothelium, cornea guttata, primary acute angle-closure glaucoma, specular microscopy.] Ophthalmology
89:596-599, 1982
Acute angle-closure glaucoma may lead to corneal damage, as well as to iris atrophy and changes of the lens. The corneal thickness may be increased and there may be striate keratopathy and epithelial edema. After a medically induced reduction of intraocular pressure, the corneal thickness returns to normal after a shorter or longer interval. An attack of acute glaucoma may be followed years later by a gradual From the Department of Ophthalmology, University of Zurich, Zurich, Switzerland. Presented at the Eighty-sixth Annual Meeting of the American Academy of Ophthalmology, Atlanta, Georgia, November 1-6,
1981.
Supported by the Albrecht Bruppacher Foundation of the Zurich Eye Bank. Reprint requests to Francis Bigar, MD, Department of Ophthalmology, University of Zurich, 100, Ramistrasse, CH-8091 Zurich, Switzerland.
596
onset of a corneal degeneration; this may be accelerated by a cataract extraction. To what an extent an attack of acute glaucoma damages the corneal endothelium has therefore been a question of interest.
MATERIALS AND METHODS The central corneal endothelium of both eyes of 20 patients was examined and photographed with a Leitz Biophthal contact specular microscope after successful medical treatment of their first attack of primary acute angle-closure glaucoma, as soon as the cornea had cleared up and before performing any surgery. The specular photomicrographs of each eye were analyzed with a semiautomatic image analyzer Leitz-ASM. The mean number of endothelial cells measured in the affected eye was 126 (range 54- 216) and in the nonaffected eye 171 (range 57-255). Seven patients were 0161-6420/82/0600/596/$00.70
© American
Academy of Ophthalmology
BIGAR AND WITMER • ACUTE GLAUCOMA
Table 1. Corneal Endothelial Cell Loss in 20 Patients with Primary Acute Angle-Closure Glaucoma Cell/mm 2
Mean Range Standard deviation
Affected Eye
Fellow Eye
1534 239-2701
2243 1152-3058
818
401
Difference 33%
men, 13 were women. The mean age of the patients was 74 years (range 54-90). Thirteen of the patients were hypermetropes, with a mean refractive error of +3 diopters (range + 1.0 to +6.0 diopters). The mean increase in intraocular pressure at admission was 55 mrn Hg (range 32-70). From the onset of symptoms (ocular pain, decrease in visual acuity), the pressure was normalized within an average of 47 hours (5-192). In the mean the cornea cleared up within 4.8 days (range 2-12 days) sufficiently so that specular photomicrography could be performed before any surgery. Statistical analysis was made for different variables. In a separate retrospective specular microscopic study, 20 patients with a history of unilateral acute angle-closure glaucoma were examined to check the incidence of cornea guttata. More than 12 central nonconfluent excrescences had to be present to be considered a cornea guttata.
RESULTS The mean central endothelial cell density was 1534 (range 239-2701, SD 818) per mm2 in the affected eye, and 2243 cells/mm2 (range 1152- 3058, SD 401) in the normotensive fellow eye (Table 1). The mean difference in the cell count is 709/mm 2 (range + 101 down to -2297) or 33% (range +4 down to -91) which is statistically highly significant (Kolmogorov-Smirnov
cellloss/mm 2
•
2000
• 1000
••••••
••
0 0
•
• •
• •
•• SO
150 100 200 duration oT;;ff;;ck (hrs)
Fig 1. Endothelial cell loss plotted against the duration of acute angle-closure glaucoma attack.
test, P = 0.002). The amount of cell loss for all 20 patients correlates with the duration of the intraocular pressure increase (Kendall's correlation coefficient r = 0.40, P = 0.004) (Fig 1). In six patients there was a cell count below 800/mm2 • In four of these patients, medical therapy waS not started or not successful until four to eight days after the onset of symptoms. In seven (35%) of these patients dark, nonreflective, round- to oval-shaped areas, typical for cornea grtttata, were found in both eyes within the endothelial mosaic (Fig 2). There was no difference in the amount of cell loss between eyes with and without cornea guttata (Kolmogorov-Smirnov test, P = 0.29). In the retrospective study of 20 additional patients with a previous attack of unilateral acute angle-closure glaucoma, six (30%) were found to have a bilateral cornea guttata. These corneas with low cell counts showed an even cell size distribution over the whole posterior corneal surface. All corneas were clear and had a normal thickness.
DISCUSSION The acute increase in intraocular pressure in acute angle-closure glaucoma leads to massive damage of the corneal endothelium. It has been shown that there is no significant difference of the central endothelial cell density in all age groups of a normal population between pairs of eyes. 1 ,2 This fact allows one to draw retrospective conclusions on the extent of cell damage in a unilateral condition using the unaffected and presumably normal, fellow eye as a control. Olsen found in 23 patients a cell loss of 23% without indicating the length of the glaucoma attack. 3 Setala4 reported a difference of the endothelial cell count between the affected and the fellow eye of 10% in 25 patients after an acute glaucoma of a mean duration of 2.5 days. In our present study, the mean cell loss was 33% after a similar length of time. In four of the 20 patients, the attack lasted between four and 12 days. The cell loss in this group averaged 77%, and the mean cell count was reduced to 448 cells/mm2 (Figs 3, 4). The longer the attack of acute glaucoma lasts the greater the endothelial cell loss is . In contrast, endothelial cell densities in far advanced open-angle glaucoma with fully excavated discs, reduced visual acuities, and visual fields never reached such a low level. 1 They remained as in unilateral capsular glaucoma and in glaucornato-cyclitic crises (PosnerSchlossmann) within the large range of individual variation of the normal population. 5 ,6 Finally, the reduced cell count in congenital glaucoma was in accordance with the increase of the estimated enlargement of the posterior corneal surface. 1 Therefore, it can be assumed that the corneal endothelium tolerates a chronic increase in intraocular pressure much better than an attack of angle-closure glaucoma. This endothelial damage is due most probably to hypoxia resulting from the impaired flow of aqueous humor in acute 597
OPHTHALMOLOGY. JUNE 1982 • VOLUME 89 • NUMBER 6
Fig 2. Cornea guttata in the nonaffected fellow eye of a 74-year-old patient with acute angle-closure glaucoma: round to oval dark areas correspond to the base of wart-like excrescences on Descemet's membrane. Bar = 100 /Lm.
Fig 3. Enlargement of endothelial cells after longlasting acute angle-closure glaucoma (144 hours) in a 61-year-old woman. Photomicrography was performed before iridectomy on the eighth day after the attack. Top, normal endothelium in the fellow eye 2000 cells/ mm 2 • Bottom, enlarged endothelial cells with pleomorphic forms and some curved cell borders with abnormal cell intersections, 700 cells/ mm 2 • Bar = 100 /Lm.
glaucoma. This assumption is supported by Sperling's experimental findings in human donor eyes that high pressure under normal oxygenation does not affect the endothelial cell density. 7 A coincidental, and so far unreported, finding is the frequent incidence of bilateral cornea guttata in one third of the patients with acute angle-closure glaucoma, whereas the incidence of this endothelial
as.
o.D. 2044
730
me.m celldensity/mm 2
30
dystrophy is given in the literature as between 4 and 10%.1,8-10 The attack of acute glaucoma, the peripheral iridectomy, and the age of patient contribute to cataract formation. 1o ,11 These patients then require a cataract extraction. With an already irreversibly decreased cell count, frequently associated with bilateral cornea guttata, the operation constitutes a risk to the maintenance of corneal transparency. The additional trauma of cataract extraction may lead to immediate or late endothelial decompensation and corneal edema in spite of perfect technique and a smoothly performed operation. Specular microscopy has thus brought new understanding of a possible entity in which endothelial damage due to angle-closure glaucoma together with a frequent bilateral endothelial dystrophy accounts for the corneal degeneration in these patients following a later cataract extraction.
ACKNOWLEDGMENTS eel/area
o
1000
2000
3000
,#m2
Fig 4. Histogram of patient of Figure 3. Black columns: healthy fellow eye with a normal frequency distribution curve of cell areas. Empty columns: flat frequency distribution after acute glaucoma. The histograms of both eyes show a significantly different distribution (Kolmogorov-Smirnov test P < 0.01).
598
The authors thank Miss Ruth Hiirzeler for her constant excellent technical assistance and Dr. P. Rejrnan for the statistical work-up.
REFERENCES 1. Bigar F. Specular microscopy of the corneal endothelium. Optical solutions and clinical results. Dev Ophthalmol 1982; 6:
1-94.
BIGAR AND WITMER • ACUTE GLAUCOMA
2. Sturrock GO, Sherrard ES, Rice NSC. Specular microscopy of the corneal endothelium. Br J Ophthalmol 1978; 62:809-14. 3. Olsen T. The endothelial cell damage in acute glaucoma. On the corneal thickness response to intraocular pressure. Acta Ophthalmol 1980; 58:257-66. 4. Setala K. Corneal endothelial cell density after an attack of acute glaucoma. Acta Ophthalmol 1979; 57:1004-13. 5. Vannas A, Setiilii K, Ruusuvaara P. Endothelial cells in capsular glaucoma. Acta Ophthalmol 1977; 55:951-8. 6. Setala K, Vannas A. Endothelial cells in the glaucomatocyclitic crisis. Adv Ophthalmol 1978; 36:218-24. 7. Sperling S. Human corneal endothelial cell density after an in vitro imitation of elevated intraocular pressure In preparation,
1981.
8. Goar EL. Dystrophy of the corneal endothelium (cornea guttata) with a report of a histological examination. Am. J. Ophthalmol.
1934; 17:215- 20. 9. Moeschler H. Untersuchungen tiber Pigmentierung der Hornhautruckflache bei 395 am Spaltlampenmikroskop untersuchten Augen gesunder Personen. Z Augenhei Ik 1922;
48:195-202. 10. Godel V, Regenbogen L. Cataractogenic factors in patients with primary angle-closure glaucoma after peripheral iridectomy. Am J Ophthalmol 1977; 83:180-4.
11 . Krupin T, Mitchell KB, Johnson MF, Becker B. The long-term effects of iridectomy for primary acute angle -closure glaucoma Am J Ophthalmol 1978; 86:506 - 9
599