- Email: [email protected]
Cataract Extraction and the Corneal Endothelium
CATARACT EXTRACTION AND T H E CORNEAL E N D O T H E L I U M W I L L I A M M. B O U R N E , M.D.,
AND H E R B E R T E. K A U F M A N ,
M.D.
Gainesville, Florida
Bullous keratopathy develops in a sig nificant percentage of eyes after cataract extraction. Most of these cases are thought to be due to corneal endothelial damage occurring during the cataract ex traction, postoperative vitreous contact, or preexisting Fuchs' dystrophy. Until recently, however, inadequate visualiza tion of the corneal endothelium has pre vented the precise evaluation of endothe lial damage. With the clinical specular microscope, 1 the endothelium can be ex amined and photographed in a patient before and after the procedure, allowing objective estimates of cell loss. In this study we examined the corneal endotheli um of 16 patients, before and at intervals after cataract extraction, to determine the measurable endothelial cell loss associ ated with complicated and uncomplicat ed cataract extraction.
neas, two had clear penetrating keratoplasties, and one had a clear cornea fol lowing blunt trauma. Intracapsular cryoextraction was performed on 12 patients, two of whom had surgical or postsurgical complications: one had vitreous loss and the other had severe postoperative in flammation with sterile hypopyon that cleared after intensive corticosteroid ther apy. Four younger patients with soft cata racts underwent phacoemulsification pro cedures, and the actual ultrasonification time was minimal. All the operations were performed by staff and residents here during a four-month period. The preoperative endothelial photographs were generally taken on the day before extraction, and the first postoperative ex aminations were usually taken the day after the operation. Several were taken a week or more later. Follow-up photo graphs were taken every few weeks there after, the longest period of study being 14 weeks. The corneal epithelium was intact at all examinations.
M A T E R I A L AND M E T H O D S
We examined and photographed the central corneal endothelium of 16 pa tients with the clinical specular micro scope before and at intervals after cataract extraction. The central endothelial cell density was determined for each cornea as previously described. 1 The central cor neal thickness was measured at each ex amination by the same investigator with a modified Haag-Streit pachometer. 2 The patients' eyes had not been patched for several hours before the thickness read ings. Thirteen patients had normal cor-
RESULTS
The corneas of 13 patients were normal by specular microscopy, and three (Cases 1, 13, and 16) were abnormal before sur gery (Table). These three corneas had abnormally low endothelial cell densities, although no cornea guttata or other changes were seen. Four patients (Cases 10 to 13) had a statistically significant decrease in endo thelial cell density after cataract extrac tion (Student's f-test, P<.05), although all 16 corneas remained clear postoperatively. Two of these four patients had a significant loss of endothelial cells during the ensuing weeks, although the corneas remained clear (analysis of variance with regression, P<.05). These were the same
From the Department of Ophthalmology, College of Medicine, University of Florida, Gainesville, Florida. This study was supported by Public Health Service grants EY-00033-12 and EY-00266-12, Na tional Institutes of Health. Reprint requests to William Μ. Bourne, M.D., Department of Ophthalmology, Mayo Clinic, Roch ester, MN 55901. 44
VOL. 82, NO. 1
ENDOTHELIUM AND CATARACT EXTRACTION
45
TABLE CATARACT EXTRACTIONS*
Case No., Age, yrs
Cause of Cataract
1,47 2,78 3,78 4,75 5,64 6,55 7,75 8,57 9,80 10,57 11,73 12,51 13,25
PK (Fuchs') Senile Senile Senile Senile Senile Senile Senile Senile Senile Senile Senile PK (keratoconus)
14,29
Diabetes
15,46
Senile?
16, 16
Blunt trauma
Procedure ICCE ICCE ICCE ICCE ICCE ICCE ICCE ICCE ICCE ICCE ICCE ICCE Phacoemulsification Phacoemulsifi cation Phacoemulsiflcation Phacoemulsiflcation
Endothelial Cell Density, cells/mm2 Pre Post operative operative 467
Corneal Swelling, mmf
Complications
413
0.04
None
2,369 2,963 2,397 2,660 2,984 2,459 2,700 2,234 2,296 2,860 2,366 1,103
2,771 2,570 2,307 2,636 2,873 2,105 2,395 2,060 1,423} 1,415} l,679t 687}
0 0.04 0.03 0.04 0.04 0.04 0.04 0.03 0.07 0.04 0.06 0.06
None None None None None None None None Iridocyclitis Vitreous loss None None
2,211
2,368
0.04
None
2,650
2,615
0.02
None
1,498
1,582
0.03
None
*ICCE indicates intracapsular cryoextraction, and PK, penetrating keratoplasty. t Increase in central corneal thickness at first postoperative measurement. } Statistically different from the preoperative value (P<.05).
two patients who suffered surgical or postsurgical complications (Cases 10 and 11) (Figure). One patient (Case 12) did not have further significant endothelial cell loss, and the fourth patient (Case 13) had a statistically significant increase in endothelial cell density during the ensu ing weeks, so that six weeks after cataract extraction, the endothelial cell density had returned to preoperative levels. The remaining 12 patients showed no signifi cant change in endothelial cell density during the entire period of study (P>.05). Fifteen patients had an increase in cen tral corneal thickness on the first exami nation after cataract extraction. Most thicknesses eventually returned to their preoperative levels. However, the four corneas that had a significant loss of en dothelial cells (Cases 10 to 13) also had a
statistically greater mean increase in cor neal thickness when compared to the re maining 12 corneas (f-test, P<.002). DISCUSSION
No presently detectable loss of central endothelial cells occurred in most of the uncomplicated intracapsular cataract ex tractions. Cell loss is more likely to occur if there are operative or postoperative complications such as vitreous loss or severe iridocyclitis. T w o patients (Cases 10 and 11) contin ued to lose central endothelial cells for weeks after cataract extraction. One of these patients had severe postoperative iridocyclitis, which seems to be responsi ble for the continued cell loss. The other patient had no demonstrable abnormal postoperative inflammation, although he
46
AMERICAN JOURNAL OF OPHTHALMOLOGY
JULY, 1976
Figure (Bourne and Kaufman). Case 10. The central endothelium before (top) and seven weeks after (bottom) intracapsular cataract extraction. A severe postoperative, sterile iridocyclitis with hypopyon occurred and cleared with intensive corticosteroid therapy.
had suffered vitreous loss. Undetected inflammation may have been present in this patient. A further decrease in the number of central endothelial cells may be due to spreading of cells to the periph eral cornea to cover areas of cell loss, for example, at the corneoscleral incision site. A significant increase in endothelial cell density occurred after the initial decrease in a 25-year-old man with an 11-month-old clear penetrating corneal transplant (Case 13). The cataract extrac tion was done by phacoemulsification. The increased number of central endothe lial cells presumably is the result of cell division or sliding of cells from the pe riphery of the graft or recipient. Healing by sliding is the more likely possibility, since this occurred in corneal transplants in primates 3 and central endothelial cells do not normally maintain their number with time by cell division. 1 To our knowl edge, this is the first documentation of an increase in the number of central endo thelial cells in humans.
The small transient increase in corneal thickness that occurred in the 12 patients without demonstrable endothelial cell loss was statistically significant (i-test, P<.02). This apparently represents a tem porary decrease in endothelial function, presumably due to operative manipula tion. Increases in corneal thickness after cataract extraction have been reported. 4 - 8 A greater transient mean increase in cor neal thickness was demonstrated in our four patients with significant endothelial cell loss, indicating a more severe, al though temporary, disturbance in endo thelial function in these patients. These findings indicate that in the absence of epithelial ulcération, stromal scarring, or elevated intraocular pressure, corneal thickness is an approximate measure of endothelial function, and the larger the postoperative increase in thickness, the more likely it is that endothelial cell loss occurred during the procedure. Also, no loss of endothelial cells was demonstrated in three of four phacoemul sification procedures (extracapsular).
VOL. 82, NO. 1
ENDOTHELIUM AND CATARACT EXTRACTION
They were all soft cataracts in young adults (average age, 29 years), so that this procedure appears to be safe for this group of patients. Two of the four corneas were abnormal before the phacoemulsification procedures (Cases 13 and 16), and the one that lost endothelial cells was a corneal transplant (Case 13). Although significant endothelial cell loss was demonstrated in four of the pa tients, all 16 corneas remained clear dur ing the period of observation. A long-term study will be needed to determine if the cell loss is clinically detrimental. It is clear that extremely low endothelial cell densities maintain clear corneal trans plants for years. 9 However, it seems rea sonable to assume that the corneas which lost many endothelial cells are more sus ceptible to the detrimental effects of in flammation, trauma, or aging. We would like, therefore, to minimize the endothe lial damage from our surgical procedures. This study demonstrates an excellent method for documenting this damage and perhaps for more accurately predicting those corneas at greatest risk before sur gery. SUMMARY
We examined and photographed the central corneal endothelium of 16 pa tients with the clinical specular micro scope before and at intervals after cataract extraction. No detectable loss of endothe lial cells occurred in 75% of the patients, including 12 routine intracapsular cryoextractions and four phacoemulsifications of soft cataracts in young adults. Only one of the four cases of significant endothelial cell loss occurred in a normal
47
cornea without demonstrable operative or postoperative complications. Two of the four corneas that lost central endothelial cells at cataract extraction continued to lose more cells during the ensuing weeks. A significant increase in central endothe lial cell density was demonstrated in one patient. The four corneas with endotheli al cell loss also had a significantly higher mean increase in corneal thickness postoperatively, although this was transient. All 16 corneas remained clear during the period of observation (maximum, 14 weeks), and long-term studies are needed to measure the chronic effects of the en dothelial damage. REFERENCES 1. Bourne, W. M., and Kaufman, H. E.: Specular microscopy of human corneal endothelium in vivo. Am. J. Ophthalmol. 81:319, 1976. 2. Mishima, S., and Hedbys, B. O.: Measurement of corneal thickness with the Haag-Streit pachometer. Arch. Ophthalmol. 80:710, 1968. 3. Bourne, W. M.: In vivo survival of cryopreserved endothelial cells in primates. Arch. Ophthal mol. 92:146, 1974. 4. Giardini, A., and Cambiaggi, A.: Recherches sur l'é paisseur cornèenne après extraction de la cataracte. Ophthalmologica 131:41, 1956. 5. Lavergne, G., and Kelecom, J.: Applications cliniques de la mesure de l'é paisseur de la corneè. Bull. Soc. Belge Ophtalmol. 131:323, 1962. 6. Miller, D., and Dohlman, C. H.: Effect of cataract surgery on the cornea. Trans. Am. Acad. Ophthalmol. Otolaryngol. 74:369, 1970. 7. Norn, M. S.: Pachometric study on the influ ence of corneal endothelial vital staining: Corneal thickness' after cataract extraction studied by vital staining with trypan blue. Acta Ophthalmol. 51:679, 1973. 8. Wood, W. J., and Maumenee, A. E.: Corneal thickness after cataract surgery. Trans Am. Acad. Ophthalmol. Otolaryngol. 79:631, 1975. 9. Bourne, W. M., and Kaufman, H. E.: The endothelium of clear corneal transplants. Arch. Oph thalmol. In press.
AND H E R B E R T E. K A U F M A N ,
M.D.
Gainesville, Florida
Bullous keratopathy develops in a sig nificant percentage of eyes after cataract extraction. Most of these cases are thought to be due to corneal endothelial damage occurring during the cataract ex traction, postoperative vitreous contact, or preexisting Fuchs' dystrophy. Until recently, however, inadequate visualiza tion of the corneal endothelium has pre vented the precise evaluation of endothe lial damage. With the clinical specular microscope, 1 the endothelium can be ex amined and photographed in a patient before and after the procedure, allowing objective estimates of cell loss. In this study we examined the corneal endotheli um of 16 patients, before and at intervals after cataract extraction, to determine the measurable endothelial cell loss associ ated with complicated and uncomplicat ed cataract extraction.
neas, two had clear penetrating keratoplasties, and one had a clear cornea fol lowing blunt trauma. Intracapsular cryoextraction was performed on 12 patients, two of whom had surgical or postsurgical complications: one had vitreous loss and the other had severe postoperative in flammation with sterile hypopyon that cleared after intensive corticosteroid ther apy. Four younger patients with soft cata racts underwent phacoemulsification pro cedures, and the actual ultrasonification time was minimal. All the operations were performed by staff and residents here during a four-month period. The preoperative endothelial photographs were generally taken on the day before extraction, and the first postoperative ex aminations were usually taken the day after the operation. Several were taken a week or more later. Follow-up photo graphs were taken every few weeks there after, the longest period of study being 14 weeks. The corneal epithelium was intact at all examinations.
M A T E R I A L AND M E T H O D S
We examined and photographed the central corneal endothelium of 16 pa tients with the clinical specular micro scope before and at intervals after cataract extraction. The central endothelial cell density was determined for each cornea as previously described. 1 The central cor neal thickness was measured at each ex amination by the same investigator with a modified Haag-Streit pachometer. 2 The patients' eyes had not been patched for several hours before the thickness read ings. Thirteen patients had normal cor-
RESULTS
The corneas of 13 patients were normal by specular microscopy, and three (Cases 1, 13, and 16) were abnormal before sur gery (Table). These three corneas had abnormally low endothelial cell densities, although no cornea guttata or other changes were seen. Four patients (Cases 10 to 13) had a statistically significant decrease in endo thelial cell density after cataract extrac tion (Student's f-test, P<.05), although all 16 corneas remained clear postoperatively. Two of these four patients had a significant loss of endothelial cells during the ensuing weeks, although the corneas remained clear (analysis of variance with regression, P<.05). These were the same
From the Department of Ophthalmology, College of Medicine, University of Florida, Gainesville, Florida. This study was supported by Public Health Service grants EY-00033-12 and EY-00266-12, Na tional Institutes of Health. Reprint requests to William Μ. Bourne, M.D., Department of Ophthalmology, Mayo Clinic, Roch ester, MN 55901. 44
VOL. 82, NO. 1
ENDOTHELIUM AND CATARACT EXTRACTION
45
TABLE CATARACT EXTRACTIONS*
Case No., Age, yrs
Cause of Cataract
1,47 2,78 3,78 4,75 5,64 6,55 7,75 8,57 9,80 10,57 11,73 12,51 13,25
PK (Fuchs') Senile Senile Senile Senile Senile Senile Senile Senile Senile Senile Senile PK (keratoconus)
14,29
Diabetes
15,46
Senile?
16, 16
Blunt trauma
Procedure ICCE ICCE ICCE ICCE ICCE ICCE ICCE ICCE ICCE ICCE ICCE ICCE Phacoemulsification Phacoemulsifi cation Phacoemulsiflcation Phacoemulsiflcation
Endothelial Cell Density, cells/mm2 Pre Post operative operative 467
Corneal Swelling, mmf
Complications
413
0.04
None
2,369 2,963 2,397 2,660 2,984 2,459 2,700 2,234 2,296 2,860 2,366 1,103
2,771 2,570 2,307 2,636 2,873 2,105 2,395 2,060 1,423} 1,415} l,679t 687}
0 0.04 0.03 0.04 0.04 0.04 0.04 0.03 0.07 0.04 0.06 0.06
None None None None None None None None Iridocyclitis Vitreous loss None None
2,211
2,368
0.04
None
2,650
2,615
0.02
None
1,498
1,582
0.03
None
*ICCE indicates intracapsular cryoextraction, and PK, penetrating keratoplasty. t Increase in central corneal thickness at first postoperative measurement. } Statistically different from the preoperative value (P<.05).
two patients who suffered surgical or postsurgical complications (Cases 10 and 11) (Figure). One patient (Case 12) did not have further significant endothelial cell loss, and the fourth patient (Case 13) had a statistically significant increase in endothelial cell density during the ensu ing weeks, so that six weeks after cataract extraction, the endothelial cell density had returned to preoperative levels. The remaining 12 patients showed no signifi cant change in endothelial cell density during the entire period of study (P>.05). Fifteen patients had an increase in cen tral corneal thickness on the first exami nation after cataract extraction. Most thicknesses eventually returned to their preoperative levels. However, the four corneas that had a significant loss of en dothelial cells (Cases 10 to 13) also had a
statistically greater mean increase in cor neal thickness when compared to the re maining 12 corneas (f-test, P<.002). DISCUSSION
No presently detectable loss of central endothelial cells occurred in most of the uncomplicated intracapsular cataract ex tractions. Cell loss is more likely to occur if there are operative or postoperative complications such as vitreous loss or severe iridocyclitis. T w o patients (Cases 10 and 11) contin ued to lose central endothelial cells for weeks after cataract extraction. One of these patients had severe postoperative iridocyclitis, which seems to be responsi ble for the continued cell loss. The other patient had no demonstrable abnormal postoperative inflammation, although he
46
AMERICAN JOURNAL OF OPHTHALMOLOGY
JULY, 1976
Figure (Bourne and Kaufman). Case 10. The central endothelium before (top) and seven weeks after (bottom) intracapsular cataract extraction. A severe postoperative, sterile iridocyclitis with hypopyon occurred and cleared with intensive corticosteroid therapy.
had suffered vitreous loss. Undetected inflammation may have been present in this patient. A further decrease in the number of central endothelial cells may be due to spreading of cells to the periph eral cornea to cover areas of cell loss, for example, at the corneoscleral incision site. A significant increase in endothelial cell density occurred after the initial decrease in a 25-year-old man with an 11-month-old clear penetrating corneal transplant (Case 13). The cataract extrac tion was done by phacoemulsification. The increased number of central endothe lial cells presumably is the result of cell division or sliding of cells from the pe riphery of the graft or recipient. Healing by sliding is the more likely possibility, since this occurred in corneal transplants in primates 3 and central endothelial cells do not normally maintain their number with time by cell division. 1 To our knowl edge, this is the first documentation of an increase in the number of central endo thelial cells in humans.
The small transient increase in corneal thickness that occurred in the 12 patients without demonstrable endothelial cell loss was statistically significant (i-test, P<.02). This apparently represents a tem porary decrease in endothelial function, presumably due to operative manipula tion. Increases in corneal thickness after cataract extraction have been reported. 4 - 8 A greater transient mean increase in cor neal thickness was demonstrated in our four patients with significant endothelial cell loss, indicating a more severe, al though temporary, disturbance in endo thelial function in these patients. These findings indicate that in the absence of epithelial ulcération, stromal scarring, or elevated intraocular pressure, corneal thickness is an approximate measure of endothelial function, and the larger the postoperative increase in thickness, the more likely it is that endothelial cell loss occurred during the procedure. Also, no loss of endothelial cells was demonstrated in three of four phacoemul sification procedures (extracapsular).
VOL. 82, NO. 1
ENDOTHELIUM AND CATARACT EXTRACTION
They were all soft cataracts in young adults (average age, 29 years), so that this procedure appears to be safe for this group of patients. Two of the four corneas were abnormal before the phacoemulsification procedures (Cases 13 and 16), and the one that lost endothelial cells was a corneal transplant (Case 13). Although significant endothelial cell loss was demonstrated in four of the pa tients, all 16 corneas remained clear dur ing the period of observation. A long-term study will be needed to determine if the cell loss is clinically detrimental. It is clear that extremely low endothelial cell densities maintain clear corneal trans plants for years. 9 However, it seems rea sonable to assume that the corneas which lost many endothelial cells are more sus ceptible to the detrimental effects of in flammation, trauma, or aging. We would like, therefore, to minimize the endothe lial damage from our surgical procedures. This study demonstrates an excellent method for documenting this damage and perhaps for more accurately predicting those corneas at greatest risk before sur gery. SUMMARY
We examined and photographed the central corneal endothelium of 16 pa tients with the clinical specular micro scope before and at intervals after cataract extraction. No detectable loss of endothe lial cells occurred in 75% of the patients, including 12 routine intracapsular cryoextractions and four phacoemulsifications of soft cataracts in young adults. Only one of the four cases of significant endothelial cell loss occurred in a normal
47
cornea without demonstrable operative or postoperative complications. Two of the four corneas that lost central endothelial cells at cataract extraction continued to lose more cells during the ensuing weeks. A significant increase in central endothe lial cell density was demonstrated in one patient. The four corneas with endotheli al cell loss also had a significantly higher mean increase in corneal thickness postoperatively, although this was transient. All 16 corneas remained clear during the period of observation (maximum, 14 weeks), and long-term studies are needed to measure the chronic effects of the en dothelial damage. REFERENCES 1. Bourne, W. M., and Kaufman, H. E.: Specular microscopy of human corneal endothelium in vivo. Am. J. Ophthalmol. 81:319, 1976. 2. Mishima, S., and Hedbys, B. O.: Measurement of corneal thickness with the Haag-Streit pachometer. Arch. Ophthalmol. 80:710, 1968. 3. Bourne, W. M.: In vivo survival of cryopreserved endothelial cells in primates. Arch. Ophthal mol. 92:146, 1974. 4. Giardini, A., and Cambiaggi, A.: Recherches sur l'é paisseur cornèenne après extraction de la cataracte. Ophthalmologica 131:41, 1956. 5. Lavergne, G., and Kelecom, J.: Applications cliniques de la mesure de l'é paisseur de la corneè. Bull. Soc. Belge Ophtalmol. 131:323, 1962. 6. Miller, D., and Dohlman, C. H.: Effect of cataract surgery on the cornea. Trans. Am. Acad. Ophthalmol. Otolaryngol. 74:369, 1970. 7. Norn, M. S.: Pachometric study on the influ ence of corneal endothelial vital staining: Corneal thickness' after cataract extraction studied by vital staining with trypan blue. Acta Ophthalmol. 51:679, 1973. 8. Wood, W. J., and Maumenee, A. E.: Corneal thickness after cataract surgery. Trans Am. Acad. Ophthalmol. Otolaryngol. 79:631, 1975. 9. Bourne, W. M., and Kaufman, H. E.: The endothelium of clear corneal transplants. Arch. Oph thalmol. In press.