Documenta OphthalmoloNca 46,2:201-206, 1979 C E L L U L A R DENSITY OF NORMAL C O R N E A L ENDOTHELIUM O. LOPES CARDOZO
(Amsterdam) Key words: Cornea, Endothelium, Specular microscope. ABSTRACT With the aid of the specular microscope photographs were made of the endothelium of 72 subjects, 41 men and 31 women, aged from 24 to 86 years. The number of endothelial cells was counted by projecting the photographs and counting the cells in a Nven area. The average variation per eye was 4.9% with a standard deviation of 3.7. A difference hi magnification associated with differences in the thickness of the cornea was not found. There was a significant negative correlation (Pearson) between age and cell density: correlation coefficient 0.395, significance P = 0.001. Regression analysis of this correlation gave the following equation: CL = 3081 - 8.8 L, where CL = cell density at age L. The error is 97 cells more or less than this figure. The advent of the clinical specular microscope is a step forwards in the study of the physiology and pathology of the corneal endothelium. Although the instrument is technically far from perfect, With a tittle practice the examination can be carried out without worrying the patient too much. The information obtained has little morphological value as only the outlines of the cells are shown; the only morphological information concerns the shape of the cells, the size of the intercellular spaces and the presence or absence of small quantities of subcellular fluid. On the other hand the quantitative information, the number of cells per mm 2, has been found to be of great clinical importance. Studies by Laing (1976) and Bourne (1976) have shown that the cell density decreases with age. According to Van Horn (1975) this is because human endothelium no longer undergoes cell division so that surface defects are only filled up by increase in size of the cells. This process does n o t affect the capacity of the endothelial pumping function and the normal corneal thickness is maintained until a minimum number of endothelial cells is reached. This number lies between the 400 and 700 cells per m m 2 (Kaufman, personal communication). Further study shows that intraocular operations are associated with loss of endothelial cells. This is particularly true for artificial lens implantation, where the loss can be between 20 and 70% (Lance Forstot, 1977; Binklrorst, 1977). Since lens implantation is chiefly indicated in elderly patients, the question arises which endothelia are able to accept this great loss. In order
201
4000 cellenlmm 2 9
,~
.
i
3000,
2000,
100
1o
2' 0
3 'o
4' 0
s'o
6o '
70'
8'o
9o '
I~ftiid Fig. 1. Results of measurements on 144 eyes with the calculated regression line.
to answer this question it is necessary to know something about the distribution of the cell density over the normal population: this is the purpose of this study. METHODS A group of 72 subjects was examined, 41 men and 31 women, aged from 24 to 86 years, all without any history of keratitis, iritis, glaucoma, perforation of the eyeball or ocular surgery. The youngest age group was mainly recruited from personnel from the Out Patients' Department and colleagues, the middle group consisted of patients who came for reading glasses, while the oldest group was formed by relations accompanying patients to the Out Patients' Department 9 After the instillation of local anaesthetic five photographs were made of each eye, two of which were usually suitable. After being developed the films were projected onto a frosted glass screen, on which a field had been drawn; the cells crossed by the left side line and the bottom line of the field were counted and those crossed by the top line and the right side line were not.
202
The magnification was controlled by p h o t o g r a p h i n g an o b j e c t - m i c r o m e t e r under water w h e n the system was focused for different thicknesses of the cornea. This film was used to d e t e r m i n e the boundaries of the c o u n t i n g area; these were 0.36 x 0.03 m m . Two p h o t o g r a p h s of each eye were counted. An average c o u n t and variation per eye were calculated, on the basis of which an average n u m b e r of ceils per m m 2 was found. The average variation percentage and the variation percentage and the age were then determined. Regression analysis was p e r f o r m e d on the correlation b e t w e e n age and cell density. RESULTS No measurable difference in magnification was f o u n d within the range of corneal thicknesses in which the m e a s u r e m e n t s were p e r f o r m e d . The average variation percentage of the counts was 4.9%, with a standard deviation of 3.7. Significant negative correlation was f o u n d b e t w e e n age and cell density: correlation coefficient 0.395, sign ifican ce P = 0.001. No significant correlation was found b e t w e e n age and variation percentage. There was no significant difference b e t w e e n m e n and w o m e n or b e t w e e n right and left eyes.
Fig. 2. Regular densely cellular endothelium. Top: 3148 cells/ram 2, age 26 years. Bottom: 2917 ceUs/mm~ age 75 years.
203
Fig. 3. Irregular endothelium. Top: 3102 cells/mm2, age 24 years. Bottom: 2083 ceUs/mm 2, age 72 years.
Fig. 4. Top: regular endothelium. Bottom: polymorphic cells. Both from the same eye, 2407 cells/mm ~, age 83 years.
Regression analysis of the correlation b e t w e e n age and cell density showed that the n u m b e r of cells per m m 2 at a given age equals 3081 - 8.8 x the age, within a range of 97 cells m o r e or less. DISCUSSION The figures give an impression of the variation in a n o r m a l p o p u l a t i o n . It appears that a regular pattern of small endothelial cells can be f o u n d up to old age. There can also be a high degree of cell p o l y m o r p h i s m in b o t h densely cellular and sparsely cellular endothelia.
204
Our results are in complete agreement with those of Laing who finds, with a slightly more refined counting technique, a correlation coefficient of 0.444 and the same significance. Bourne also finds comparable values. The counting method used seems to be reasonably reliable. On account of the shape of the image a relatively large number of cells overlie the edge of the counting area. The error due to this increases as the cells become larger. We did not find a correlation, however, between age and variation percentage. This is probably because the polymorphism of the cells, which contributes to the variation in a similar manner, occurs with both great and small cellular density. The counting method used was not refined enough to provide particulars about the polymorphism. For this purpose it would be necessary to determine the surface area of a large number of cells. The standard deviation would then be a measure of the polymorphism. We hope this year to have an instrument at our disposal which measures and records this completely automatically. The intriguing aspect of the polymorphism is not so much the large cells, which spread out to fill up a lacuna, as the small cells. Are these cells with a large capacity for expansion, or just the opposite? In the former case, markedly polymorphic endothellum with many small cells will have a better prognosis for lens implantation than endothelium with uniform large cells. If, on the other hand, the polymorphism is an expression of the fact that the small cells have lost their capacity for regeneration, polymorphic endothelium would have a poorer prognosis. I hope to report further on this point in the future.
Fig. 5. Top: sparsely cellular but regular; 1990 cells/mm2, age 49 years. Bottom: densely cellular but extreme polymorphism; 3056 cells/mm2, age 41 years.
205
REFERENCES Binkhorst, C.D., L.H. Loones & P. Nygaard. The clinical specular microscope. Doc. Ophthalmoi. 44: 57- 75 (1977). Bourne, W.M., & H.E. Kaufman. Specular microscopy of human corneal endothelium in vivo. Amer. J. OphthalmoL 81:319 - 323 (1976). Laing, R.A., Sandstrom, M.M., Berrospi, A.R. & Leibowitz, H.M. Changes in the corneal endothelium as a function of age. Exp. Eye Res. 22:58 7- 594 (1976). Lance Forstot, S., Blackwell, W.L., Jaffe, N.S. & Kaufman, H.E. The effect of intraocular lens implantation on the corneal endothelium. Trans. Amer. Acad. Ophthai. & Otolaryngoi. 83:0P195-203 (1977). Horn, D.L. van & Hyndink, R.A. Experimental wound repair in primate cornea. Exp. EyeRes. 21:113-124 (1975). Author's address: Ophthalmological Department Vrij Universiteit Hospital, Amsterdam The Netherlands
206
(Amsterdam) Key words: Cornea, Endothelium, Specular microscope. ABSTRACT With the aid of the specular microscope photographs were made of the endothelium of 72 subjects, 41 men and 31 women, aged from 24 to 86 years. The number of endothelial cells was counted by projecting the photographs and counting the cells in a Nven area. The average variation per eye was 4.9% with a standard deviation of 3.7. A difference hi magnification associated with differences in the thickness of the cornea was not found. There was a significant negative correlation (Pearson) between age and cell density: correlation coefficient 0.395, significance P = 0.001. Regression analysis of this correlation gave the following equation: CL = 3081 - 8.8 L, where CL = cell density at age L. The error is 97 cells more or less than this figure. The advent of the clinical specular microscope is a step forwards in the study of the physiology and pathology of the corneal endothelium. Although the instrument is technically far from perfect, With a tittle practice the examination can be carried out without worrying the patient too much. The information obtained has little morphological value as only the outlines of the cells are shown; the only morphological information concerns the shape of the cells, the size of the intercellular spaces and the presence or absence of small quantities of subcellular fluid. On the other hand the quantitative information, the number of cells per mm 2, has been found to be of great clinical importance. Studies by Laing (1976) and Bourne (1976) have shown that the cell density decreases with age. According to Van Horn (1975) this is because human endothelium no longer undergoes cell division so that surface defects are only filled up by increase in size of the cells. This process does n o t affect the capacity of the endothelial pumping function and the normal corneal thickness is maintained until a minimum number of endothelial cells is reached. This number lies between the 400 and 700 cells per m m 2 (Kaufman, personal communication). Further study shows that intraocular operations are associated with loss of endothelial cells. This is particularly true for artificial lens implantation, where the loss can be between 20 and 70% (Lance Forstot, 1977; Binklrorst, 1977). Since lens implantation is chiefly indicated in elderly patients, the question arises which endothelia are able to accept this great loss. In order
201
4000 cellenlmm 2 9
,~
.
i
3000,
2000,
100
1o
2' 0
3 'o
4' 0
s'o
6o '
70'
8'o
9o '
I~ftiid Fig. 1. Results of measurements on 144 eyes with the calculated regression line.
to answer this question it is necessary to know something about the distribution of the cell density over the normal population: this is the purpose of this study. METHODS A group of 72 subjects was examined, 41 men and 31 women, aged from 24 to 86 years, all without any history of keratitis, iritis, glaucoma, perforation of the eyeball or ocular surgery. The youngest age group was mainly recruited from personnel from the Out Patients' Department and colleagues, the middle group consisted of patients who came for reading glasses, while the oldest group was formed by relations accompanying patients to the Out Patients' Department 9 After the instillation of local anaesthetic five photographs were made of each eye, two of which were usually suitable. After being developed the films were projected onto a frosted glass screen, on which a field had been drawn; the cells crossed by the left side line and the bottom line of the field were counted and those crossed by the top line and the right side line were not.
202
The magnification was controlled by p h o t o g r a p h i n g an o b j e c t - m i c r o m e t e r under water w h e n the system was focused for different thicknesses of the cornea. This film was used to d e t e r m i n e the boundaries of the c o u n t i n g area; these were 0.36 x 0.03 m m . Two p h o t o g r a p h s of each eye were counted. An average c o u n t and variation per eye were calculated, on the basis of which an average n u m b e r of ceils per m m 2 was found. The average variation percentage and the variation percentage and the age were then determined. Regression analysis was p e r f o r m e d on the correlation b e t w e e n age and cell density. RESULTS No measurable difference in magnification was f o u n d within the range of corneal thicknesses in which the m e a s u r e m e n t s were p e r f o r m e d . The average variation percentage of the counts was 4.9%, with a standard deviation of 3.7. Significant negative correlation was f o u n d b e t w e e n age and cell density: correlation coefficient 0.395, sign ifican ce P = 0.001. No significant correlation was found b e t w e e n age and variation percentage. There was no significant difference b e t w e e n m e n and w o m e n or b e t w e e n right and left eyes.
Fig. 2. Regular densely cellular endothelium. Top: 3148 cells/ram 2, age 26 years. Bottom: 2917 ceUs/mm~ age 75 years.
203
Fig. 3. Irregular endothelium. Top: 3102 cells/mm2, age 24 years. Bottom: 2083 ceUs/mm 2, age 72 years.
Fig. 4. Top: regular endothelium. Bottom: polymorphic cells. Both from the same eye, 2407 cells/mm ~, age 83 years.
Regression analysis of the correlation b e t w e e n age and cell density showed that the n u m b e r of cells per m m 2 at a given age equals 3081 - 8.8 x the age, within a range of 97 cells m o r e or less. DISCUSSION The figures give an impression of the variation in a n o r m a l p o p u l a t i o n . It appears that a regular pattern of small endothelial cells can be f o u n d up to old age. There can also be a high degree of cell p o l y m o r p h i s m in b o t h densely cellular and sparsely cellular endothelia.
204
Our results are in complete agreement with those of Laing who finds, with a slightly more refined counting technique, a correlation coefficient of 0.444 and the same significance. Bourne also finds comparable values. The counting method used seems to be reasonably reliable. On account of the shape of the image a relatively large number of cells overlie the edge of the counting area. The error due to this increases as the cells become larger. We did not find a correlation, however, between age and variation percentage. This is probably because the polymorphism of the cells, which contributes to the variation in a similar manner, occurs with both great and small cellular density. The counting method used was not refined enough to provide particulars about the polymorphism. For this purpose it would be necessary to determine the surface area of a large number of cells. The standard deviation would then be a measure of the polymorphism. We hope this year to have an instrument at our disposal which measures and records this completely automatically. The intriguing aspect of the polymorphism is not so much the large cells, which spread out to fill up a lacuna, as the small cells. Are these cells with a large capacity for expansion, or just the opposite? In the former case, markedly polymorphic endothellum with many small cells will have a better prognosis for lens implantation than endothelium with uniform large cells. If, on the other hand, the polymorphism is an expression of the fact that the small cells have lost their capacity for regeneration, polymorphic endothelium would have a poorer prognosis. I hope to report further on this point in the future.
Fig. 5. Top: sparsely cellular but regular; 1990 cells/mm2, age 49 years. Bottom: densely cellular but extreme polymorphism; 3056 cells/mm2, age 41 years.
205
REFERENCES Binkhorst, C.D., L.H. Loones & P. Nygaard. The clinical specular microscope. Doc. Ophthalmoi. 44: 57- 75 (1977). Bourne, W.M., & H.E. Kaufman. Specular microscopy of human corneal endothelium in vivo. Amer. J. OphthalmoL 81:319 - 323 (1976). Laing, R.A., Sandstrom, M.M., Berrospi, A.R. & Leibowitz, H.M. Changes in the corneal endothelium as a function of age. Exp. Eye Res. 22:58 7- 594 (1976). Lance Forstot, S., Blackwell, W.L., Jaffe, N.S. & Kaufman, H.E. The effect of intraocular lens implantation on the corneal endothelium. Trans. Amer. Acad. Ophthai. & Otolaryngoi. 83:0P195-203 (1977). Horn, D.L. van & Hyndink, R.A. Experimental wound repair in primate cornea. Exp. EyeRes. 21:113-124 (1975). Author's address: Ophthalmological Department Vrij Universiteit Hospital, Amsterdam The Netherlands
206
