Corneal Topographic Changes after Retinal and Vitreous Surgery Dov Weinberger, MD, Henia Lichter, MD, Nissim Loya, MD, Ruth Axer–Siegel, MD, Larisa Muzmacher, MD, Uri Gabbay, MD, Yuval Yassur, MD Purpose: To investigate the topographic changes in the cornea after retinal and vitreous operations. Design: Observational prospective case series. Participants: The study population included 46 patients after vitreoretinal surgery: 11 underwent pneumatic retinopexy, 10 underwent vitrectomy, and 25 underwent scleral buckling procedure. Methods: The corneal topography was measured by videokeratography with the absolute program and evaluated statistically by a quantitative comparative method, which was developed for this study, for the whole and the central cornea. Main Outcome Measures: The corneal topographic changes were measured in diopters (D), evaluating and comparing the preoperative and postoperative measurements. Results: None of the operative procedures changed the shape of the whole cornea. Vitrectomy induced radial steepening of the central cornea 1.2 to 1.6 D, corresponding to the scleral sutures. Central steepening (average, 2.2 D) was also noted in the first week after circular buckling, but it flattened (average, 1.4 D) after 1 to 3 months. When an additional radial or circumferential buckling element was added to the circular buckle, steepening of the entire cornea and radial steepening of the central cornea (average, 0.6 – 0.8 D) occurred in the first week and flattened or returned to baseline after 1 to 3 months. There was no correlation between the location of the additional buckling element and the corneal topographic change. Conclusions: Corneal videokeratography is a useful tool for evaluating the postoperative corneal curvature. It showed that vitreoretinal surgery alters the shape of the cornea when buckling or scleral sutures are used, but pneumatic retinopexy does not. Ophthalmology 1999;106:1521–1524 Retinal detachment surgery is a well-known cause of ocular refractive changes. The encircling band creates a circular indentation of the eye, thereby increasing its anteroposterior axial length; the myopic shift may be up to 3 diopters (D).1– 4 Although several authors have also reported changes in the corneal curvature after scleral buckling,5–9 the results were based on measurements with a keratometer combined with ultrasonography. Only one study published by Hayashi et al10 applied videokeratography in patients after buckling procedure for retinal detachment. Computer-assisted analysis of photokeratoscope images was introduced by Klyce and Wilson11 in 1989. It has since been used to investigate corneal changes in various corneal diseases,12 such as keratoconus,13 astigmatism after penetrating keratoplasty,14 fitting contact lenses, and after keratotomy.15 In the current study, we used videokeratography to investigate topographic changes of the cornea after scleral buckling operation, pneumatic retinopexy, and vitrectomy.
Originally received: October 12, 1998. Revision accepted: May 5, 1999. Manuscript no. 98679. From the Department of Ophthalmology, Rabin Medical Center, Beilinson Campus, Petah Tiqva and Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel. Address correspondence to Dov Weinberger, MD, Department of Ophthalmology, Rabin Medical Center, Beilinson Campus, Petah Tiqva 49 100, Israel.
Materials and Methods Patients Forty-six patients (32 men, 14 women) after retinal or vitreous surgery were included in the study: 11 pneumatic retinopexy, 10 vitrectomy, and 25 scleral buckling procedure. Ages ranged between 40 and 72 years (mean, 56 years). Patients who had had a previous ocular operation or who underwent more than one procedure were excluded. Pneumatic retinopexy consisted of the injection of 0.4 ml perfluoropropane gas into the vitreous cavity through the pars plana. In all the patients, the left eye was investigated for vitrectomy; three ports of entry were used—the upper nasal, upper temporal, and lower temporal quadrants—3.5 mm posterior to the limbus. These ports were closed at the end of the procedure with 7– 0 silk mattress sutures. No buckling elements were added. In the scleral buckling group (Table 1), a 2.5-mm-wide encircling extrascleral silicone band was placed at the equator with four 6 – 0 Dacron sutures, one in each quadrant. The band edges were tied in the upper nasal quadrant by a Supramid knot. In 11 of the 25 patients, a 7-mm-wide circumferential solid grooved silicone rubber band or sponge was added in different quadrants as needed, and in 4 patients a 5-mm cylindrical sponge was added radially beneath the retinal hole (Table 1).
Corneal Topography Assessment Corneal topography was measured by videokeratography (Topographic Modeling System; TMS-1, Computed Anatomy, New York, NY) with the absolute program.
1521
Ophthalmology Volume 106, Number 8, August 1999 Table 1. Scleral Buckle Procedure: Location of Additional Elements Patient No. 1 2 3 4
Radial Element Location
Circumferential Element
Eye
Lower temporal Lower nasal Upper nasal Lower temporal
Location
Eye
Right
Temporal half
Right
Left
Temporal half
Left
Left
Upper nasal sponge
Right
Right
270° temporal half ⫹ upper nasal Lower temporal Lower nasal Lower nasal Lower half Upper temporal sponge Lower half groove Lower nasal sponge
Right
5 6 7 8 9 10 11
Left Left Right Right Right Right Left
The color-coded map developed for the Tomey instrument represents a qualitative description of the cornea derived from a single measurement and precludes quantitative comparisons among patients or between timepoints for individual patients. To counter this problem and to investigate various topographic locations in the cornea, we developed a comparative technique using the Topographic Modeling System (TMS-1) data file. The corneal curvature is divided into 30 concentric circles. The first circle contains 256 points that start at the 3-o’clock position and move counterclockwise, passing into the second circle at the 3-o’clock position, and so on. The curvature is measured in diopters, with each point representing 1 dioptric value, for a total of 7680 values (30 ⫻ 256). The distance between every two points is 1.4° of the circle. The numeric data of the topographic maps were computerized as an ASCII file and processed by a mathematical algorithm developed by our group (Dr. Gabbay, Gerthner Institute for Epidemiology and Health Service Research, Israel). The algorithm creates a matrix of 30 ⫻ 256 value points (representing each of the corneal topographic dioptric values) for each topographic map and subtracts the identical corneal analog points between the two different matrices, creating a new matrix representing the difference values. The program enables the quantitative assessment of changes in the whole cornea or chosen sectors over time. In the current study, measurements were made before surgery and at 1, 4, and 12 weeks after surgery. Comparisons were conducted for the entire cornea (30 rings) and for the central cornea (5 central rings), which represent the visual axis. For the
latter, the 5 central rings were divided into 16 equal radial wedges, each representing 22.5° of the ring, starting at 0° and rotating counterclockwise (such that wedge 1 represented 0 –22.5°, wedge 4 represented 67.5–90°, and so on). We calculated both the average differences between the 30 rings on repeated measurements (the whole cornea) and the average differences between corresponding radial wedges in the first 5 rings (central cornea). A change of 0.5 D in the corneal shape was considered significant.
Results The corneal curvature changes from baseline to 1, 4, and 12 weeks after surgery in the three groups of patients are shown in Tables 2 and 3. The whole cornea measurements showed an average steepening of 0.3 to 0.5 D for pneumatic retinopexy, 0.2 D for vitrectomy, and 0.4 D for the buckling procedure with no additional elements; none of these findings was significant. In the patients with circular buckling combined with a partial circumferential element, significant steepening of an average of 0.8 D was noted at 1 week, with flattening after 1 month and repeated steepening after 3 months. Patients with circular buckling combined with a radial element showed an average steepening of 0.6 D at 1 week after surgery followed by flattening of 0.3 D at 4 and 12 weeks. Central cornea (radial wedge) measurements yielded the following: pneumatic retinopexy—average steepening of 0.4 D at all postoperative timepoints; vitrectomy—average steepening of 1.2 to 1.6 D at 67° and 157° at 1 week and flattening at 67° at 12 weeks; circular buckling alone—average steepening of 2.2 D between 67° and 135° and of 2 D at 270° to 315° at 1 week with flattening of 1.1 to 1.6 D at 4 and 12 weeks; circular buckling with partial circumference element—average steepening of 0.8 D at 1 week, which flattened at 4 weeks and then remained stable; and circular buckling plus radial element—average steepening of 0.7 D at 1 week, which flattened by 0.7 D at 4 weeks and then remained stable. There was no correlation between the location of the circumferential or radial buckle and the radial curvature changes.
Discussion In this study, we analyzed the central cornea and whole cornea topographic changes that occur after gas injection, vitrectomy, and buckling procedures with the use of videokeratography and a mathematical algorithm specially developed for the study.
Table 2. Whole Cornea Changes: D Postoperative Time (wks)
Pneumatic Retinopexy (n ⴝ 11)
Vitrectomy (n ⴝ 10)
SB Circular (n ⴝ 10)
SB ⴙ Circumferential Element (n ⴝ 11)
SB ⴙ Radial Element (n ⴝ 4)
1 4 12
⫹0.5 ⫹0.4 ⫹0.3
⫹0.2 ⫹0.2 0
⫹0.4 0 0
⫹0.8 0 ⫹0.5
⫹0.6 ⫺0.3 ⫺0.3
SB ⫽ scleral buckle, 360°; ⫹ ⫽ steepening of the cornea; ⫺ ⫽ flattening of the cornea.
1522
Weinberger et al 䡠 Corneal Topography after Vitreoretinal Surgery Table 3. Radial Changes of the Central Cornea Pneumatic Retinopexy
Postoperative Time (wks)
(D)
1
Vitrectomy
SB
SB ⴙ Circumferential Element (D)
SB ⴙ Radial Element (D)
(°)
(D)
(°)
(D)
(°)
⫹0.4
0, 180
67–135 270–315
⫹0.8
0.7
⫹0.5
0–180 upper half 90, 270
135–180 337–360 225–270 22–45 157–180 337–360
⫺0.7
⫹0.5
⫺1.2 ⫺1.6 ⫺1.4 ⫺1.3 ⫺1.1 ⫺1.4
⫺0.5
12
45–90 157–202 337–360 0–67 157–315 67–112 247–360
⫹2.2 ⫹2
4
⫹1.2 ⫹1 ⫹0.8 ⫹1.4 ⫹1.6 ⫺1.4 ⫺0.5
0
0
SB ⫽ scleral buckle, 360°.
The effect of a retinal detachment operation on the corneal curvature has barely been investigated. The induction astigmatism was reported in some studies5– 8 but not in others3,4; when it was noted, it was usually infrequent2– 4 and transitory.6,8 Induced astigmatism has been associated with radial rather than circumferential scleral buckles,4,5,7 greater scleral buckle height,5 anterior location of the scleral buckle,4 and use of a spongy material rather than silicone.8 In one study of corneal curvature changes, Watanabe et al9 found that circumferential buckling caused steepening of the corneal curvature in the direction of the buckling, and radial buckling caused heightening of the corneal curvature in the direction of the buckling. These changes, however, disappeared within 2 months after the surgical procedure. All previous studies on postoperative corneal curvature changes have been based on keratometer and keratoscope measurements, except those of Hayashi et al,10 who used videokeratography after scleral retinal buckling and found that the surgery induced prolonged irregular and asymmetric changes in the corneal shape. Our videokeratographic investigation comparing the effect of several types of retinal and vitreous procedures yielded the following findings: pneumatic retinopexy, vitrectomy, and circular buckle operation, without additional element, do not alter the entire corneal curvature. The only changes noted were after buckling operations when additional elements, either circumferential or radial, were added to the circular band. In these cases, corneal steepening occurred in the first postoperative week and then disappeared or flattened after 3 months. Changes in the central cornea, which represent the visual axis, were as follows: pneumatic retinopexy did not alter the curvature of the central cornea, but vitrectomy induced significant steepening in correlation with the entry ports that were sutured. After 3 months, the central cornea flattened in approximately the same meridians, possibly owing to changes in scleral elasticity or loosening of the sutures. Surgeons who use different types of sutures, with different elastic properties, to close the sclerotomy sites should consider corneal topographic changes after surgery. After the encircling buckle procedure, with or without additional
buckling elements, the central cornea steepened in the first week and then flattened at the fourth and twelfth weeks. There was no correlation between the location of the additional elements and the corneal radial changes. Further studies that compare visual acuity results after scleral buckling surgery might benefit from reporting results after 12 weeks, for before that period of time, visual acuity is likely affected by corneal changes. In conclusion, corneal videotopography is a useful tool for evaluating the postoperative corneal curvature. It showed that vitrectomy and buckling surgery induce significant central curvature changes. Pneumatic retinopexy does not induce either whole cornea or central corneal curvature changes.
References 1. Burton TC, Herron BE, Ossoinig KC. Axial length changes after retinal detachment surgery. Am J Ophthalmol 1977;83: 59 – 62. 2. Smiddy WE, Loupe DN, Michels RG, et al. Refractive changes after scleral buckling surgery. Arch Ophthalmol 1989;107:1469 –71. 3. Larsen JS, Syrdalen P. Ultrasonographic study on changes in axial eye dimensions after encircling procedure in retinal detachment surgery. Acta Ophthalmol (Copenh) 1979;57: 337– 43. 4. Rubin ML. The induction of refractive errors by retinal detachment surgery. Trans Am Ophthalmol Soc 1976;73:452– 90. 5. Goel R, Crewdson J, Chignell AH. Astigmatism following retinal detachment surgery. Br J Ophthalmol 1983;67:327–9. 6. Fiore JV Jr, Newton JC. Anterior segment changes following the scleral buckling procedure. Arch Ophthalmol 1970;84: 284 –7. 7. Burton TC. Irregular astigmatism following episcleral buckling procedure with the use of silicone rubber sponges. Arch Ophthalmol 1973;90:447– 8. 8. Mensher JH, Burton TC. Corneal curvature changes after scleral buckling. In: Blodi F, ed. Current Concepts in Ophthalmology. St. Louis, MO: C.V. Mosby Co., 1974; v. IV, chap. 4.
1523
Ophthalmology Volume 106, Number 8, August 1999 9. Watanabe K, Emi K, Hamano T, et al. Corneal topographic evaluation of retinal detachment surgery. Nippon Ganka Gakkai Zasshi 1988;92:367–71. 10. Hayashi H, Hayashi K, Nakao F, Hayashi F. Corneal shape changes after scleral buckling surgery. Ophthalmology 1997; 104:831–7. 11. Klyce SD, Wilson SE. Methods of analysis of corneal topography. Refract Corneal Surg 1989;5:368 –71. 12. Missotten L. Corneal topography. Curr Opin Ophthalmol 1994;5:68 –74.
13. Rabinowitz YS, Nesburn AB, McDonnell PJ. Videokeratography of the fellow eye in unilateral keratoconus. Ophthalmology 1993;100:181– 6. 14. Khong AM, Mannis MJ, Plotnik RD, Johnson CA. Computerized topographic analysis of the healing graft after penetrating keratoplasty for keratoconus. Am J Ophthalmol 1993;115:209–15. 15. McDonnell PJ, Garbus JJ, Caroline P, Yoshinaga PD. Computerized analysis of corneal topography as an aid in fitting contact lenses after radial keratotomy. Ophthalmic Surg 1992; 23:55–9.
Historical Image Postage stamp commemorating Frans Donders.
* Courtesy of the Museum of Ophthalmology, Foundation of the American Academy of Ophthalmology, San Francisco, California.
1524