- Email: [email protected]
Corneal Astigmatism After Phacoemulsification and Lens Implantation Through Unsutured Scleral and Corneal Tunnel Incisions
Corneal Astigmatism After Phacoemulsification and Lens Implantation Through Unsutured Scierai and Corneal Tunnel Incisions ROBERT H. GROSS, MD., AND KEVIN M. MILLER, M.D.
• PURPOSE: We compared the changes in corneal astigmatism after phacoemulsification and intraoc ular lens implantation in 93 consecutive eyes with unsutured 4-mm superior scierai tunnel incisions to those through 105 consecutive eyes with unsutured 3.2- to 3.5-mm temporal corneal tunnel incisions. • METHODS: Keratometry measurements were ob tained preoperatively and at postoperative day 1, week 1, and week 6. Group differences in scalar and vector astigmatism were compared by using analysis of variance methods. • RESULTS: Mean scalar astigmatism in the scierai incision group changed from preoperative astigma tism by 0.65 diopter at postoperative day 1, 0.37 diopter at postoperative week 1, and 0.13 diopter at postoperative week 6. Mean scalar astigmatism in the corneal incision group changed from preop erative astigmatism by 0.39 diopter at postopera tive day 1, 0.21 diopter at postoperative week 1, and 0.13 diopter at postoperative week 6. Mean vector astigmatism in the scierai incision group changed 1.26 diopters at 8 0 degrees at postopera tive day 1, 1.05 diopters at 83 degrees at postoper-
Accepted for publication July 20, 1995. From the Jules Stein Eye Institute and the Department of Ophthal mology, University of California at Los Angeles School of Medicine, Los Angeles, California. This study was supported by a grant from the Karl Kirchgessner Foundation, Los Angeles, California. It was pre sented, in part, at the annual meeting of the American Sociery of Cataract and Refractive Surgery, San Diego, California, April 2, 1995. Reprint requests to Kevin M. Miller, M.D., Jules Stein Eye Institute, 100 Stein Plaza, UCLA, Los Angeles, CA 90095-7002; fax: (310) 825-0841; E-mail: [email protected].
VOL.121, No. 1
ative week 1, and 0.42 diopter at 103 degrees at postoperative week 6. Mean vector astigmatism in the corneal incision group changed 0.77 diopter at 90 degrees at postoperative day 1, 0.75 diopter at 88 degrees at postoperative week 1, and 0.61 diopter at 8 9 degrees at postoperative week 6. The differences were statistically significant (P = .003) only by vector analysis at the postoperative day 1 examination. • CONCLUSIONS: We found significantly greater with-the-rule change in astigmatism in the scierai incision group than in the corneal incision group on the first postoperative day. The effect disap peared by the sixth postoperative week.
C
ATARACT EXTRACTION BY THE PHACOEMULSIFI-
cation technique with implantation of foldable intraocular lenses through unsutured scierai tunnel incisions 4 mm or less in width has been shown to induce a smaller change in corneal astigma tism than surgery through larger scierai incisions.1'7 Nielsen8 found a difference in surgically induced astigmatism in a comparison of phacoemulsification through 3.5-mm scierai incisions to that through 3.5-mm corneal incisions, but the study had a limited number of patients and the difference was not statisti cally significant. In this study we compared the astigmatic effects of phacoemulsification and intraoc ular lens implantation through unsutured 4-mm scier ai tunnel incisions located in the superior vertical meridian (12 o'clock meridian) to those through unsutured 3.2- to 3.5-mm corneal tunnel incisions located in a temporal (3 or 9 o'clock) meridian.
© AMERICAN JOURNAL OF OPHTHALMOLOGY i996,-i2i:57-64
57
PATIENTS AND METHODS WE REVIEWED THE RECORDS OF 198 CONSECUTIVE EYES
that underwent phacoemulsification and foldable in traocular lens implantation by one of us (K.M.M.). Scierai incision surgery had been performed on 93 eyes, and corneal incision surgery on 105 eyes. Most of the scierai incision operations were performed before the first corneal incision procedure was per formed. In all of the eyes that underwent scierai incision, anesthesia was attained by posterior orbital injection of 3 to 5 ml of a 50:50 mixture of 2% lidocaine with epinephrine 1:100,000 and 0.75% bupivacaine with one ampule (150 USP units) of hyaluronidase. In the eyes that underwent corneal incision, anesthesia was attained either by posterior orbital injection or by topical administration of four drops of 1 % tetracaine. In the eyes that underwent scierai incision, a twoto three-clock hour limbal peritomy centered at the 12 o'clock meridian was fashioned with Wescott scissors and Colibri forceps. Wetfield bipolar cautery was applied to coagulate episcleral vessels. A 4-mmlong partial-thickness scierai groove was fashioned 2.5 mm posterior to the 12 o'clock meridian corneoscleral limbus by using a No. 69 Beaver blade. A scierai tunnel was dissected anteriorly 0.5 to 1.0 mm into clear cornea with the same blade. A second instrument port was fashioned at the 2:30 o'clock meridian with a 15-degree angle blade, and the anterior chamber was re-entered through the scierai tunnel incision by using a straight 3.2-mm keratome to form a three-stepped self-sealing tunnel incision. Hyaluronic acid was injected to reform the anterior chamber. In the patients who underwent corneal incision, a 3.2- to 3.5-mm partial-thickness corneal groove was fashioned at the temporal corneoscleral limbus in the 3 or 9 o'clock meridian with a 1-mm two-step diamond knife with the blade set to a depth of 300 to 400 μιτι. A second instrument site was established for the left hand at either the 10:30 or the 4:30 o'clock meridian, depending on the eye undergoing surgery. The same diamond knife was used with the blade fully extended. Hyaluronic acid was injected through the side port incision to increase intraocular pressure
58
and facilitate the incision. The eye was re-entered through the temporal corneal incision with a 3.5/3.2mm or 3.2/2.7-mm trapezoidal diamond keratome to form a three-stepped self-sealing tunnel incision. After the incision was made, a continuous curvi linear capsulorhexis was started by using a bent 25-gauge cystotome and was completed with Utrata forceps. Hydrodissection was performed by injecting balanced saline solution between the nucleus and cortex through a 27-gauge cannula. The nucleus was removed with an Alcon Series Ten Thousand phaco emulsification unit by using a standard two-handed nuclear cracking technique (Alcon Surgical, Irvine, California). Residual cortex was removed and the posterior capsule was polished, if necessary, by using the irrigation and aspiration probe. The capsule and anterior chamber were filled with sodium hyaluronate, and a foldable silicone intraocular lens was implanted in the capsule through an unenlarged incision. After removal of the sodium hyaluronate, the anterior chamber was reformed with balanced saline solution, and the wound was tested with absorbent sponges. In the eyes that underwent corneal incision, if the wound was not self-sealing and watertight, the corne al stroma was hydrated with additional balanced saline solution. In the eyes that underwent scierai incision, the conjunctiva was brought over the limbal area and secured with bipolar forceps cautery. Keratometry was performed preoperatively, and at postoperative day 1, postoperative week 1, and post operative week 6, with a manual keratometer. In the scalar analysis, astigmatism was calculated as the absolute value of the difference in keratometric measurements. Changes in scalar astigmatism were calculated by subtracting the preoperative amount from that measured at each postoperative examina tion. In the vector analysis, the magnitude assigned to the astigmatism vector was that calculated in the scalar analysis. The angle assigned to the astigmatism vector was that of the greater keratometric value. Changes in vector astigmatism were determined by subtracting the preoperative vector from that mea sured at each postoperative examination with a double-angle polar coordinate system for analysis of toric surfaces (Fig. 1). Custom software for perform ing the vector calculations was written and imple-
AMERICAN JOURNAL OF OPHTHALMOLOGY
JANUARY 1996
one-way multiple analysis of variance, with the F statistic based on Hotelling's T2-test method." (0 α> Q. O
RESULTS
S
0/180 c
90'
With-the-rule
135 u
Against-the-rule
Fig. 1 (Gross and Miller). Graphic method for calculat ing the corneal astigmatism change vector. Representa tive preoperative and postoperative astigmatism vectors are plotted in polar coordinates at twice the actual angle. The change vector sums with the preoperative vector to produce the postoperative vector. It can be drawn ex tending from the plot origin. Vectors falling on the left side of the graph represent with-the-rule astigmatism. Vectors falling on the right side of the graph represent against-the-rule astigmatism.
mented on a Microsoft Excel spreadsheet according to the general methodology of Jaffe, Jaffe, and JafFe.9 Scleral and corneal incision group mean scalar astig matism changes were compared by using one-way analysis of variance. At each examination, average change vectors were calculated by converting the individual polar vectors into rectangular coordinate system components (rx, r y ), calculating averages (rx = 1/ΠΣΓΧ, ry = l/n2r y ), and reverse transforming the resultant [r = V (rx2 4ry2), Θ = tan" 1 (r x /r y )]. Approximate 95% confidence intervals for the average change vector magnitudes and angles were computed by forming the confidence ellipse (Fig. 2) in terms of rx and ry and then transforming the ellipse to polar coordinates (Fig. 3). 10 The ellipse model assumes that the rx and ry means have an approximate bivariate gaussian distri bution. Scleral and corneal incision group mean astigmatism change vectors were compared using
VOL.121,
No. 1
IN THE SCALAR ANALYSIS, THE PREOPERATIVE CORNEAL
astigmatism in the 4-mm scleral incision and 3.2- to 3.5-mm corneal incision groups was not statistically different (P = .126). There was slightly more surgical ly induced astigmatism in the scleral incision group than in the corneal incision group at postoperative day 1 and postoperative week 1 (Fig. 4), but the difference was not statistically significant either at these examinations or at the postoperative week 6 examination (Table 1). The changes in scalar astig matism at postoperative week 6 were virtually identi cal in the two groups. In the vector analysis, there was a mean with-therule change in corneal astigmatism in the scleral and corneal incision groups at all postoperative examina tions (Fig. 5) that was statistically greater (P = .003) for the scleral incision group on the postoperative day 1 examination (Table 2). In both groups there was a gradual decay in the average amount of with-the-rule
!
r (diopters)
Fig. 2 (Gross and Miller). A representative average astigmatism change vector plotted in Cartesian coordi nates surrounded by the 95% confidence ellipse. rx is the projection of the vector onto the axis at 0/180 degrees. ry is the projection of the vector onto the axis at 45 degrees (in a double-angle polar coordinate system).
ASTIGMATISM AFTER SCLERAL AND CORNEAL TUNNEL INCISIONS
59
I 90'
0/180
135° Fig. 3 (Gross and Miller). A representative change vector plotted in a double-angle polar coordinate system sur rounded by the approximate 95% confidence interval for vector magnitude (r95) and angle (θ95). The area repre sented by the shaded wedge should include 95% of all change vectors in the group.
vector astigmatism change (Fig. 6). The postoperative drift toward against-the-rule corneal astigmatism was more pronounced for the scierai incision group. At the postoperative week 6 examination, when final prescriptions for spectacles were written, there was an average vector change of 0.42 diopter in the scierai incision group and 0.61 diopter in the corneal incision group.
change as it steepened the corneal meridian 90 degrees from the incision. The astigmatic change caused by the temporal corneal cataract incision was what one would have expected for a peripheral arcuate keratotomy incision. The small decrease in astigma tism that was observed on subsequent examinations was the result of wound healing. The with-the-rule or with-the-wound change in astigmatism that occurred in the scierai incision group, however, is not analo gous to the change induced by arcuate keratotomy, because the astigmatism would then have gone against the wound, as it did in the corneal incision group. Rather, the early with-the-wound change can probably be explained either by the episcleral cautery, by the edema that developed in the scierai and corneal tissue surrounding the scierai tunnel, by the compressive effects of the eyelids, or by a combination of factors. In most respects, except for the application of cautery (including wound architecture, wound hydration, operating time, absence of suture place ment, and postoperative management), the two groups were treated similarly. Cataract surgeons who use cautery during surgery can often see the episclera contracting beneath the tip of the wet-field probe. In a large series of patients, Grabow12 found a 50% reduction in induced astigmatism at the postoperative week 1 examination when cautery was limited to the
0.8 0.7
• — Scierai incision * — Corneal incision
0.6
DISCUSSION IN THIS STUDY WE FOUND A SIGNIFICANT DIFFERENCE
between unsutured 4-mm scierai tunnel incisions and unsutured 3.2- to 3.5-mm corneal tunnel incisions with respect to early postoperative keratometric astig matism by vector methods. The difference disap peared by the sixth postoperative week. There was no difference between the two groups when the change in astigmatism was measured by scalar methods. Both surgical procedures produced an immediate with-the-rule change in astigmatism that decreased on subsequent examinations. For the temporal corne al incision group, the with-the-rule change in astig matism was simultaneously an against-the-wound
60
1
10 Days after surgery
100
Fig. 4 (Gross and Miller). Average change in scalar astigmatism (± S.E.) in the s*cleral and corneal incision groups at postoperative day 1, postoperative week 1, and postoperative week 6. The differences between the two groups were not statistically significant at any postopera tive time point.
AMERICAN JOURNAL OF OPHTHALMOLOGY
JANUARY 1996
Scierai Incision Postoperative week 6
Postoperative week 1
Postoperative day 1
45°
o Q
90°
0/180 '
0/180 ° 90"
0/180° 90"
135 l
135'
135'
Corneal Incision
I
o a
0/180° 90°
135 °
0/180°
90°
0/180"
135 °
135 °
Fig. 5 (Gross and Miller). Individual change vectors represented as points at each postoperative examination. The scierai incision group is shown in the upper half, and the corneal incision group in the lower half. Most change vectors are with-the-rule at each postoperative examination, although there is a clustering toward the origin with less dispersion over time.
Postoperative day 1
Postoperative week 1
0/180°
135'
Postoperative week 6
0/180°
0/180"
135"
135° lόlj Scierai Incision
LV Corneal Incision
Fig. 6 (Gross and Miller). Average change vectors represented as 95% confidence intervals for magnitude and axis at each postoperative examination. The areas represented by the shaded wedges should include 95% of all change vectors in each group. The difference at postoperative day 1 is significant (P = .003) and the difference at postoperative week 1 approaches significance (P = .08).
VOL.121, No. I
ASTIGMATISM AFTER SCLERAL AND CORNEAL TUNNEL INCISIONS
61
TABLE 1 CHANGE IN SCALAR CORNEAL ASTIGMATISM WITH SCLERAL AND CORNEAL TUNNEL INCISION PHACOEMULSIFICATION AND INTRAOCULAR LENS IMPLANTATION ASTIGMATISM CHANGE IN SCLERAL INCISION GROUP (D)
ASTIGMATISM CHANGE IN CORNEAL INCISION GROUP (D)
p
NO.
MEAN ± S.E.
NO.
MEAN ± S.E.
F STATISTIC
VALUE
91 90 73
0.651 ± 0.136 0.371 ± 0.107 0.134 ± 0.071
101 102 87
0.388 ± 0.097 0.211 ± 0.082 0.130 ± 0.077
2.56 1.46 0.00
.111 .229 .966
Postoperative day 1 Postoperative week 1 Postoperative week 6
episclera posterior to the insertion of Tenon's fascia and not carried all the way to the corneoscleral limbus. The relatively marked decrease in astigmatism observed on follow-up examinations in our scierai incision group is most readily explained by scierai relaxation, as either the cautery effect or stromal edema wore off. If it is possible for these effects to wear off completely, and if 4-mm scierai incisions produce the same amount of corneal relaxation as 3.2- to 3.5-mm corneal incisions, we would anticipate an eventual against-the-wound astigmatic drift of about 0.6 diopter, as occurred in the corneal incision group. The follow-up interval of this study was insufficient to make this determination. The astigmatic benefits of unsutured small-incision phacoemulsification incisions have been documented extensively. Levy, Pisacano, and Chadwick 7 found a vector astigmatism change at one month of 1.35 ± 0.98 diopters in a 5.1-mm scierai incision group in contrast to 0.95 ± 0.62 diopter in a 3.5-mm scierai incision group. The difference between their two groups was statistically significant. Oshika and associ ates6 found a vector astigmatism change at one month
of 0.30 ± 0.68 diopter in a 5.5-mm scierai incision group in contrast to 0.08 ± 0.58 diopter in a 3.2-mm scierai incision group. This difference was statistically significant. Brint, Ostrick, and Bryan1 found a scalar astigmatism change at six weeks of 1.03 ± 0.51 diopters in a 4-mm scierai incision group in contrast to 1.33 ± 0.79 diopters in a 7-mm scierai incision group. This difference was also statistically significant. El-Maghraby and associates5 found less induced vec tor astigmatism at the postoperative day 2 and postop erative week 1 examinations in a group of patients who had surgery through 3.5-mm scierai incisions than in patients who had surgery through 5.5- and 6.5-mm scierai incisions. All the incisions in their study, even the 3.5-mm group, were closed with X-shaped sutures. Gills and Sanders2 found less vector astigmatism two to three weeks after surgery in a group of 55 eyes receiving 3-mm scierai incisions than in a group of 48 eyes receiving 6- to 7-mm incisions. Other investigators observed similar effects with small scierai incisions.3,4 By contrast, Neumann and associates13 found no difference between 56 eyes undergoing 6-mm scierai
TABLE 2 CHANGE IN VECTOR CORNEAL ASTIGMATISM WITH SCLERAL AND CORNEAL TUNNEL INCISION PHACOEMULSIFICATION AND INTRAOCULAR LENS IMPLANTATION ASTIGMATISM CHANGE IN SCLERAL INCISION GROUP (D)
AXIS (DEGREES)
ASTIGMATISM CHANGE IN CORNEAL INCISION GROUP (D)
AXIS (DEGREES)
95% CONFIDENCE 95% CONFIDENCE 95% CONFIDENCE 95% CONFIDENCE F P NO. MEAN INTERVAL MEAN INTERVAL NO. MEAN INTERVAL MEAN INTERVAL STATISTIC VALUE
Postoperative day 1 34 1.26 Postoperative week 1 36 1.05 Postoperative week 6 28 0.424
62
0.86-1.67 0.67-1.42 0.44-0.72
80.0 83.0 103
69-92 70-100 58-161
95 0.766 99 0.749 84 0.608
0.59--0.94 0.59-0.91 0.46-0.76
AMERICAN JOURNAL OF OPHTHALMOLOGY
90.0 87.9 88.9
79-102 81-95 78-100
6.06 2.58 1.38
.0031 .0799 .256
JANUARY
1996
incision surgery and 67 eyes undergoing 3- to 4-mm scleral incision surgery with regard to vector astigma tism at three months. Grabow12 found no difference between 250 eyes undergoing 5.2-mm scleral incision surgery and 280 eyes undergoing 4-mm scleral inci sion surgery with respect to scalar astigmatism at one month. No difference in postoperative vector astigma tism was found at any postoperative examination in a study by Pfleger, Scholz, and Skorpik,14 who com pared a 4.5-mm scleral incision group with a 3.5-mm scleral incision group. Davison1' found no difference between 5.5- and 4-mm scleral incision groups closed with X-shaped sutures with regard to induced vector astigmatism. From the investigations reviewed, it appears that the difference in scleral incision size in comparison groups has to be 2 mm or more to be statistically significant, given the numbers of patients typically enrolled. In Nielsen's8 study, which compared the astigmatic effects of corneal incision phacoemulsification to that of scleral incision phacoemulsification, 3.5-mm tem poral corneal incisions were found to induce a greater change in vector corneal astigmatism at all postopera tive examinations than 3.5-mm corneoscleral inci sions located in the superior meridian. However, there were only 16 eyes in that corneal incision group and 13 eyes in that scleral incision group, and the difference between groups was not statistically signifi cant. Many studies using the case series format have documented the natural history of corneal astigma tism after sutureless, small-incision phacoemulsifica tion. Kondrot16 found a gradual increase in the percentage of patients with 1 diopter or less of astigmatism from postoperative week 1 to postopera tive month 1 to postoperative year 1 examinations, and a corresponding decrease in the number of patients with 1 to 2 diopters or more of astigmatism at each examination. A 5.2-mm scleral tunnel incision was used for surgery in this series. Surprisingly, there was a 10% incidence of hyphema. Menapace and associates17 found a minimal change in vector corneal astigmatism from the postoperative week 1 examina tion to the postoperative month 3 examination (0.23 ± 0.91 diopter vs 0.20 ± 0.69 diopter) in their series of 100 consecutive cases performed through 4-mm scleral tunnel incisions, suggesting that stable refrac tions can be obtained one week postoperatively. In a VOL.121, No. I
small series of 22 consecutive eyes, Feil, Crandall, and Olson18 also found minimal differences in keratometric and topographic astigmatism from postoperative week 1 to postoperative month 1 examinations. The effect of suture placement on 4-mm scleral tunnel incisions was studied by Masket.19 With regard to astigmatism, one year postoperatively there was no difference between 20 eyes closed with sutures and 40 eyes closed without sutures. Ernest, Lavery, and Kiessling20 determined that square scleral tunnel incisions 4 mm wide by 4 mm long, including a 1.5-mm corneal lip, provided the best protection against leakage from external pressure in cadaver eyes. Such incisions do not change the immediate postoperative corneal topography of ca daver eyes, as shown separately by Frieling and Steinert.21 In our scleral tunnel group, incisions were 4.0 mm wide X 3.0 to 3.5 mm long, including a 0.5to 1.0-mm corneal lip. In another study, Ernest, Lavery, and Kiessling22 determined that square 3.2 X 3.2-mm scleral incisions and that square 3.2 X 3.2 and 2 X 2-mm clear corneal incisions provide greater stability and safety than conventional 3.2 X 2-mm clear corneal incisions. The 3.2-mm square is imprac tical, however, because it encroaches on the visual axis, and the 2-mm square is too small for the current generation of phacoemulsification instruments and intraocular lenses. Stepped and hinged techniques were found to provide greater stability than beveled or paracentesis incisions.23 Changes in corneal astigmatism should be ana lyzed by vector means whenever the effects of a particular type of cataract or refractive surgery are being reported. Simple scalar analysis of the data in this study failed to identify a difference in the two incision groups. Computed topography is replacing keratometry as the preferred method for analyzing corneal astigmatism, and it is probably more sensitive to axis than keratometry. Future studies from this group will compare preoperative and postoperative corneal topography. By vector analysis, a greater with-the-rule change in the scleral incision group was identified in the immediate postoperative period, a change that was both statistically and clinically significant. Scleral incision surgery, as performed in this study, was associated with more keratometric instability in the immediate postoperative period than corneal incision surgery.
ASTIGMATISM AFTER SCLERAL AND CORNEAL TUNNEL INCISIONS
63
ACKNOWLEDGMENT
Jeffrey A. Gornbein, Dr.P.H., Department of Biomathematics, UCLA School of Medicine, Los Ange les, California, provided statistical consultation.
REFERENCES 1. Brint SF, Ostrick DM, Bryan JE. Keratometric cylinder and visual performance following phacoemulsification and im plantation with silicone small-incision or poly(methyl methacrylate) intraocular lenses. J Cataract Refract Surg 1991;17:32-6. 2. Gills JP, Sanders DR. Use of small incisions to control induced astigmatism and inflammation following cataract surgery. J Cataract Refract Surg 1991 ; 17 Suppl:740-4. 3. Steinen RF, Brint SF, White SM, Fine IH. Astigmatism after small incision cataract surgery: a prospective, randomized, multicenter comparison of 4- and 6.5-mm incisions. Oph thalmology 1991;98:417-23. 4. Martin RG, Sanders DR, Van der Karr MA, DeLuca M. Effect of small incision intraocular lens surgery on postopera tive inflammation and astigmatism: a study of the A M O SI-18NB small incision lens. J Cataract Refract Surg 1992;18:51-7. 5. el-Maghrahy A, Anwar M, el-Sayyad F, Matheen M, Marzouky A, Gazayerli E, et al. Effect of incision size on early postoperative visual rehabilitation after cataract surgery and intraocular lens implantation. J Cataract Refract Surg 1993;19:494-8. 6. Oshika T, Tsuboi S, Yaguchi S, Yoshitomi F, Nagamoto T, Nagahara K, et al. Comparative study of intraocular lens implantation through 3.2- and 5.5-mm incisions. Ophthal mology 1994;101:1183-90. 7. Levy JH, Pisacano AM, Chadwick K. Astigmatic changes after cataract surgery with 5.1 mm and 3.5 mm sutureless incisions. J Cataract Refract Surg 1994;20:630-3. 8. Nielsen PJ. Prospective evaluation of surgically induced astigmatism and astigmatic keratotomy effects of various self-sealing small incisions. ] Cataract Refract Surg 1995;21:43-8.
64
9. Jaffe NS, Jaffe MS, Jaffe GF. Cataract surgery and its complications, 5th ed. St. Louis: CV Mosby, 1990:114-9. 10. Batschelet E. Circular statistics in biology. London: Academ ic Press, 1981. 11. Hotelling H. Generalization of Student's ratio. A n n Math Stat 1931;2:360-78. 12. Grabow HB. Early results of 500 cases of no-stitch cataract surgery. J Cataract Refract Surg 1991;17 Suppl:726-30. 13. Neumann AC, McCarty GR, Sanders DR, Raanan MG. Small incisions to control astigmatism during cataract sur gery. J Cataract Refract Surg 1989;15:78-84. 14- Pfleger T, Scholz U, Skorpik C. Postoperative astigmatism after no-stitch, small incision cataract surgery with 3.5 mm and 4-5 mm incisions. J Cataract Refract Surg 1994;20: 400-5. 15. Davison JA. Keratometric comparison of 4.0 mm and 5.5 mm scierai tunnel cataract incisions. J Cataract Refract Surg 1993;19:3-8. 16. Kondrot EC. Keratometric cylinder and visual recovery following phacoemulsification and intraocular lens implanta tion using a self-sealing cataract incision. J Cataract Refract Surg 1991;17 Suppl:731-3. 17. Menapace R, Radax U, Amon M, Papapanos P. No-stitch, small incision cataract surgery with flexible intraocular lens implantation. J Cataract Refract Surg 1994;20:534-42. 18. Feil SH, Crandall AS, Olson RJ. Astigmatic decay following small incision, self-sealing cataract surgery. J Cataract Refract Surg 1994;20:40-3. 19. Masket S. O n e year postoperative astigmatic comparison of sutured and unsutured 4-0 mm scierai pocket incisions. J Cataract Refract Surg 1993;19:453-6. 20. Ernest PH, Lavery KT, Kiessling LA. Relative strength of scierai tunnel incisions with internal corneal lips constructed in cadaver eyes. J Cataract Refract Surg 1993;19:457-61. 21. Frieling E, Steinert RF. Intrinsic stability of 'self-sealing' unsutured cataract wounds. Arch Ophthalmol 1993;111: 381-3. 22. Ernest PH, Lavery KT, Kiessling LA. Relative strength of scierai corneal and clear corneal incisions constructed in cadaver eyes. J Cataract Refract Surg 1994;20:626-9. 23. Ernest PH, Fenzl R, Lavery KT, Sensoli A. Relative stability of clear corneal incisions in a cadaver eye model. J Cataract Refract Surg 1995;21:39-42.
AMERICAN JOURNAL OF OPHTHALMOLOGY
JANUARY I 996
• PURPOSE: We compared the changes in corneal astigmatism after phacoemulsification and intraoc ular lens implantation in 93 consecutive eyes with unsutured 4-mm superior scierai tunnel incisions to those through 105 consecutive eyes with unsutured 3.2- to 3.5-mm temporal corneal tunnel incisions. • METHODS: Keratometry measurements were ob tained preoperatively and at postoperative day 1, week 1, and week 6. Group differences in scalar and vector astigmatism were compared by using analysis of variance methods. • RESULTS: Mean scalar astigmatism in the scierai incision group changed from preoperative astigma tism by 0.65 diopter at postoperative day 1, 0.37 diopter at postoperative week 1, and 0.13 diopter at postoperative week 6. Mean scalar astigmatism in the corneal incision group changed from preop erative astigmatism by 0.39 diopter at postopera tive day 1, 0.21 diopter at postoperative week 1, and 0.13 diopter at postoperative week 6. Mean vector astigmatism in the scierai incision group changed 1.26 diopters at 8 0 degrees at postopera tive day 1, 1.05 diopters at 83 degrees at postoper-
Accepted for publication July 20, 1995. From the Jules Stein Eye Institute and the Department of Ophthal mology, University of California at Los Angeles School of Medicine, Los Angeles, California. This study was supported by a grant from the Karl Kirchgessner Foundation, Los Angeles, California. It was pre sented, in part, at the annual meeting of the American Sociery of Cataract and Refractive Surgery, San Diego, California, April 2, 1995. Reprint requests to Kevin M. Miller, M.D., Jules Stein Eye Institute, 100 Stein Plaza, UCLA, Los Angeles, CA 90095-7002; fax: (310) 825-0841; E-mail: [email protected].
VOL.121, No. 1
ative week 1, and 0.42 diopter at 103 degrees at postoperative week 6. Mean vector astigmatism in the corneal incision group changed 0.77 diopter at 90 degrees at postoperative day 1, 0.75 diopter at 88 degrees at postoperative week 1, and 0.61 diopter at 8 9 degrees at postoperative week 6. The differences were statistically significant (P = .003) only by vector analysis at the postoperative day 1 examination. • CONCLUSIONS: We found significantly greater with-the-rule change in astigmatism in the scierai incision group than in the corneal incision group on the first postoperative day. The effect disap peared by the sixth postoperative week.
C
ATARACT EXTRACTION BY THE PHACOEMULSIFI-
cation technique with implantation of foldable intraocular lenses through unsutured scierai tunnel incisions 4 mm or less in width has been shown to induce a smaller change in corneal astigma tism than surgery through larger scierai incisions.1'7 Nielsen8 found a difference in surgically induced astigmatism in a comparison of phacoemulsification through 3.5-mm scierai incisions to that through 3.5-mm corneal incisions, but the study had a limited number of patients and the difference was not statisti cally significant. In this study we compared the astigmatic effects of phacoemulsification and intraoc ular lens implantation through unsutured 4-mm scier ai tunnel incisions located in the superior vertical meridian (12 o'clock meridian) to those through unsutured 3.2- to 3.5-mm corneal tunnel incisions located in a temporal (3 or 9 o'clock) meridian.
© AMERICAN JOURNAL OF OPHTHALMOLOGY i996,-i2i:57-64
57
PATIENTS AND METHODS WE REVIEWED THE RECORDS OF 198 CONSECUTIVE EYES
that underwent phacoemulsification and foldable in traocular lens implantation by one of us (K.M.M.). Scierai incision surgery had been performed on 93 eyes, and corneal incision surgery on 105 eyes. Most of the scierai incision operations were performed before the first corneal incision procedure was per formed. In all of the eyes that underwent scierai incision, anesthesia was attained by posterior orbital injection of 3 to 5 ml of a 50:50 mixture of 2% lidocaine with epinephrine 1:100,000 and 0.75% bupivacaine with one ampule (150 USP units) of hyaluronidase. In the eyes that underwent corneal incision, anesthesia was attained either by posterior orbital injection or by topical administration of four drops of 1 % tetracaine. In the eyes that underwent scierai incision, a twoto three-clock hour limbal peritomy centered at the 12 o'clock meridian was fashioned with Wescott scissors and Colibri forceps. Wetfield bipolar cautery was applied to coagulate episcleral vessels. A 4-mmlong partial-thickness scierai groove was fashioned 2.5 mm posterior to the 12 o'clock meridian corneoscleral limbus by using a No. 69 Beaver blade. A scierai tunnel was dissected anteriorly 0.5 to 1.0 mm into clear cornea with the same blade. A second instrument port was fashioned at the 2:30 o'clock meridian with a 15-degree angle blade, and the anterior chamber was re-entered through the scierai tunnel incision by using a straight 3.2-mm keratome to form a three-stepped self-sealing tunnel incision. Hyaluronic acid was injected to reform the anterior chamber. In the patients who underwent corneal incision, a 3.2- to 3.5-mm partial-thickness corneal groove was fashioned at the temporal corneoscleral limbus in the 3 or 9 o'clock meridian with a 1-mm two-step diamond knife with the blade set to a depth of 300 to 400 μιτι. A second instrument site was established for the left hand at either the 10:30 or the 4:30 o'clock meridian, depending on the eye undergoing surgery. The same diamond knife was used with the blade fully extended. Hyaluronic acid was injected through the side port incision to increase intraocular pressure
58
and facilitate the incision. The eye was re-entered through the temporal corneal incision with a 3.5/3.2mm or 3.2/2.7-mm trapezoidal diamond keratome to form a three-stepped self-sealing tunnel incision. After the incision was made, a continuous curvi linear capsulorhexis was started by using a bent 25-gauge cystotome and was completed with Utrata forceps. Hydrodissection was performed by injecting balanced saline solution between the nucleus and cortex through a 27-gauge cannula. The nucleus was removed with an Alcon Series Ten Thousand phaco emulsification unit by using a standard two-handed nuclear cracking technique (Alcon Surgical, Irvine, California). Residual cortex was removed and the posterior capsule was polished, if necessary, by using the irrigation and aspiration probe. The capsule and anterior chamber were filled with sodium hyaluronate, and a foldable silicone intraocular lens was implanted in the capsule through an unenlarged incision. After removal of the sodium hyaluronate, the anterior chamber was reformed with balanced saline solution, and the wound was tested with absorbent sponges. In the eyes that underwent corneal incision, if the wound was not self-sealing and watertight, the corne al stroma was hydrated with additional balanced saline solution. In the eyes that underwent scierai incision, the conjunctiva was brought over the limbal area and secured with bipolar forceps cautery. Keratometry was performed preoperatively, and at postoperative day 1, postoperative week 1, and post operative week 6, with a manual keratometer. In the scalar analysis, astigmatism was calculated as the absolute value of the difference in keratometric measurements. Changes in scalar astigmatism were calculated by subtracting the preoperative amount from that measured at each postoperative examina tion. In the vector analysis, the magnitude assigned to the astigmatism vector was that calculated in the scalar analysis. The angle assigned to the astigmatism vector was that of the greater keratometric value. Changes in vector astigmatism were determined by subtracting the preoperative vector from that mea sured at each postoperative examination with a double-angle polar coordinate system for analysis of toric surfaces (Fig. 1). Custom software for perform ing the vector calculations was written and imple-
AMERICAN JOURNAL OF OPHTHALMOLOGY
JANUARY 1996
one-way multiple analysis of variance, with the F statistic based on Hotelling's T2-test method." (0 α> Q. O
RESULTS
S
0/180 c
90'
With-the-rule
135 u
Against-the-rule
Fig. 1 (Gross and Miller). Graphic method for calculat ing the corneal astigmatism change vector. Representa tive preoperative and postoperative astigmatism vectors are plotted in polar coordinates at twice the actual angle. The change vector sums with the preoperative vector to produce the postoperative vector. It can be drawn ex tending from the plot origin. Vectors falling on the left side of the graph represent with-the-rule astigmatism. Vectors falling on the right side of the graph represent against-the-rule astigmatism.
mented on a Microsoft Excel spreadsheet according to the general methodology of Jaffe, Jaffe, and JafFe.9 Scleral and corneal incision group mean scalar astig matism changes were compared by using one-way analysis of variance. At each examination, average change vectors were calculated by converting the individual polar vectors into rectangular coordinate system components (rx, r y ), calculating averages (rx = 1/ΠΣΓΧ, ry = l/n2r y ), and reverse transforming the resultant [r = V (rx2 4ry2), Θ = tan" 1 (r x /r y )]. Approximate 95% confidence intervals for the average change vector magnitudes and angles were computed by forming the confidence ellipse (Fig. 2) in terms of rx and ry and then transforming the ellipse to polar coordinates (Fig. 3). 10 The ellipse model assumes that the rx and ry means have an approximate bivariate gaussian distri bution. Scleral and corneal incision group mean astigmatism change vectors were compared using
VOL.121,
No. 1
IN THE SCALAR ANALYSIS, THE PREOPERATIVE CORNEAL
astigmatism in the 4-mm scleral incision and 3.2- to 3.5-mm corneal incision groups was not statistically different (P = .126). There was slightly more surgical ly induced astigmatism in the scleral incision group than in the corneal incision group at postoperative day 1 and postoperative week 1 (Fig. 4), but the difference was not statistically significant either at these examinations or at the postoperative week 6 examination (Table 1). The changes in scalar astig matism at postoperative week 6 were virtually identi cal in the two groups. In the vector analysis, there was a mean with-therule change in corneal astigmatism in the scleral and corneal incision groups at all postoperative examina tions (Fig. 5) that was statistically greater (P = .003) for the scleral incision group on the postoperative day 1 examination (Table 2). In both groups there was a gradual decay in the average amount of with-the-rule
!
r (diopters)
Fig. 2 (Gross and Miller). A representative average astigmatism change vector plotted in Cartesian coordi nates surrounded by the 95% confidence ellipse. rx is the projection of the vector onto the axis at 0/180 degrees. ry is the projection of the vector onto the axis at 45 degrees (in a double-angle polar coordinate system).
ASTIGMATISM AFTER SCLERAL AND CORNEAL TUNNEL INCISIONS
59
I 90'
0/180
135° Fig. 3 (Gross and Miller). A representative change vector plotted in a double-angle polar coordinate system sur rounded by the approximate 95% confidence interval for vector magnitude (r95) and angle (θ95). The area repre sented by the shaded wedge should include 95% of all change vectors in the group.
vector astigmatism change (Fig. 6). The postoperative drift toward against-the-rule corneal astigmatism was more pronounced for the scierai incision group. At the postoperative week 6 examination, when final prescriptions for spectacles were written, there was an average vector change of 0.42 diopter in the scierai incision group and 0.61 diopter in the corneal incision group.
change as it steepened the corneal meridian 90 degrees from the incision. The astigmatic change caused by the temporal corneal cataract incision was what one would have expected for a peripheral arcuate keratotomy incision. The small decrease in astigma tism that was observed on subsequent examinations was the result of wound healing. The with-the-rule or with-the-wound change in astigmatism that occurred in the scierai incision group, however, is not analo gous to the change induced by arcuate keratotomy, because the astigmatism would then have gone against the wound, as it did in the corneal incision group. Rather, the early with-the-wound change can probably be explained either by the episcleral cautery, by the edema that developed in the scierai and corneal tissue surrounding the scierai tunnel, by the compressive effects of the eyelids, or by a combination of factors. In most respects, except for the application of cautery (including wound architecture, wound hydration, operating time, absence of suture place ment, and postoperative management), the two groups were treated similarly. Cataract surgeons who use cautery during surgery can often see the episclera contracting beneath the tip of the wet-field probe. In a large series of patients, Grabow12 found a 50% reduction in induced astigmatism at the postoperative week 1 examination when cautery was limited to the
0.8 0.7
• — Scierai incision * — Corneal incision
0.6
DISCUSSION IN THIS STUDY WE FOUND A SIGNIFICANT DIFFERENCE
between unsutured 4-mm scierai tunnel incisions and unsutured 3.2- to 3.5-mm corneal tunnel incisions with respect to early postoperative keratometric astig matism by vector methods. The difference disap peared by the sixth postoperative week. There was no difference between the two groups when the change in astigmatism was measured by scalar methods. Both surgical procedures produced an immediate with-the-rule change in astigmatism that decreased on subsequent examinations. For the temporal corne al incision group, the with-the-rule change in astig matism was simultaneously an against-the-wound
60
1
10 Days after surgery
100
Fig. 4 (Gross and Miller). Average change in scalar astigmatism (± S.E.) in the s*cleral and corneal incision groups at postoperative day 1, postoperative week 1, and postoperative week 6. The differences between the two groups were not statistically significant at any postopera tive time point.
AMERICAN JOURNAL OF OPHTHALMOLOGY
JANUARY 1996
Scierai Incision Postoperative week 6
Postoperative week 1
Postoperative day 1
45°
o Q
90°
0/180 '
0/180 ° 90"
0/180° 90"
135 l
135'
135'
Corneal Incision
I
o a
0/180° 90°
135 °
0/180°
90°
0/180"
135 °
135 °
Fig. 5 (Gross and Miller). Individual change vectors represented as points at each postoperative examination. The scierai incision group is shown in the upper half, and the corneal incision group in the lower half. Most change vectors are with-the-rule at each postoperative examination, although there is a clustering toward the origin with less dispersion over time.
Postoperative day 1
Postoperative week 1
0/180°
135'
Postoperative week 6
0/180°
0/180"
135"
135° lόlj Scierai Incision
LV Corneal Incision
Fig. 6 (Gross and Miller). Average change vectors represented as 95% confidence intervals for magnitude and axis at each postoperative examination. The areas represented by the shaded wedges should include 95% of all change vectors in each group. The difference at postoperative day 1 is significant (P = .003) and the difference at postoperative week 1 approaches significance (P = .08).
VOL.121, No. I
ASTIGMATISM AFTER SCLERAL AND CORNEAL TUNNEL INCISIONS
61
TABLE 1 CHANGE IN SCALAR CORNEAL ASTIGMATISM WITH SCLERAL AND CORNEAL TUNNEL INCISION PHACOEMULSIFICATION AND INTRAOCULAR LENS IMPLANTATION ASTIGMATISM CHANGE IN SCLERAL INCISION GROUP (D)
ASTIGMATISM CHANGE IN CORNEAL INCISION GROUP (D)
p
NO.
MEAN ± S.E.
NO.
MEAN ± S.E.
F STATISTIC
VALUE
91 90 73
0.651 ± 0.136 0.371 ± 0.107 0.134 ± 0.071
101 102 87
0.388 ± 0.097 0.211 ± 0.082 0.130 ± 0.077
2.56 1.46 0.00
.111 .229 .966
Postoperative day 1 Postoperative week 1 Postoperative week 6
episclera posterior to the insertion of Tenon's fascia and not carried all the way to the corneoscleral limbus. The relatively marked decrease in astigmatism observed on follow-up examinations in our scierai incision group is most readily explained by scierai relaxation, as either the cautery effect or stromal edema wore off. If it is possible for these effects to wear off completely, and if 4-mm scierai incisions produce the same amount of corneal relaxation as 3.2- to 3.5-mm corneal incisions, we would anticipate an eventual against-the-wound astigmatic drift of about 0.6 diopter, as occurred in the corneal incision group. The follow-up interval of this study was insufficient to make this determination. The astigmatic benefits of unsutured small-incision phacoemulsification incisions have been documented extensively. Levy, Pisacano, and Chadwick 7 found a vector astigmatism change at one month of 1.35 ± 0.98 diopters in a 5.1-mm scierai incision group in contrast to 0.95 ± 0.62 diopter in a 3.5-mm scierai incision group. The difference between their two groups was statistically significant. Oshika and associ ates6 found a vector astigmatism change at one month
of 0.30 ± 0.68 diopter in a 5.5-mm scierai incision group in contrast to 0.08 ± 0.58 diopter in a 3.2-mm scierai incision group. This difference was statistically significant. Brint, Ostrick, and Bryan1 found a scalar astigmatism change at six weeks of 1.03 ± 0.51 diopters in a 4-mm scierai incision group in contrast to 1.33 ± 0.79 diopters in a 7-mm scierai incision group. This difference was also statistically significant. El-Maghraby and associates5 found less induced vec tor astigmatism at the postoperative day 2 and postop erative week 1 examinations in a group of patients who had surgery through 3.5-mm scierai incisions than in patients who had surgery through 5.5- and 6.5-mm scierai incisions. All the incisions in their study, even the 3.5-mm group, were closed with X-shaped sutures. Gills and Sanders2 found less vector astigmatism two to three weeks after surgery in a group of 55 eyes receiving 3-mm scierai incisions than in a group of 48 eyes receiving 6- to 7-mm incisions. Other investigators observed similar effects with small scierai incisions.3,4 By contrast, Neumann and associates13 found no difference between 56 eyes undergoing 6-mm scierai
TABLE 2 CHANGE IN VECTOR CORNEAL ASTIGMATISM WITH SCLERAL AND CORNEAL TUNNEL INCISION PHACOEMULSIFICATION AND INTRAOCULAR LENS IMPLANTATION ASTIGMATISM CHANGE IN SCLERAL INCISION GROUP (D)
AXIS (DEGREES)
ASTIGMATISM CHANGE IN CORNEAL INCISION GROUP (D)
AXIS (DEGREES)
95% CONFIDENCE 95% CONFIDENCE 95% CONFIDENCE 95% CONFIDENCE F P NO. MEAN INTERVAL MEAN INTERVAL NO. MEAN INTERVAL MEAN INTERVAL STATISTIC VALUE
Postoperative day 1 34 1.26 Postoperative week 1 36 1.05 Postoperative week 6 28 0.424
62
0.86-1.67 0.67-1.42 0.44-0.72
80.0 83.0 103
69-92 70-100 58-161
95 0.766 99 0.749 84 0.608
0.59--0.94 0.59-0.91 0.46-0.76
AMERICAN JOURNAL OF OPHTHALMOLOGY
90.0 87.9 88.9
79-102 81-95 78-100
6.06 2.58 1.38
.0031 .0799 .256
JANUARY
1996
incision surgery and 67 eyes undergoing 3- to 4-mm scleral incision surgery with regard to vector astigma tism at three months. Grabow12 found no difference between 250 eyes undergoing 5.2-mm scleral incision surgery and 280 eyes undergoing 4-mm scleral inci sion surgery with respect to scalar astigmatism at one month. No difference in postoperative vector astigma tism was found at any postoperative examination in a study by Pfleger, Scholz, and Skorpik,14 who com pared a 4.5-mm scleral incision group with a 3.5-mm scleral incision group. Davison1' found no difference between 5.5- and 4-mm scleral incision groups closed with X-shaped sutures with regard to induced vector astigmatism. From the investigations reviewed, it appears that the difference in scleral incision size in comparison groups has to be 2 mm or more to be statistically significant, given the numbers of patients typically enrolled. In Nielsen's8 study, which compared the astigmatic effects of corneal incision phacoemulsification to that of scleral incision phacoemulsification, 3.5-mm tem poral corneal incisions were found to induce a greater change in vector corneal astigmatism at all postopera tive examinations than 3.5-mm corneoscleral inci sions located in the superior meridian. However, there were only 16 eyes in that corneal incision group and 13 eyes in that scleral incision group, and the difference between groups was not statistically signifi cant. Many studies using the case series format have documented the natural history of corneal astigma tism after sutureless, small-incision phacoemulsifica tion. Kondrot16 found a gradual increase in the percentage of patients with 1 diopter or less of astigmatism from postoperative week 1 to postopera tive month 1 to postoperative year 1 examinations, and a corresponding decrease in the number of patients with 1 to 2 diopters or more of astigmatism at each examination. A 5.2-mm scleral tunnel incision was used for surgery in this series. Surprisingly, there was a 10% incidence of hyphema. Menapace and associates17 found a minimal change in vector corneal astigmatism from the postoperative week 1 examina tion to the postoperative month 3 examination (0.23 ± 0.91 diopter vs 0.20 ± 0.69 diopter) in their series of 100 consecutive cases performed through 4-mm scleral tunnel incisions, suggesting that stable refrac tions can be obtained one week postoperatively. In a VOL.121, No. I
small series of 22 consecutive eyes, Feil, Crandall, and Olson18 also found minimal differences in keratometric and topographic astigmatism from postoperative week 1 to postoperative month 1 examinations. The effect of suture placement on 4-mm scleral tunnel incisions was studied by Masket.19 With regard to astigmatism, one year postoperatively there was no difference between 20 eyes closed with sutures and 40 eyes closed without sutures. Ernest, Lavery, and Kiessling20 determined that square scleral tunnel incisions 4 mm wide by 4 mm long, including a 1.5-mm corneal lip, provided the best protection against leakage from external pressure in cadaver eyes. Such incisions do not change the immediate postoperative corneal topography of ca daver eyes, as shown separately by Frieling and Steinert.21 In our scleral tunnel group, incisions were 4.0 mm wide X 3.0 to 3.5 mm long, including a 0.5to 1.0-mm corneal lip. In another study, Ernest, Lavery, and Kiessling22 determined that square 3.2 X 3.2-mm scleral incisions and that square 3.2 X 3.2 and 2 X 2-mm clear corneal incisions provide greater stability and safety than conventional 3.2 X 2-mm clear corneal incisions. The 3.2-mm square is imprac tical, however, because it encroaches on the visual axis, and the 2-mm square is too small for the current generation of phacoemulsification instruments and intraocular lenses. Stepped and hinged techniques were found to provide greater stability than beveled or paracentesis incisions.23 Changes in corneal astigmatism should be ana lyzed by vector means whenever the effects of a particular type of cataract or refractive surgery are being reported. Simple scalar analysis of the data in this study failed to identify a difference in the two incision groups. Computed topography is replacing keratometry as the preferred method for analyzing corneal astigmatism, and it is probably more sensitive to axis than keratometry. Future studies from this group will compare preoperative and postoperative corneal topography. By vector analysis, a greater with-the-rule change in the scleral incision group was identified in the immediate postoperative period, a change that was both statistically and clinically significant. Scleral incision surgery, as performed in this study, was associated with more keratometric instability in the immediate postoperative period than corneal incision surgery.
ASTIGMATISM AFTER SCLERAL AND CORNEAL TUNNEL INCISIONS
63
ACKNOWLEDGMENT
Jeffrey A. Gornbein, Dr.P.H., Department of Biomathematics, UCLA School of Medicine, Los Ange les, California, provided statistical consultation.
REFERENCES 1. Brint SF, Ostrick DM, Bryan JE. Keratometric cylinder and visual performance following phacoemulsification and im plantation with silicone small-incision or poly(methyl methacrylate) intraocular lenses. J Cataract Refract Surg 1991;17:32-6. 2. Gills JP, Sanders DR. Use of small incisions to control induced astigmatism and inflammation following cataract surgery. J Cataract Refract Surg 1991 ; 17 Suppl:740-4. 3. Steinen RF, Brint SF, White SM, Fine IH. Astigmatism after small incision cataract surgery: a prospective, randomized, multicenter comparison of 4- and 6.5-mm incisions. Oph thalmology 1991;98:417-23. 4. Martin RG, Sanders DR, Van der Karr MA, DeLuca M. Effect of small incision intraocular lens surgery on postopera tive inflammation and astigmatism: a study of the A M O SI-18NB small incision lens. J Cataract Refract Surg 1992;18:51-7. 5. el-Maghrahy A, Anwar M, el-Sayyad F, Matheen M, Marzouky A, Gazayerli E, et al. Effect of incision size on early postoperative visual rehabilitation after cataract surgery and intraocular lens implantation. J Cataract Refract Surg 1993;19:494-8. 6. Oshika T, Tsuboi S, Yaguchi S, Yoshitomi F, Nagamoto T, Nagahara K, et al. Comparative study of intraocular lens implantation through 3.2- and 5.5-mm incisions. Ophthal mology 1994;101:1183-90. 7. Levy JH, Pisacano AM, Chadwick K. Astigmatic changes after cataract surgery with 5.1 mm and 3.5 mm sutureless incisions. J Cataract Refract Surg 1994;20:630-3. 8. Nielsen PJ. Prospective evaluation of surgically induced astigmatism and astigmatic keratotomy effects of various self-sealing small incisions. ] Cataract Refract Surg 1995;21:43-8.
64
9. Jaffe NS, Jaffe MS, Jaffe GF. Cataract surgery and its complications, 5th ed. St. Louis: CV Mosby, 1990:114-9. 10. Batschelet E. Circular statistics in biology. London: Academ ic Press, 1981. 11. Hotelling H. Generalization of Student's ratio. A n n Math Stat 1931;2:360-78. 12. Grabow HB. Early results of 500 cases of no-stitch cataract surgery. J Cataract Refract Surg 1991;17 Suppl:726-30. 13. Neumann AC, McCarty GR, Sanders DR, Raanan MG. Small incisions to control astigmatism during cataract sur gery. J Cataract Refract Surg 1989;15:78-84. 14- Pfleger T, Scholz U, Skorpik C. Postoperative astigmatism after no-stitch, small incision cataract surgery with 3.5 mm and 4-5 mm incisions. J Cataract Refract Surg 1994;20: 400-5. 15. Davison JA. Keratometric comparison of 4.0 mm and 5.5 mm scierai tunnel cataract incisions. J Cataract Refract Surg 1993;19:3-8. 16. Kondrot EC. Keratometric cylinder and visual recovery following phacoemulsification and intraocular lens implanta tion using a self-sealing cataract incision. J Cataract Refract Surg 1991;17 Suppl:731-3. 17. Menapace R, Radax U, Amon M, Papapanos P. No-stitch, small incision cataract surgery with flexible intraocular lens implantation. J Cataract Refract Surg 1994;20:534-42. 18. Feil SH, Crandall AS, Olson RJ. Astigmatic decay following small incision, self-sealing cataract surgery. J Cataract Refract Surg 1994;20:40-3. 19. Masket S. O n e year postoperative astigmatic comparison of sutured and unsutured 4-0 mm scierai pocket incisions. J Cataract Refract Surg 1993;19:453-6. 20. Ernest PH, Lavery KT, Kiessling LA. Relative strength of scierai tunnel incisions with internal corneal lips constructed in cadaver eyes. J Cataract Refract Surg 1993;19:457-61. 21. Frieling E, Steinert RF. Intrinsic stability of 'self-sealing' unsutured cataract wounds. Arch Ophthalmol 1993;111: 381-3. 22. Ernest PH, Lavery KT, Kiessling LA. Relative strength of scierai corneal and clear corneal incisions constructed in cadaver eyes. J Cataract Refract Surg 1994;20:626-9. 23. Ernest PH, Fenzl R, Lavery KT, Sensoli A. Relative stability of clear corneal incisions in a cadaver eye model. J Cataract Refract Surg 1995;21:39-42.
AMERICAN JOURNAL OF OPHTHALMOLOGY
JANUARY I 996