- Email: [email protected]
Postoperative intraocular lens rotation
Postoperative Intraocular Lens Rotation A Randomized Comparison of Plate and Loop Haptic Implants C. K. Patel, FRCOphth, Sue Ormonde, FRCOphth, Paul H. Rosen, FRCOphth, Anthony J. Bron, FRCOphth Objective: To compare the postoperative rotation of plate and loop haptic implants of spherical power to ascertain the optimal design appropriate for toric intraocular lenses (IOLs). Design: Randomized, controlled trial. Participants: Forty-eight patients attending for routine cataract surgery by phacoemulsification. Method: Patients with cataract as the only ocular disease were randomly implanted with plate or loop haptic implants after uncomplicated phacoemulsification. The baseline position of the IOL was determined from a video frame acquired at the conclusion of surgery. Postoperative IOL position was documented using digital retroillumination images at 2 weeks and 6 months after surgery. Capsular fusion patterns were recorded using slit-lamp biomicroscopy. Correlation of IOL rotation with axial length, capsular contraction, and fusion was attempted. Main Outcome Measures: Early IOL rotation, occurring between surgery and 2 weeks after surgery, was graded as mild (⬍10°), moderate (10°⬍ to ⬍30°), or severe (⬎30°) by a semiobjective online comparison of the images. Late IOL rotation, occurring between 2 weeks and 6 months, was measured more precisely using software developed specifically for the study. Results: Twenty-three patients were allocated the loop haptic and 25 the plate haptic IOL. The groups were comparable for demographic variables and numbers of patients excluded from analysis (P ⬎ 0.05). Five (24%) of 21 of plate haptic IOLs underwent severe early rotation compared to 2 (9%) of 22 loop haptics (P ⫽ 0.36). The median late rotation was 6.8° for loop haptics compared to 0.6° for plate haptics (P ⫽ 0.0073). Between 2 weeks and 6 months, anticlockwise rotation had occurred in 16 (89%) of 18 loop haptic IOLs compared to 11 (52%) of 21 plate haptic IOLs (P ⫽ 0.0081). Conclusions: Plate haptic IOLs show greater rotational stability than do loop haptics made from polypropylene once capsular fusion has taken place. Loop haptics invariably rotate anticlockwise after 2 weeks. Ophthalmology 1999;106:2190 –2196 The surgical management of cataract is a two-stage procedure. The first involves removal of the opaque lens and the second involves correction of the resultant refractive error. The evolution of astigmatically neutral cataract incisions has led to patients expecting emmetropia as the optimal refractive outcome when corneal astigmatism does not coexist. In 20% of patients in whom astigmatism is manifest,1 a toric implant can also yield emmetropia without recourse to additional corneal surgery provided the lens does not rotate after surgery. The peer-reviewed literature is unclear regarding whether the loop or plate haptic design is the more stable. The purpose of this prospective, randomized,
Originally received: November 10, 1998. Revision accepted: July 7, 1999. Manuscript no. 98594. From Oxford Eye Hospital, Radcliffe Infirmary, Oxford, England. Presented in part at the American Academy of Ophthalmology annual meeting, New Orleans, Louisiana, November 1998. Reprint requests to C. K. Patel, FRCOphth, c/o Office of Alan Berger, MD, Sunny Brook Health Science Centre, 2075 Bayview Ave., Suite M1-202A, Toronto, Ontario, Canada M4N 3MS. E-mail: [email protected].
2190
controlled study is to compare postoperative rotation of silicone intraocular lenses (IOLs) of spherical power, engineered either as a one-piece plate haptic or a three-piece loop design.
Methods Patients attending for cataract surgery at the Oxford Eye Hospital, a teaching center in the United Kingdom, were prospectively assessed for the study. Exclusion criteria included previous intraocular surgery, evidence of previous uveitis, and presence of pseudoexfoliation, iridodonesis, diabetes, and marked corneal opacity. Patients were included in the study if they gave informed consent. Before surgery, the pupils were dilated with G cyclopentolate 1% and G phenylephrine 10%; G amethocaine 0.5% was used to achieve ocular surface anesthesia. In the operating room, with the patient recumbent, the head was stabilized in a standardized position before a surgeon (CKP) dried and marked 6-o’clock meridian on the limbal conjunctiva. A standard peribulbar anesthetic was then administered. A single surgeon (CKP) performed surgery using a phacotechnique described elsewhere (Patel et al. Post-
Patel et al 䡠 Rotation of Plate and Loop Haptic IOLs
Figure 1. A, straight line connecting the haptic insertions is copied and pasted such that it bisects the corresponding line in the 2-week image. Angle @ is used to grade rotation. “X” is the ink mark defining the 6-o‘clock meridian. B, orientation of the plate haptic lens is defined by selecting one end of its minor axis. The axis is then constructed by drawing right-angled chords across the optic. The 6-o‘clock meridian (X) is here highlighted by placing a lens dialer over the fading ink mark.
operative changes in the capsulorrhexis aperture: a prospective, randomised comparison between loop and plate haptic silicone intraocular lenses. Eye 1999 [accepted, September 1999]). Cases in which the capsulorrhexis was incomplete or in which posterior capsule rupture occurred were excluded before randomization. A permuted block-restricted randomization dictated the IOL that was implanted. The plate haptic lens (C10UB, Bausch & Lomb Surgical, Rochester, NY) was a single-piece plate haptic silicone IOL, 10.5 mm in length, with a 6-mm optic and 0.3-mm diameter fenestrations. The SI30NB lens (Allergan-Medical-Optics, Irvine, CA) was a 13.5-mm long, three-piece lens with a 6-mm silicone optic and “C” shaped polypropylene loops. A video clip of the eye was acquired before a subconjunctival injection of 2 mg betamethasone and 20 mg of gentamicin was administered. After surgery, g dexamethasone 0.1% and g chloramphenicol 0.5% were used four times a day for 2 weeks and then reduced over 2 weeks or as clinically indicated. Patients were assessed at 2 weeks and 6 months after surgery. The pupil was dilated with G tropicamide 1% and G phenylephrine 10%. The relations of the IOL and capsule were documented in detail using biomicroscopy. Particular attention was paid to the manner in which the anterior and posterior capsules fused after surgery. A digital retroillumination image of the red reflex was acquired using a computerized system (CASE 2000, Marcher Enterprises, Hereford, UK) previously developed for cataract analysis.2 It had been modified for IOL assessment by the company in conjunction with one of the authors (CKP). Early rotation of the IOL occurring between the conclusion of surgery and the 2-week postoperative visit was graded by online comparison of the operative video frame and digital retroillumination images. Registration of the two images was necessary when local anesthetic had caused the eye to rotate. This involved digitally rotating the video image so that the ink mark on the limbal conjunctiva was restored to the 6-o’clock meridian. Points on the
IOL to be tracked after surgery differed for the two lenses. For the loop haptic IOL, a line was drawn on the computer linking the haptic insertions (Fig 1A). The line on the video frame was copied and pasted to bisect the line on the 2-week image. The acute angle formed by the lines was assumed to represent the magnitude of rotation and measured using NIH Image 1.47 software (developed by the National Institutes of Health, Bethesda, MD). Plate haptic rotation was measured by tracking its minor axis. One end of the axis was defined by estimating the point on the circumference of the optic at which the tangeant to the optic was parallel to the long axis of the IOL. The minor axis could then be defined by constructing a right-angled triangle as shown in Figure 1B. The magnitude of rotation was graded as mild (⬍10°), moderate (10°⬍ to ⬍30°), or severe (⬍30°). Late rotation was measured more precisely. Perfect registration of images was assumed because a central fixation light on the retroillumination camera ensured repeatable ocular alignment. The CASE 2000 software determined the center of the optic when the latter’s boundary was mapped with a circle. Another cursor marked a peripheral point on the optic. The software calculated the angular subtense of the line connecting these two points with reference to the vertical (Figs 2A, B, and 3A, B). The intraobserver repeatability of determining IOL orientation was tested in 14 patients with each lens design before the study. For early rotation, the mean difference and 95% confidence interval (CI) in test–retest results of orientation for the loop IOL was ⫺0.72° (standard deviation [SD], 2.3; upper 95% CI ⫽ 0.83°; lower 95% CI ⫽ ⫺2.28°) and for the plate haptic was 0.46° (SD, 3.12; upper 95% CI ⫽ 2.55°; lower 95% CI ⫽ ⫺1.62°). Grading this rotation resulted in a 100% match in test–retest data. For late rotation, the mean difference for the loop IOL was 0.35° (SD, 1.29; upper 95% CI ⫽ 1.09°; lower 95% CI ⫽ ⫺0.4°) and for the plate haptic was ⫺0.28° (SD, 2.38; upper 95% CI ⫽ 1.10°; lower 95% CI ⫽ ⫺1.65°).
Figure 2. A, 2-week and 6-month retroillumination images of the Chiron-Vision C10UB plate haptic lens are shown to the left and right, respectively. B, late rotation is defined as @’minus@. The angles are calculated by the CASE 2000 software after the center of the optic and a peripheral landmark are defined.
2191
Ophthalmology Volume 106, Number 11, November 1999 was excluded because of inadequate mydriasis and an additional two in each group for failure to attend the 6-month visit. One patient in the loop haptic group developed asymmetric fixation of haptics and another died before the 6-month visit. Another exclusion in the plate haptic group resulted from the absence of the 2-week image.
Early Rotation For the loop haptic IOL, 13 (59%) of 22, 7 (32%) of 22, and 2 (9%) of 22 underwent mild, moderate, and severe rotation, respectively, contrasting with 9 (43%) of 21, 7 (33%) of 21, and 5 (24%) of 21 for the plate haptic IOL (P ⫽ 0.36). In 10 (45%) of 22 and 10 (48%) of 21 cases, the rotation was clockwise for the loop and plate haptic IOLs, respectively (P ⫽ 0.86).
Late Rotation
Figure 3. A, 2-week and 6-month retroillumination images of the AMO SI30NB loop haptic lens are shown to the left and right, respectively. B, late rotation is defined as @’minus@. The angles are calculated by the CASE 2000 software after the center of the optic and a peripheral landmark are defined.
Statistical analysis was performed on a Macintosh computer (Apple, Cupertino, CA) using JMP software (SAS, Cary, NC). Categoric data were tested using the chi-square test. Nonparametric analysis was used to compare the median late rotation and one-way analysis of variance to compare the relationship of axial length to early rotation. A P value of 0.05 was considered significant.
Results Fifty-four patients entered the study. Five patients had an incomplete capsulorrhexis and 1 patient had a posterior capsule rupture, leaving 48 eyes for randomization.
Demographic Data Twenty-three patients received the loop haptic lens and 25 the plate haptic design. There was no statistically significant difference between the groups for mean age, laterality, or gender distribution (Patel et al. Post-operative changes in the capsulorrhexis aperture: a prospective, randomised comparison between loop and plate haptic silicone intraocular lenses. Eye 1999 [accepted, September, 1999]). The number of patients excluded from analysis for early and late rotation did not differ significantly between groups (P ⬎ 0.05). Four patients with plate haptic IOLs were excluded when early rotation was analyzed: one patient in whom the pupil dilated poorly at 2 weeks, another patient whose 2-week image had not been taken because of equipment failure, and two patients in whom the video images were of poor quality. One patient, for whom a video image had not been captured, was the only exclusion from the loop haptic group. For late rotation, one patient in each group
2192
The distribution of the late rotation is shown in Figures 4A and B. A mean rotation of 7.61° (7.25° SD) occurred for the loop haptic IOL contrasting with a mean rotation of 1.8 (12.9° SD) for the plate haptic. The median rotation was 6.8° and 0.6° for loop and plate haptics, respectively (P ⫽ 0.0073). Sixteen (89%) of 18 loop haptic IOLs underwent anticlockwise rotation compared to 11 (52%) of 21 plate haptic IOLs (P ⫽ 0.0081). The modal pattern of capsular fusion observed at 2 weeks for plate (11 [46%] of 24) and loop haptics (16 [73%] of 22) is shown in Figures 5A and B. No comments were made regarding fusion for one patient in each group. The fusion, for the plate haptic, was symmetric and complete to the edge of the lens by 2 weeks. For the loop lens, the fusion was asymmetric beginning behind the convexity of the loop and proceeding clockwise. The pupil was dilated inadequately to allow assessment of fusion in 2 (8%) of 24 and 3 (14%) of 22 patients with plate and loop haptics, respectively. In 6 (25%) of 24 patients with plate haptics, the characteristic fusion pattern was only observed on one side of the lens, the other side being obscured by the pupil. For 5 (21%) of 24 plate haptics, capsular fusion was not complete. Two of these cases showed severe early rotation. It is noteworthy that the only plate haptic IOL to show severe movement between 2 weeks and 6 months also had incomplete fusion. For the loop haptic, fusion was complete for 360° in two patients, and postoperative fibrin accumulation prevented adequate assessment in another 3 (13%) of 22 patients. No significant relationship was detected between rotation and axial length, initial alignment of the lens, or capsular contraction, which has been the subject of a previous report (Patel et al. Post-operative changes in the capsulorrhexis aperture: a prospective, randomised comparison between loop and plate haptic silicone intraocular lenses. Eye 1999 [submitted]) (P ⬎ 0.05). If rotation in excess of 30° was considered for the plate haptic lens over the 6-month follow-up, a tenuous link existed between axial length and rotation (P ⫽ 0.12) (Fig 6).
Discussion Astigmatism occurs in approximately 22% of patients with cataract.1 Achieving emmetropia for these cases has become possible because of the evolution of astigmatically neutral cataract incisions. One option is to combine corneal refractive surgery with the use of a spherical implant. The drawbacks of this approach are unpredictability and the risk of
Patel et al 䡠 Rotation of Plate and Loop Haptic IOLs
Figure 4. A, late rotation for Chiron-Vision C10UB. B, late rotation for AMO SI30NB. Rotation in degrees is plotted on the vertical. A positive number denotes anticlockwise rotation. Zero represents the position of a patient’s lens at 2 weeks, and the top of the bar represents the magnitude and direction of rotation.
violating corneal anatomy. A toric IOL would overcome these drawbacks provided it did not rotate after surgery. The magnitude of cylinder corrected is inversely proportional to the degree of axis misalignment between 0° and 30°.3 Rotation in excess of 30° accentuates preoperative astigmatism and is unacceptable for toric IOLs. Rotation less than 30°, although not providing full correction of the cylinder, has been well-tolerated in clinical studies involving implantation of 2- and 3-diopter toric IOLs.4,5 Such rotation is, however, less likely to be tolerated if larger diopter IOLs
Figure 5. A, capsular fusion develops first behind the “C” loops and progresses clockwise. Arrows mark the boundary of the fusion at 2 weeks after surgery. B, capsular fusion typically is completed by 2 weeks and occurs symmetrically on either side of the long axis of the plate haptic lenses. Fusion lines are sometimes seen (arrows).
were to be used. Plate and loop haptic and toric IOLs have been developed independently without evaluation of which haptic configuration is more stable. The plate haptic torus is the first to be widely marketed by Staar Surgical Inc, Monrovia, CA. The plate haptic design has been chosen after showing that its insertion does not induce postoperative astigmatism.6 Insertion of loop lenses with folding forceps tends to require wound enlargement and will induce astigmatism, which is undesirable when implanting toric lenses. The development of introducers for loop lenses has been shown to produce postinsertion wound lengths that are compatible with astigmatic neutrality.7 A need has therefore arisen to evaluate which design is more stable and better suited to serve the use of toric optics. The current study has shown that a plate haptic silicone lens has the tendency to be less stable in the early postoperative period than a conventional three-piece lens with polypropylene loop haptics. In the late postoperative period, there is no doubt that the plate lens is more stable. The theoretical basis for IOL rotation is worth discussing before comparing our data with other studies. Translational stability of the IOL has improved significantly after the development of capsulorrhexis. Several clinical and ex vivo studies show that decentration of lenses is less common if there is symmetric in-the-bag fixation.8,9 If rotation is to occur when a lens is entirely sequestered in the bag, there must be resultant torque acting on it. Coincidence of center of the torque with center of the lens generates rotation alone. If this were not the case, then translation would be expected as well. The forces available to generate this torque are gravity and those resulting from ocular movement and events associated with epithelial proliferation. Resisting this torque are the tension developed in the haptics and friction, which may be equatorial or central. Central friction is conceived to be caused by interaction of the capsule with the anterior and posterior surfaces of the optic. Shimizu et al4 and Werblin10 are the only authors to have published the results of loop lens rotation in peer-reviewed literature. Werblin,10 implanting 14-mm, 3-piece, J-loop
2193
Ophthalmology Volume 106, Number 11, November 1999
Figure 6. Axial length versus rotation: plate haptics.
polymethylmethacrylate (PMMA) lenses, observed significant early rotation in 1 of 12 eyes, whereas late rotation greater than 4° was never seen. Werblin identified difficulties in achieving postoperative image registration, using slit-lamp photography and an operating reticule alone, to measure rotation. The difficulties related to torsion induced by factors including local anesthesia, variable head position, and Bell’s phenomenon induced by flash photography on the slit lamp. We addressed this in the current study in the following way. First, we marked the limbal conjunctiva with ink to define orientation before administering local anesthetic. This allowed us to improve registration of surgical and subsequent postoperative images. The grading system used for early rotation was deemed to represent a good compromise between objectivity, accuracy, and repeatability. Second, the CASE 2000 digital retroillumination camera uses dim constant illumination so that Bell’s phenomenon does not occur. A chin rest and an axial fixation light ensure repeatable head and ocular alignment. The greater late stability of the lens observed by Werblin differs from the behavior of the loop lens used in our study. Differing surgical techniques apart, it is possible that the better stability observed by Werblin is related to greater friction arising from the interaction between PMMA and capsule than that found between silicone and capsule.11 In addition, polypropylene haptics used in our study are known to develop less tension than PMMA haptics used in the study by Werblin and will resist rotation less. The AMO SI30NB silicone lens used in this study has now been replaced with the SI40NB lens that has PMMA haptics. Shimizu’s first-generation toric lens was a 13.5-mm, three-piece oval optic with C-shaped loop haptics. Slit-lamp photography was used with the same inherent limitations discussed earlier. Operative photographs were not taken. In 10 (21%) of 47 patients, rotation of 30° or greater was observed by 3 months. Although significant error in orientation can occur without fixation devices, Shimizu’s obser-
2194
vations are likely to be real because there was a correlation between astigmatic correction and degree of measured rotation, although this was not masked. He suggested that instability of the lens is principally related to the three-piece design and that a one-piece lens would rotate less. The explanation might be that one-piece lenses generally are stiffer than three-piece counterparts made from the same biomaterial and will provide more resistance to rotational torques.12 There is no obvious explanation for the differences in the data by Shimizu and Werblin regarding PMMA lenses. I would also have expected lenses in the study by Shimizu to be more stable than the silicone lenses used in our study because of the differences in adhesion referred to earlier. Other potential determinants affecting rotation must then be considered when looking for an explanation. Grabow5 reported a 5% incidence of rotation greater than 30° in a 6-month follow-up of 81 cases implanted with the plate haptic toric lens made by Staar Surgical Inc. (Model 4203T). Assuming that the method of measuring rotation in the study by Grabow was reliable, the lower incidence of instability in his study may be related to the longer length of the toric haptic (10.8 mm vs 10.5 mm), which would result in greater equatorial friction and tension in the haptic. The other major difference between the lens we used and the torus is the presence of large diameter (1.3 mm vs 0.3 mm) fenestrations in the toric lens. There is histologic evidence in animals and humans that transfenestration fusion of capsules does take place, but this did not increase lens stability in an in vitro study by Kent et al.13 This small study did not have the statistical power to answer certain critical questions about the factors influencing IOL rotation. However, certain trends are worth exploring further. Large diameter capsular bags might be associated with a reduction of equatorial friction for a given lens and therefore a decrease in lens stability. Using axial length as a surrogate of capsular bag size,14 we were only able to show a tenuous link at the 12% level for severe rotators in
Patel et al 䡠 Rotation of Plate and Loop Haptic IOLs the plate haptic group to have larger mean axial lengths (Fig 6). That there was no association of axial length for rotation with our loop haptic is not surprising given that the haptic diameter was 13.5 mm instead of 10.5 mm for the plate haptic lens. Postoperative anterior and posterior capsular apposition influences friction between IOL and capsule and could affect lens rotation. One hypothesis that could explain the late stability of the plate lens is how capsular apposition, complete by 2 weeks, traps the lens in the bag. In two of our cases in which early instability was noted despite complete fusion seen at 2 weeks, the lens movement could still have predated fusion. The only severe rotation to occur late for the plate lens had incomplete fusion in keeping with the hypothesis. The only way to disprove the hypothesis is to document severe rotation after capsular fusion is noted to be complete. The fusion pattern was asymmetric for the loop lens and may explain why anticlockwise rotation is the rule, a finding that has also been observed previously.4 Clockwise rotation of the lens would require the fused capsule to separate so that the observed tendency for anticlockwise rotation is in the direction of least resistance into areas of open capsular bag. Haptic compression or expansion may induce a torque on the optic causing it to rotate independently of the haptic.12 Compression will cause clockwise rotation provided a moment is generated about the center of the IOL. Expansion would, in contrast, produce anticlockwise rotation. A clinical study has shown that a reduction in circumference of the capsular bag is universal, implying that compression of the haptic is more likely than expansion.15 The anticlockwise rotation observed in our study is therefore likely to represent rotation of the haptic within the bag rather than that caused by haptic expansion. Anterior capsular contraction in this series of patients has been reported elsewhere (Patel et al. Post-operative changes in the capsulorrhexis aperture: a prospective, randomised comparison between loop and plate haptic silicone intraocular lenses. Eye 1999 [submitted]). Using this as a surrogate of the lens epithelial cell response, it was not possible to show a relationship between rotation and epithelial proliferation for either the plate or loop haptic lens. Initial lens position could affect the way gravity influences early rotation. No relationship was observed, but once again the low numbers in the current study may have been a significant obstacle in terms of the power of the study. The ideal toric lens should not rotate at all, although it is argued that the effects of early instability would be corrected easily by modifiable components of the toric IOL.16 There is evidence that the presence of miniloops, on plate haptics, increases their stability,17 and this modification perhaps should be considered. Greater bioadhesion with capsule associated with acrylic materials may also aid stability.18 Similar considerations apply to loop IOLs for which the major change required is to promote symmetric capsular fusion. How this could be done while maintaining the ease with which the lens is inserted needs to be consid-
ered. Whatever change in haptic design is contemplated, we would recommend that studies of the kind described in this article be used in spherical implants before proceeding with the development of toric lenses to treat combined cataract and astigmatism.
References 1. Ninn–Pedersen K, Stenevi U, Ehinger B. Cataract patients in a defined Swedish population 1986 –1990. II. Preoperative observations. Acta Ophthalmol (Copenh) 1994;72:10 –5. 2. Sparrow JM, Brown NA, Shun–Shin GA, Bron AJ. The Oxford Modular Cataract Image Analysis System. Eye 1990;4: 638 – 48. 3. Stevens JD. Astigmatic excimer laser treatment: theoretical effects of axis misalignment. European Journal of Implant and Refractive Surgery 1994;6:310 – 8. 4. Shimizu K, Misawa A, Suzuki Y. Toric intraocular lenses: correcting astigmatism while controlling axis shift. J Cataract Refract Surg 1994;20:523– 6. 5. Grabow HB. Toric intraocular lens report. Annals of Ophthalmology and Glaucoma 1997;29:161–3. 6. Sanders D, Grabow H, Shepherd J. The toric IOL. In: Gills JP, Martin RG, Sanders DR, eds. Sutureless Cataract Surgery: An Evolution Toward Minimally Invasive Technique. Thorofare, NJ: SLACK, 1992;183–97. 7. Olson R, Cameron R, Hovis T, et al. Clinical evaluation of the Unfolder. J Cataract Refract Surg 1997;23:1384 –9. 8. Wasserman D, Apple DJ, Castaneda VE, et al. Anterior capsular tears and loop fixation of posterior chamber intraocular lenses. Ophthalmology 1991;98:425–31. 9. Pollock WST, Casswell AG. Decentration of the posterior chamber lens implant: a comparison of capsulorhexis with endocapsular surgery. Eye 1994;8:680 –3. 10. Werblin TP. Do three-piece PMMA IOLs rotate after implantation in the capsular bag? J Refract Surg 1995;11:468 –71. 11. Oshika T, Nagata T, Ishii Y. Adhesion of lens capsule to intraocular lenses of polymethylmethacrylate, silicone, and acrylic foldable materials: an experimental study. Br J Ophthalmol 1998;82:549 –53. 12. Parssinen O, Raty J, Klemetti A, et al. Compression forces of haptics of selected posterior chamber lenses. J Cataract Refract Surg 1997;23:1237– 46. 13. Kent DG, Peng Q, Isaacs RT, et al. Security of capsular fixation: small-versus large-hole plate-hepatic lenses. J Cataract Refract Surg 1997;23:1371–5. 14. Lim SJ, Kang SJ, Kim HB, et al. Analysis of zonular-free zone and lens size in relation to axial length of eye with age. J Cataract Refract Surg 1998;24:390 – 6. 15. Strenn K, Menapace R, Vass C. Capsular bag shrinkage after implantation of an open-loop silicone lens and a poly(methyl methacrylate) capsule tension ring. J Cataract Refract Surg 1997;23:1543–7. 16. Werblin TP. Multicomponent intraocular lens. J Refract Surg 1996;12:187–9. 17. Kent DG, Peng Q, Isaacs RT, et al. Mini-haptics to improve capsular fixation of plate-haptic silicone intraocular lenses. J Cataract Refract Surg 1998;24:666 –71. 18. Nagata T, Minakata A, Watanabe I. Adhesiveness of AcrySof to a collagen film. J Cataract Refract Surg 1998;24:367–70.
2195
Ophthalmology Volume 106, Number 11, November 1999 Discussion by David J. Apple, MD I congratulate Dr. Patel and his colleagues from Oxford for their consideration of an important topic, namely the issue of rotation of toric intraocular lens (IOL) implants. This is a very relevant topic since, as they noted, in the 22% of postcataract operation patients in whom astigmatism is manifest, a toric implant can serve to yield emmetropia without recourse to additional corneal surgery provided the lens does not rotate after surgery. The authors’ main finding was that 5 (24%) of the 21 plate haptic IOLs underwent severe early rotation compared to 2 (9%) of 22 loop haptic IOLs (P ⫽ 0.36). They also noted that the mean late rotation was 6.8° for loop haptic lenses compared to 0.6° for plate haptic lenses (P ⫽ 0.0073). The authors concluded that plate haptic implants tend to rotate more so than those with polypropylene loop haptics in the initial postoperative period but are more stable thereafter once capsular fusion has taken place. I agree with these conclusions, but it is useful to note the following:
Address correspondence to David J. Apple, MD, Storm Eye Institute, Department of Ophthalmology, Medical College of South Carolina, 167 Ashley Avenue, Box 250676, Charleston, SC 29425.
The lenses studied in this project included the Allergan SI30 IOL (Allergan, Irvine, CA), a three-piece silicone-optic IOL with polypropylene haptics. The plate lens used in this study was a Chiron-Vision CI0UB design (Chiron-Vision, Claremont, CA), a plate lens with two small holes (0.3-mm diameter). It was 10.5 mm in length. The data and conclusions derived from this are correct and should be documented. However, it must also be recognized that the two lenses studied here are now being supplanted by more modern, analogous designs, namely (1) the Allergan SI40 series, a silicon optic lens with PMMA haptics and (2) the more modern plate lens, which now has two large positioning holes (1.1-mm) diameter. It is also longer, 10.8 mm versus 10.5 mm with the older design. The modern plate lens design should be even more efficacious in reducing rotation according to two mechanisms. The first is that the increased length (10.8 mm) of this design results in a tighter “fit” in the bag. The second is that histologic studies have shown that the large hole allows far better, more stable, and permanent fixation by providing an orifice through which lens epithelial cells and retained cortex can grow, this forming a scar or synechia that serves to lock the lens in a secure position. I congratulate the author for this excellent report. I would welcome a continuation and update of the study by Dr. Patel’s group to test the efficacy of these newer designs.
STATEMENT OF OWNERSHIP, MANAGEMENT AND CIRCULATION (Act of August 12, 1970: Section 3685, Title 39 United States Code) Date of Filing—October 1, 1999. Title of Publication—Ophthalmology; Frequency of Issue—Monthly; Annual Subscription Price—$175.00; Location of Known Office of Publication—Lippincott Williams & Wilkins, 12107 Insurance Way, Hagerstown, MD 21740; Location of the Headquarters or General Business Offices of the Publisher—227 East Washington Square, Philadelphia, PA 19106; Publisher—Lippincott Williams & Wilkins, 227 East Washington Square, Philadelphia, PA 19106; Editor—Donald S. Minckler, MD, Doheny Eye Institute, 1450 San Pablo St, Los Angeles, CA 90033; Managing Editor—Ann Dawson, Doheny Eye Institute, 1450 San Pablo St, DEI-4900, Los Angeles, CA 90033; Owner—American Academy of Ophthalmology, 655 Beach St, San Francisco, CA 94120-9424; Known Bondholders, Mortgagees, and other security holders owning or holding 1 percent or more of total amount of bonds, mortgages, or other securities—None. A. Total no. copies printed (net press run), average 27,571, actual 28,000; B. Paid circulation 1. Sales through dealers and carriers, street vendors and counter sales, average None, actual None; 2. Mail subscriptions, average 26,921, actual 27,325; C. Total paid circulation, average 26,921, actual 27,325; D. Free distribution by mail, carrier or other means. Samples complimentary, and other free copies, average 523, actual 519; E. Total distribution (sum of C and D), average 27,444, actual 27,844; F. Copies not distributed 1. Office use, leftover, unaccounted, spoiled after printing, average 127, actual 156; 2. Returns from news agents, None; G. Total (sum of E & F—should equal net press runs shown in A), average 27,571, actual 28,000. I certify that the statements made by me above are correct and complete. Beverly V. Dietrich, Circulation Director.
2196
Originally received: November 10, 1998. Revision accepted: July 7, 1999. Manuscript no. 98594. From Oxford Eye Hospital, Radcliffe Infirmary, Oxford, England. Presented in part at the American Academy of Ophthalmology annual meeting, New Orleans, Louisiana, November 1998. Reprint requests to C. K. Patel, FRCOphth, c/o Office of Alan Berger, MD, Sunny Brook Health Science Centre, 2075 Bayview Ave., Suite M1-202A, Toronto, Ontario, Canada M4N 3MS. E-mail: [email protected].
2190
controlled study is to compare postoperative rotation of silicone intraocular lenses (IOLs) of spherical power, engineered either as a one-piece plate haptic or a three-piece loop design.
Methods Patients attending for cataract surgery at the Oxford Eye Hospital, a teaching center in the United Kingdom, were prospectively assessed for the study. Exclusion criteria included previous intraocular surgery, evidence of previous uveitis, and presence of pseudoexfoliation, iridodonesis, diabetes, and marked corneal opacity. Patients were included in the study if they gave informed consent. Before surgery, the pupils were dilated with G cyclopentolate 1% and G phenylephrine 10%; G amethocaine 0.5% was used to achieve ocular surface anesthesia. In the operating room, with the patient recumbent, the head was stabilized in a standardized position before a surgeon (CKP) dried and marked 6-o’clock meridian on the limbal conjunctiva. A standard peribulbar anesthetic was then administered. A single surgeon (CKP) performed surgery using a phacotechnique described elsewhere (Patel et al. Post-
Patel et al 䡠 Rotation of Plate and Loop Haptic IOLs
Figure 1. A, straight line connecting the haptic insertions is copied and pasted such that it bisects the corresponding line in the 2-week image. Angle @ is used to grade rotation. “X” is the ink mark defining the 6-o‘clock meridian. B, orientation of the plate haptic lens is defined by selecting one end of its minor axis. The axis is then constructed by drawing right-angled chords across the optic. The 6-o‘clock meridian (X) is here highlighted by placing a lens dialer over the fading ink mark.
operative changes in the capsulorrhexis aperture: a prospective, randomised comparison between loop and plate haptic silicone intraocular lenses. Eye 1999 [accepted, September 1999]). Cases in which the capsulorrhexis was incomplete or in which posterior capsule rupture occurred were excluded before randomization. A permuted block-restricted randomization dictated the IOL that was implanted. The plate haptic lens (C10UB, Bausch & Lomb Surgical, Rochester, NY) was a single-piece plate haptic silicone IOL, 10.5 mm in length, with a 6-mm optic and 0.3-mm diameter fenestrations. The SI30NB lens (Allergan-Medical-Optics, Irvine, CA) was a 13.5-mm long, three-piece lens with a 6-mm silicone optic and “C” shaped polypropylene loops. A video clip of the eye was acquired before a subconjunctival injection of 2 mg betamethasone and 20 mg of gentamicin was administered. After surgery, g dexamethasone 0.1% and g chloramphenicol 0.5% were used four times a day for 2 weeks and then reduced over 2 weeks or as clinically indicated. Patients were assessed at 2 weeks and 6 months after surgery. The pupil was dilated with G tropicamide 1% and G phenylephrine 10%. The relations of the IOL and capsule were documented in detail using biomicroscopy. Particular attention was paid to the manner in which the anterior and posterior capsules fused after surgery. A digital retroillumination image of the red reflex was acquired using a computerized system (CASE 2000, Marcher Enterprises, Hereford, UK) previously developed for cataract analysis.2 It had been modified for IOL assessment by the company in conjunction with one of the authors (CKP). Early rotation of the IOL occurring between the conclusion of surgery and the 2-week postoperative visit was graded by online comparison of the operative video frame and digital retroillumination images. Registration of the two images was necessary when local anesthetic had caused the eye to rotate. This involved digitally rotating the video image so that the ink mark on the limbal conjunctiva was restored to the 6-o’clock meridian. Points on the
IOL to be tracked after surgery differed for the two lenses. For the loop haptic IOL, a line was drawn on the computer linking the haptic insertions (Fig 1A). The line on the video frame was copied and pasted to bisect the line on the 2-week image. The acute angle formed by the lines was assumed to represent the magnitude of rotation and measured using NIH Image 1.47 software (developed by the National Institutes of Health, Bethesda, MD). Plate haptic rotation was measured by tracking its minor axis. One end of the axis was defined by estimating the point on the circumference of the optic at which the tangeant to the optic was parallel to the long axis of the IOL. The minor axis could then be defined by constructing a right-angled triangle as shown in Figure 1B. The magnitude of rotation was graded as mild (⬍10°), moderate (10°⬍ to ⬍30°), or severe (⬍30°). Late rotation was measured more precisely. Perfect registration of images was assumed because a central fixation light on the retroillumination camera ensured repeatable ocular alignment. The CASE 2000 software determined the center of the optic when the latter’s boundary was mapped with a circle. Another cursor marked a peripheral point on the optic. The software calculated the angular subtense of the line connecting these two points with reference to the vertical (Figs 2A, B, and 3A, B). The intraobserver repeatability of determining IOL orientation was tested in 14 patients with each lens design before the study. For early rotation, the mean difference and 95% confidence interval (CI) in test–retest results of orientation for the loop IOL was ⫺0.72° (standard deviation [SD], 2.3; upper 95% CI ⫽ 0.83°; lower 95% CI ⫽ ⫺2.28°) and for the plate haptic was 0.46° (SD, 3.12; upper 95% CI ⫽ 2.55°; lower 95% CI ⫽ ⫺1.62°). Grading this rotation resulted in a 100% match in test–retest data. For late rotation, the mean difference for the loop IOL was 0.35° (SD, 1.29; upper 95% CI ⫽ 1.09°; lower 95% CI ⫽ ⫺0.4°) and for the plate haptic was ⫺0.28° (SD, 2.38; upper 95% CI ⫽ 1.10°; lower 95% CI ⫽ ⫺1.65°).
Figure 2. A, 2-week and 6-month retroillumination images of the Chiron-Vision C10UB plate haptic lens are shown to the left and right, respectively. B, late rotation is defined as @’minus@. The angles are calculated by the CASE 2000 software after the center of the optic and a peripheral landmark are defined.
2191
Ophthalmology Volume 106, Number 11, November 1999 was excluded because of inadequate mydriasis and an additional two in each group for failure to attend the 6-month visit. One patient in the loop haptic group developed asymmetric fixation of haptics and another died before the 6-month visit. Another exclusion in the plate haptic group resulted from the absence of the 2-week image.
Early Rotation For the loop haptic IOL, 13 (59%) of 22, 7 (32%) of 22, and 2 (9%) of 22 underwent mild, moderate, and severe rotation, respectively, contrasting with 9 (43%) of 21, 7 (33%) of 21, and 5 (24%) of 21 for the plate haptic IOL (P ⫽ 0.36). In 10 (45%) of 22 and 10 (48%) of 21 cases, the rotation was clockwise for the loop and plate haptic IOLs, respectively (P ⫽ 0.86).
Late Rotation
Figure 3. A, 2-week and 6-month retroillumination images of the AMO SI30NB loop haptic lens are shown to the left and right, respectively. B, late rotation is defined as @’minus@. The angles are calculated by the CASE 2000 software after the center of the optic and a peripheral landmark are defined.
Statistical analysis was performed on a Macintosh computer (Apple, Cupertino, CA) using JMP software (SAS, Cary, NC). Categoric data were tested using the chi-square test. Nonparametric analysis was used to compare the median late rotation and one-way analysis of variance to compare the relationship of axial length to early rotation. A P value of 0.05 was considered significant.
Results Fifty-four patients entered the study. Five patients had an incomplete capsulorrhexis and 1 patient had a posterior capsule rupture, leaving 48 eyes for randomization.
Demographic Data Twenty-three patients received the loop haptic lens and 25 the plate haptic design. There was no statistically significant difference between the groups for mean age, laterality, or gender distribution (Patel et al. Post-operative changes in the capsulorrhexis aperture: a prospective, randomised comparison between loop and plate haptic silicone intraocular lenses. Eye 1999 [accepted, September, 1999]). The number of patients excluded from analysis for early and late rotation did not differ significantly between groups (P ⬎ 0.05). Four patients with plate haptic IOLs were excluded when early rotation was analyzed: one patient in whom the pupil dilated poorly at 2 weeks, another patient whose 2-week image had not been taken because of equipment failure, and two patients in whom the video images were of poor quality. One patient, for whom a video image had not been captured, was the only exclusion from the loop haptic group. For late rotation, one patient in each group
2192
The distribution of the late rotation is shown in Figures 4A and B. A mean rotation of 7.61° (7.25° SD) occurred for the loop haptic IOL contrasting with a mean rotation of 1.8 (12.9° SD) for the plate haptic. The median rotation was 6.8° and 0.6° for loop and plate haptics, respectively (P ⫽ 0.0073). Sixteen (89%) of 18 loop haptic IOLs underwent anticlockwise rotation compared to 11 (52%) of 21 plate haptic IOLs (P ⫽ 0.0081). The modal pattern of capsular fusion observed at 2 weeks for plate (11 [46%] of 24) and loop haptics (16 [73%] of 22) is shown in Figures 5A and B. No comments were made regarding fusion for one patient in each group. The fusion, for the plate haptic, was symmetric and complete to the edge of the lens by 2 weeks. For the loop lens, the fusion was asymmetric beginning behind the convexity of the loop and proceeding clockwise. The pupil was dilated inadequately to allow assessment of fusion in 2 (8%) of 24 and 3 (14%) of 22 patients with plate and loop haptics, respectively. In 6 (25%) of 24 patients with plate haptics, the characteristic fusion pattern was only observed on one side of the lens, the other side being obscured by the pupil. For 5 (21%) of 24 plate haptics, capsular fusion was not complete. Two of these cases showed severe early rotation. It is noteworthy that the only plate haptic IOL to show severe movement between 2 weeks and 6 months also had incomplete fusion. For the loop haptic, fusion was complete for 360° in two patients, and postoperative fibrin accumulation prevented adequate assessment in another 3 (13%) of 22 patients. No significant relationship was detected between rotation and axial length, initial alignment of the lens, or capsular contraction, which has been the subject of a previous report (Patel et al. Post-operative changes in the capsulorrhexis aperture: a prospective, randomised comparison between loop and plate haptic silicone intraocular lenses. Eye 1999 [submitted]) (P ⬎ 0.05). If rotation in excess of 30° was considered for the plate haptic lens over the 6-month follow-up, a tenuous link existed between axial length and rotation (P ⫽ 0.12) (Fig 6).
Discussion Astigmatism occurs in approximately 22% of patients with cataract.1 Achieving emmetropia for these cases has become possible because of the evolution of astigmatically neutral cataract incisions. One option is to combine corneal refractive surgery with the use of a spherical implant. The drawbacks of this approach are unpredictability and the risk of
Patel et al 䡠 Rotation of Plate and Loop Haptic IOLs
Figure 4. A, late rotation for Chiron-Vision C10UB. B, late rotation for AMO SI30NB. Rotation in degrees is plotted on the vertical. A positive number denotes anticlockwise rotation. Zero represents the position of a patient’s lens at 2 weeks, and the top of the bar represents the magnitude and direction of rotation.
violating corneal anatomy. A toric IOL would overcome these drawbacks provided it did not rotate after surgery. The magnitude of cylinder corrected is inversely proportional to the degree of axis misalignment between 0° and 30°.3 Rotation in excess of 30° accentuates preoperative astigmatism and is unacceptable for toric IOLs. Rotation less than 30°, although not providing full correction of the cylinder, has been well-tolerated in clinical studies involving implantation of 2- and 3-diopter toric IOLs.4,5 Such rotation is, however, less likely to be tolerated if larger diopter IOLs
Figure 5. A, capsular fusion develops first behind the “C” loops and progresses clockwise. Arrows mark the boundary of the fusion at 2 weeks after surgery. B, capsular fusion typically is completed by 2 weeks and occurs symmetrically on either side of the long axis of the plate haptic lenses. Fusion lines are sometimes seen (arrows).
were to be used. Plate and loop haptic and toric IOLs have been developed independently without evaluation of which haptic configuration is more stable. The plate haptic torus is the first to be widely marketed by Staar Surgical Inc, Monrovia, CA. The plate haptic design has been chosen after showing that its insertion does not induce postoperative astigmatism.6 Insertion of loop lenses with folding forceps tends to require wound enlargement and will induce astigmatism, which is undesirable when implanting toric lenses. The development of introducers for loop lenses has been shown to produce postinsertion wound lengths that are compatible with astigmatic neutrality.7 A need has therefore arisen to evaluate which design is more stable and better suited to serve the use of toric optics. The current study has shown that a plate haptic silicone lens has the tendency to be less stable in the early postoperative period than a conventional three-piece lens with polypropylene loop haptics. In the late postoperative period, there is no doubt that the plate lens is more stable. The theoretical basis for IOL rotation is worth discussing before comparing our data with other studies. Translational stability of the IOL has improved significantly after the development of capsulorrhexis. Several clinical and ex vivo studies show that decentration of lenses is less common if there is symmetric in-the-bag fixation.8,9 If rotation is to occur when a lens is entirely sequestered in the bag, there must be resultant torque acting on it. Coincidence of center of the torque with center of the lens generates rotation alone. If this were not the case, then translation would be expected as well. The forces available to generate this torque are gravity and those resulting from ocular movement and events associated with epithelial proliferation. Resisting this torque are the tension developed in the haptics and friction, which may be equatorial or central. Central friction is conceived to be caused by interaction of the capsule with the anterior and posterior surfaces of the optic. Shimizu et al4 and Werblin10 are the only authors to have published the results of loop lens rotation in peer-reviewed literature. Werblin,10 implanting 14-mm, 3-piece, J-loop
2193
Ophthalmology Volume 106, Number 11, November 1999
Figure 6. Axial length versus rotation: plate haptics.
polymethylmethacrylate (PMMA) lenses, observed significant early rotation in 1 of 12 eyes, whereas late rotation greater than 4° was never seen. Werblin identified difficulties in achieving postoperative image registration, using slit-lamp photography and an operating reticule alone, to measure rotation. The difficulties related to torsion induced by factors including local anesthesia, variable head position, and Bell’s phenomenon induced by flash photography on the slit lamp. We addressed this in the current study in the following way. First, we marked the limbal conjunctiva with ink to define orientation before administering local anesthetic. This allowed us to improve registration of surgical and subsequent postoperative images. The grading system used for early rotation was deemed to represent a good compromise between objectivity, accuracy, and repeatability. Second, the CASE 2000 digital retroillumination camera uses dim constant illumination so that Bell’s phenomenon does not occur. A chin rest and an axial fixation light ensure repeatable head and ocular alignment. The greater late stability of the lens observed by Werblin differs from the behavior of the loop lens used in our study. Differing surgical techniques apart, it is possible that the better stability observed by Werblin is related to greater friction arising from the interaction between PMMA and capsule than that found between silicone and capsule.11 In addition, polypropylene haptics used in our study are known to develop less tension than PMMA haptics used in the study by Werblin and will resist rotation less. The AMO SI30NB silicone lens used in this study has now been replaced with the SI40NB lens that has PMMA haptics. Shimizu’s first-generation toric lens was a 13.5-mm, three-piece oval optic with C-shaped loop haptics. Slit-lamp photography was used with the same inherent limitations discussed earlier. Operative photographs were not taken. In 10 (21%) of 47 patients, rotation of 30° or greater was observed by 3 months. Although significant error in orientation can occur without fixation devices, Shimizu’s obser-
2194
vations are likely to be real because there was a correlation between astigmatic correction and degree of measured rotation, although this was not masked. He suggested that instability of the lens is principally related to the three-piece design and that a one-piece lens would rotate less. The explanation might be that one-piece lenses generally are stiffer than three-piece counterparts made from the same biomaterial and will provide more resistance to rotational torques.12 There is no obvious explanation for the differences in the data by Shimizu and Werblin regarding PMMA lenses. I would also have expected lenses in the study by Shimizu to be more stable than the silicone lenses used in our study because of the differences in adhesion referred to earlier. Other potential determinants affecting rotation must then be considered when looking for an explanation. Grabow5 reported a 5% incidence of rotation greater than 30° in a 6-month follow-up of 81 cases implanted with the plate haptic toric lens made by Staar Surgical Inc. (Model 4203T). Assuming that the method of measuring rotation in the study by Grabow was reliable, the lower incidence of instability in his study may be related to the longer length of the toric haptic (10.8 mm vs 10.5 mm), which would result in greater equatorial friction and tension in the haptic. The other major difference between the lens we used and the torus is the presence of large diameter (1.3 mm vs 0.3 mm) fenestrations in the toric lens. There is histologic evidence in animals and humans that transfenestration fusion of capsules does take place, but this did not increase lens stability in an in vitro study by Kent et al.13 This small study did not have the statistical power to answer certain critical questions about the factors influencing IOL rotation. However, certain trends are worth exploring further. Large diameter capsular bags might be associated with a reduction of equatorial friction for a given lens and therefore a decrease in lens stability. Using axial length as a surrogate of capsular bag size,14 we were only able to show a tenuous link at the 12% level for severe rotators in
Patel et al 䡠 Rotation of Plate and Loop Haptic IOLs the plate haptic group to have larger mean axial lengths (Fig 6). That there was no association of axial length for rotation with our loop haptic is not surprising given that the haptic diameter was 13.5 mm instead of 10.5 mm for the plate haptic lens. Postoperative anterior and posterior capsular apposition influences friction between IOL and capsule and could affect lens rotation. One hypothesis that could explain the late stability of the plate lens is how capsular apposition, complete by 2 weeks, traps the lens in the bag. In two of our cases in which early instability was noted despite complete fusion seen at 2 weeks, the lens movement could still have predated fusion. The only severe rotation to occur late for the plate lens had incomplete fusion in keeping with the hypothesis. The only way to disprove the hypothesis is to document severe rotation after capsular fusion is noted to be complete. The fusion pattern was asymmetric for the loop lens and may explain why anticlockwise rotation is the rule, a finding that has also been observed previously.4 Clockwise rotation of the lens would require the fused capsule to separate so that the observed tendency for anticlockwise rotation is in the direction of least resistance into areas of open capsular bag. Haptic compression or expansion may induce a torque on the optic causing it to rotate independently of the haptic.12 Compression will cause clockwise rotation provided a moment is generated about the center of the IOL. Expansion would, in contrast, produce anticlockwise rotation. A clinical study has shown that a reduction in circumference of the capsular bag is universal, implying that compression of the haptic is more likely than expansion.15 The anticlockwise rotation observed in our study is therefore likely to represent rotation of the haptic within the bag rather than that caused by haptic expansion. Anterior capsular contraction in this series of patients has been reported elsewhere (Patel et al. Post-operative changes in the capsulorrhexis aperture: a prospective, randomised comparison between loop and plate haptic silicone intraocular lenses. Eye 1999 [submitted]). Using this as a surrogate of the lens epithelial cell response, it was not possible to show a relationship between rotation and epithelial proliferation for either the plate or loop haptic lens. Initial lens position could affect the way gravity influences early rotation. No relationship was observed, but once again the low numbers in the current study may have been a significant obstacle in terms of the power of the study. The ideal toric lens should not rotate at all, although it is argued that the effects of early instability would be corrected easily by modifiable components of the toric IOL.16 There is evidence that the presence of miniloops, on plate haptics, increases their stability,17 and this modification perhaps should be considered. Greater bioadhesion with capsule associated with acrylic materials may also aid stability.18 Similar considerations apply to loop IOLs for which the major change required is to promote symmetric capsular fusion. How this could be done while maintaining the ease with which the lens is inserted needs to be consid-
ered. Whatever change in haptic design is contemplated, we would recommend that studies of the kind described in this article be used in spherical implants before proceeding with the development of toric lenses to treat combined cataract and astigmatism.
References 1. Ninn–Pedersen K, Stenevi U, Ehinger B. Cataract patients in a defined Swedish population 1986 –1990. II. Preoperative observations. Acta Ophthalmol (Copenh) 1994;72:10 –5. 2. Sparrow JM, Brown NA, Shun–Shin GA, Bron AJ. The Oxford Modular Cataract Image Analysis System. Eye 1990;4: 638 – 48. 3. Stevens JD. Astigmatic excimer laser treatment: theoretical effects of axis misalignment. European Journal of Implant and Refractive Surgery 1994;6:310 – 8. 4. Shimizu K, Misawa A, Suzuki Y. Toric intraocular lenses: correcting astigmatism while controlling axis shift. J Cataract Refract Surg 1994;20:523– 6. 5. Grabow HB. Toric intraocular lens report. Annals of Ophthalmology and Glaucoma 1997;29:161–3. 6. Sanders D, Grabow H, Shepherd J. The toric IOL. In: Gills JP, Martin RG, Sanders DR, eds. Sutureless Cataract Surgery: An Evolution Toward Minimally Invasive Technique. Thorofare, NJ: SLACK, 1992;183–97. 7. Olson R, Cameron R, Hovis T, et al. Clinical evaluation of the Unfolder. J Cataract Refract Surg 1997;23:1384 –9. 8. Wasserman D, Apple DJ, Castaneda VE, et al. Anterior capsular tears and loop fixation of posterior chamber intraocular lenses. Ophthalmology 1991;98:425–31. 9. Pollock WST, Casswell AG. Decentration of the posterior chamber lens implant: a comparison of capsulorhexis with endocapsular surgery. Eye 1994;8:680 –3. 10. Werblin TP. Do three-piece PMMA IOLs rotate after implantation in the capsular bag? J Refract Surg 1995;11:468 –71. 11. Oshika T, Nagata T, Ishii Y. Adhesion of lens capsule to intraocular lenses of polymethylmethacrylate, silicone, and acrylic foldable materials: an experimental study. Br J Ophthalmol 1998;82:549 –53. 12. Parssinen O, Raty J, Klemetti A, et al. Compression forces of haptics of selected posterior chamber lenses. J Cataract Refract Surg 1997;23:1237– 46. 13. Kent DG, Peng Q, Isaacs RT, et al. Security of capsular fixation: small-versus large-hole plate-hepatic lenses. J Cataract Refract Surg 1997;23:1371–5. 14. Lim SJ, Kang SJ, Kim HB, et al. Analysis of zonular-free zone and lens size in relation to axial length of eye with age. J Cataract Refract Surg 1998;24:390 – 6. 15. Strenn K, Menapace R, Vass C. Capsular bag shrinkage after implantation of an open-loop silicone lens and a poly(methyl methacrylate) capsule tension ring. J Cataract Refract Surg 1997;23:1543–7. 16. Werblin TP. Multicomponent intraocular lens. J Refract Surg 1996;12:187–9. 17. Kent DG, Peng Q, Isaacs RT, et al. Mini-haptics to improve capsular fixation of plate-haptic silicone intraocular lenses. J Cataract Refract Surg 1998;24:666 –71. 18. Nagata T, Minakata A, Watanabe I. Adhesiveness of AcrySof to a collagen film. J Cataract Refract Surg 1998;24:367–70.
2195
Ophthalmology Volume 106, Number 11, November 1999 Discussion by David J. Apple, MD I congratulate Dr. Patel and his colleagues from Oxford for their consideration of an important topic, namely the issue of rotation of toric intraocular lens (IOL) implants. This is a very relevant topic since, as they noted, in the 22% of postcataract operation patients in whom astigmatism is manifest, a toric implant can serve to yield emmetropia without recourse to additional corneal surgery provided the lens does not rotate after surgery. The authors’ main finding was that 5 (24%) of the 21 plate haptic IOLs underwent severe early rotation compared to 2 (9%) of 22 loop haptic IOLs (P ⫽ 0.36). They also noted that the mean late rotation was 6.8° for loop haptic lenses compared to 0.6° for plate haptic lenses (P ⫽ 0.0073). The authors concluded that plate haptic implants tend to rotate more so than those with polypropylene loop haptics in the initial postoperative period but are more stable thereafter once capsular fusion has taken place. I agree with these conclusions, but it is useful to note the following:
Address correspondence to David J. Apple, MD, Storm Eye Institute, Department of Ophthalmology, Medical College of South Carolina, 167 Ashley Avenue, Box 250676, Charleston, SC 29425.
The lenses studied in this project included the Allergan SI30 IOL (Allergan, Irvine, CA), a three-piece silicone-optic IOL with polypropylene haptics. The plate lens used in this study was a Chiron-Vision CI0UB design (Chiron-Vision, Claremont, CA), a plate lens with two small holes (0.3-mm diameter). It was 10.5 mm in length. The data and conclusions derived from this are correct and should be documented. However, it must also be recognized that the two lenses studied here are now being supplanted by more modern, analogous designs, namely (1) the Allergan SI40 series, a silicon optic lens with PMMA haptics and (2) the more modern plate lens, which now has two large positioning holes (1.1-mm) diameter. It is also longer, 10.8 mm versus 10.5 mm with the older design. The modern plate lens design should be even more efficacious in reducing rotation according to two mechanisms. The first is that the increased length (10.8 mm) of this design results in a tighter “fit” in the bag. The second is that histologic studies have shown that the large hole allows far better, more stable, and permanent fixation by providing an orifice through which lens epithelial cells and retained cortex can grow, this forming a scar or synechia that serves to lock the lens in a secure position. I congratulate the author for this excellent report. I would welcome a continuation and update of the study by Dr. Patel’s group to test the efficacy of these newer designs.
STATEMENT OF OWNERSHIP, MANAGEMENT AND CIRCULATION (Act of August 12, 1970: Section 3685, Title 39 United States Code) Date of Filing—October 1, 1999. Title of Publication—Ophthalmology; Frequency of Issue—Monthly; Annual Subscription Price—$175.00; Location of Known Office of Publication—Lippincott Williams & Wilkins, 12107 Insurance Way, Hagerstown, MD 21740; Location of the Headquarters or General Business Offices of the Publisher—227 East Washington Square, Philadelphia, PA 19106; Publisher—Lippincott Williams & Wilkins, 227 East Washington Square, Philadelphia, PA 19106; Editor—Donald S. Minckler, MD, Doheny Eye Institute, 1450 San Pablo St, Los Angeles, CA 90033; Managing Editor—Ann Dawson, Doheny Eye Institute, 1450 San Pablo St, DEI-4900, Los Angeles, CA 90033; Owner—American Academy of Ophthalmology, 655 Beach St, San Francisco, CA 94120-9424; Known Bondholders, Mortgagees, and other security holders owning or holding 1 percent or more of total amount of bonds, mortgages, or other securities—None. A. Total no. copies printed (net press run), average 27,571, actual 28,000; B. Paid circulation 1. Sales through dealers and carriers, street vendors and counter sales, average None, actual None; 2. Mail subscriptions, average 26,921, actual 27,325; C. Total paid circulation, average 26,921, actual 27,325; D. Free distribution by mail, carrier or other means. Samples complimentary, and other free copies, average 523, actual 519; E. Total distribution (sum of C and D), average 27,444, actual 27,844; F. Copies not distributed 1. Office use, leftover, unaccounted, spoiled after printing, average 127, actual 156; 2. Returns from news agents, None; G. Total (sum of E & F—should equal net press runs shown in A), average 27,571, actual 28,000. I certify that the statements made by me above are correct and complete. Beverly V. Dietrich, Circulation Director.
2196