- Email: [email protected]
Refractive lens exchange in keratoconus
J CATARACT REFRACT SURG - VOL 32, MAY 2006
Refractive lens exchange in keratoconus Antonio Leccisotti, MD, PhD
PURPOSE: To evaluate the visual results and complications of refractive lens exchange to correct myopia associated with early-stage keratoconus. SETTING: Private practice, Siena, Italy. METHODS: This prospective noncomparative interventional series comprised 34 consecutive eyes of 20 patients with stages I to II keratoconus. Mean patient age was 56.7 years G 10.4 (SD). Preoperative mean spherical equivalent (SE) was –11.0 G 4.65 diopters (D), (range –5.75 to –22). Ultrasound biometry was performed using videokeratographic central K-readings and the Holladay 2 formula. An intraoperative handheld autorefractor was used to check the power of implanted intraocular lenses. RESULTS: Intraocular lens exchange due to inaccurate power occurred in 11 eyes (32%; 9 eyes intraoperative, 2 eyes postoperative). At 12 months, mean SE was –1.31 G1.08 D and mean defocus equivalent was 1.94 G 1.57 D. Twenty-two eyes (65%) were within G2 D of defocus equivalent, 16 eyes (47%) were within G1 D, and 3 eyes (9%) were within G0.5. Mean surgically induced astigmatism (vector analysis) was 0.54 G 0.43 D. Preoperative mean best spectacle-corrected visual acuity (BSCVA) was 0.55 G 0.20, and postoperative mean BSCVA was 0.76 G 0.23; the difference was statistically significant (P<.05; 95% confidence interval, 0.19 to 0.25). Postoperative mean uncorrected visual acuity was 0.48 G 0.25. The safety index was 1.38, and the efficacy index was 0.87. Complications were posterior vitreous detachment (9%) and dysphotopsia phenomena (15%). Corneal endothelial cell density at 12 months decreased by 6.3%. CONCLUSION: Refractive lens exchange in keratoconic eyes predictably corrected myopia. However, ultrasound biometry was inaccurate in almost one third of eyes. Intraoperative autorefractometry is recommended to improve refractive outcome. J Cataract Refract Surg 2006; 32:742–746 Q 2006 ASCRS and ESCRS
Keratoconus can be associated with axial myopia or can induce myopia by increased corneal curvature.1 Laser in situ keratomileusis in keratoconus may cause corneal ectasia,2 and surgical correction of the spherical error associated with keratoconus remains a challenge. We previously evaluated phakic intraocular lenses (IOLs) for this purpose,3 but these lenses have several anatomical limitations and are generally not implanted in patients older than 45 years. Other techniques, such as intracorneal rings, can improve visual acuity and reduce corneal curvature in keratoconus,4,5 but correction of the
Accepted for publication August 14, 2005. From the Ophthalmic Surgery Unit, Casa di Cura Rugani, Siena, Italy. The author has no financial or proprietary interest in any material or method mentioned. Reprint requests to Antonio Leccisotti, MD, PhD, Piazza 5 Bersaglieri, 2 53100 Siena, Italy. E-mail: [email protected]. Q 2006 ASCRS and ESCRS Published by Elsevier Inc.
742
spherical error is limited.6 Photorefractive keratectomy has been used by some researchers to treat myopia associated with keratoconus,7 but thinning of keratoconic corneas raises long-term safety concerns.8 Refractive lens exchange (RLE), also called clear lens extraction, consists of phacoemulsification of the clear crystalline lens and in-the-bag implantation of an appropriately powered IOL. It is generally used to correct large spherical errors in patients in the presbyopic age range because it causes loss of accommodation.9 Myopia associated with keratoconus is not considered an indication for RLE because of difficult IOL power calculation.10 In a Medline search of RLE in keratoconus, only a single case report was found,11 together with the suggestion (based on 2 cases) to use videokeratographic K-readings in biometry in keratoconus12 (October 11, 2005; keywords: keratoconus, intraocular lens, and lens exchange). We therefore evaluated the efficacy and safety of RLE for the correction of myopic spherical error associated with early-stage keratoconus. 0886-3350/06/$-see front matter doi:10.1016/j.jcrs.2006.01.063
RLE IN KERATOCONUS
PATIENTS AND METHODS Patients with keratoconus having RLE to correct myopia by 1 surgeon (A.L.) between September 2002 and July 2004 were evaluated. Inclusion criteria were age greater than 45 years, keratoconus stages I to II in each operated eye, subjective contact lens intolerance (discomfort), myopic refraction, measurable preoperative subjective refraction, best spectacle-corrected visual acuity (BSCVA) 0.2 or better, refractive stability for at least 3 years, and minimum follow-up of 12 months. Exclusion criteria were corneal scars, previous corneal surgery, untreated rhegmatogenous retinal degeneration, monocularity, endothelial cell density less than 1800 cells/mm2 (evaluated by specular microscopy), glaucoma or ocular hypertension, diabetic retinopathy, ocular inflammatory diseases, and crystalline lens opacities affecting preoperative BSCVA. Informed consent was obtained from all patients. This stated that only the spherical component would be corrected, that further enhancement by excimer laser was not possible, and that spectacles would be needed after the operation to correct residual errors. The study was approved by the ethics committee of our institute. Keratoconus was diagnosed by combination of computed videokeratography, K-readings, and corneal ultrasound (US) pachymetry. Pachymetry was used to reveal differences between superior and inferior thickness (I–S difference) of more than 100 mm. The keratoconus stage was assessed by the system included in the computed videokeratographer keratograph (Oculus), as in a previous study.3 The stages were as follows: Stage I: Minimum corneal curvature (MCC), 7.5 to 6.5 mm; index of surface variance (ISV Z deviation of corneal radium from average), 30 to 55; presumed BSCVA, 20/25 to 20/15. Stage II: MCC: 6.9 to 5.3 mm; ISV, 55 to 90; BSCVA, 20/60 to 20/20. Stage III: MCC, 6.6 to 4.8 mm; ISV, 90 to 150; BSCVA, 20/125 to 20/30. Stage IV: MCC, !5 mm; ISV, O150; BSCVA, worse than 20/100. Contact lens use had been discontinued for 2 months before corneal topography. Manifest and cycloplegic refraction, slitlamp microscopy of undilated and dilated anterior segment, fundus with Goldmann mirror, applanation tonometry, and dilated binocular indirect ophthalmoscopy were also performed before surgery. Intraocular lens power was calculated by US biometry using the Allergan Humphrey 820 and Holladay 2 formula, targeting a residual myopic astigmatism. Keratometry readings were obtained by axial topographic maps. The dioptric power of the steepest meridian over the central 3 mm of the cornea was considered K1, and the dioptric power of the flattest meridian over the central 3 mm of the cornea was considered K2. Each meridian power was calculated by averaging its 2 semimeridian values, always considering the central 3 mm. Sixty and 30 minutes before surgery, ciprofloxacin (Oftacilox) and tropicamide 1% eyedrops were instilled. Surgery was performed on an outpatient basis with topical anesthesia (unpreserved lidocaine 1%). Five minutes before surgery, povidone– iodine (Betadine) diluted at 5% was instilled in the conjunctival sac. A sclerocorneal 3.2 mm 3.0 mm tunnel was made, the anterior chamber was filled with an ophthalmic viscosurgical device (OVD) (sodium hyaluronate 1.0% [Provisc]), continuous curvilinear capsulorhexis was performed by forceps, and, after hydrodissection, the lens nucleus was emulsified and the cortex aspirated.
A single-piece foldable acrylic IOL (Stabibag [Rockmerd BV]) with a 5.5 mm square-edged optic was inserted in the capsular bag by an injector, OVD was aspirated, and a single 10-0 nylon radial suture was used to close the wound. With the patient still on the operating table and the eyelid speculum on, autorefractometry was performed with a handheld instrument (Nikon Retinomax 2). If a spherical error greater than 1.5 diopters (D) was found, the IOL power was immediately exchanged. Approximately C1.5 D was added to the original IOL power for each diopter of residual hyperopia and ÿ1 D was added for each diopter of residual myopia. Exchange was performed by filling the anterior chamber with OVD, lifting the IOL in the anterior chamber by a hook, cutting the IOL in 2 with Vannas scissors, removing the IOL parts, injecting the new lens into the capsular bag, and aspirating the OVD again. Ciprofloxacin and dexamethasone 0.1% eyedrops (Visumetazone) were used 4 times a day for 15 days. Follow-up visits were at 1 and 7 days and 1, 3, 6, 12, 18, and 24 months. Evaluation included uncorrected visual acuity (UCVA), BSCVA, corneal topography, endothelial cell count, slitlamp evaluation, and fundoscopy. Statistical analysis was performed by StatView (Abacus Concepts, Inc.), and P values less than 0.05 were considered statistically significant; 95% confidence intervals (CI) were calculated by CIA Software (BMJ). Predictability was measured as attempted versus achieved spherical equivalent (SE) and as the percentage of eyes within G0.5 D, G1.0 D, and G2.0 D of defocus equivalent. Surgically induced astigmatism was evaluated by vector analysis.13 Safety was evaluated by assessing the percentage of eyes losing lines of BSCVA and by safety index (postoperative BSCVA/ preoperative BSCVA). Efficacy was evaluated by assessing postoperative UCVA and by efficacy index (postoperative UCVA/preoperative BSCVA). Visual acuity was measured and all calculations performed in logMAR units. Data were converted into decimals for presentation. RESULTS
Thirty-four eyes of 20 patients were included in the study. Mean patient age was 56.7 years G10.4 (SD) (range 45 to 76 years). Eight of 20 patients (40%) were male. Twenty eyes had keratoconus stage I, and 14 eyes had stage II. Mean follow-up was 17.4 G 5.1 months (range 12 to 24 months). Surgery was uneventful in all cases. Intraoperative in-the-bag IOL exchange after autorefractometry was performed in 9 eyes (26%) of 6 patients (4 eyes with keratoconus stage I and 5 eyes with stage II). In 2 eyes (6%) of 2 other patients (both stage II), a second operation was required 2 weeks after the first to exchange the IOL because a ÿ5 D and a ÿ6 D postoperative SE had resulted. Intraoperative autorefractometry had not revealed these large errors. Preoperative mean SE was ÿ11.0 G 4.65 D (range ÿ5.75 to ÿ22) and ÿ1.31 G 1.08 D (range, ÿ0.25 to ÿ4.5) at 12 months. The difference was statistically significant (P!.05 with Student t test and 95% CI, ÿ8.26 to ÿ11.18 D) (Figure 1). Preoperative mean defocus equivalent was 12.0 G 4.64 D (range 6.50 to 23 D) and 1.94 G 1.57 D (range 0.25 to 5.5 D) at 12 months. The difference was statistically
J CATARACT REFRACT SURG - VOL 32, MAY 2006
743
RLE IN KERATOCONUS
40 35
% of eyes
30 25 20 15 10 5 0 <-1
-1
0
1
2
3
>3
BSCVA Snellen lines gained/lost
Figure 2. Changes in BSCVA 12 months after myopic refractive lens exchange in 34 keratoconic eyes.
significant (P!.05 with Student t test; 95% CI, 8.63 to 11.46 D). Twenty-two of 34 eyes (65%) were within G2.0 D of defocus equivalent, 16 of 34 (47%) were within G1.0 D, and 3 of 34 eyes (9%) were within G0.5 D. Preoperative mean refractive cylinder was 1.86 G 1.39 D (range 0.5 to 5) and at 12 months, 1.22 G 1.37 D (range 0 to 4). The difference was not statistically significant with Student t test but was significant with 95% CI (0.4 to 0.9 D). Mean surgically induced astigmatism, calculated by vector analysis, was 0.54 G 0.43 D (range 0 to 1.80 D). Preoperative mean BSCVA was 0.55 G 0.23 (range 0.2 to 0.9) G 0.20 and 0.76 G 0.23 (range 0.4 to 1.0) at 12 months. The difference was statistically significant (P!.05 with Student t test; 95% CI, 0.19 to 0.25) (Figure 2). Mean UCVA at 12 months was 0.48 G 0.25 (range 0.1 to 0.9) (Figure 3). The safety index was 1.38, and the efficacy index was 0.87. Stability was evaluated by paired comparison of mean defocus equivalent at 3 months (1.78 D) and at 12 months (1.95 D). The variation was not statistically significant (95% CI, ÿ0.51 to 0.20 D). Posterior vitreous detachment occurred in 3 patients (9%) after 1, 6, and 12 months, respectively. Posterior capsule opacification occurred in 6 eyes (17.6%) of 4 patients. Neodymium:YAG laser capsulotomy was performed in all cases with full functional recovery. Dysphotopsia phenomena were reported by 3 of 20 patients (15%), consisting especially of light reflections
744
DISCUSSION
Although the final refractive error was acceptable (mean SE ÿ1.31 D), this result required a high number (almost 1 of every 3 eyes) of intraoperative (26%) or postoperative (6%) IOL exchanges due to inaccurate
40 35 30
% of eyes
Figure 1. Scattergram of attempted versus achieved correction (SE) 12 months after myopic refractive lens exchange in 34 keratoconic eyes with 1-diopter interval bars.
caused by lateral light sources in mesopic conditions; however, night activities were not impaired, as patients favorably compared dysphotopsia with halos generated by rigid gas-permeable contact lenses. Postoperative corneal endothelial cell density at 12 months was available in 21 of 34 eyes. In these eyes, mean preoperative density was 2652 G 421 cells/mm2 and mean postoperative density was 2487 G 327 cells/mm2, with a 6.3% reduction. However, this reduction was not statistically significant (95% CI, ÿ70 to 401 cells/mm2).
25 20
Pre-op BSCVA
15
Post-op UCVA
10 5 0 <0,1
0,2
0,4
0,6
0,8
1
Visual Acuity
Figure 3. Efficacy of myopic RLE in 34 keratoconic eyes 12 months after surgery. Black columns, preoperative BSCVA and white columns, postoperative UCVA.
J CATARACT REFRACT SURG - VOL 32, MAY 2006
RLE IN KERATOCONUS
calculation of the IOL power, more commonly in eyes with keratoconus stage II. Intraoperative autorefractometry improved predictability, although in 2 eyes it was totally inaccurate. This can be explained by the following considerations: (1) The IOL position at the end of surgery was not definitive. (2) Patient fixation into the autorefractor was affected by prolonged retinal illumination during surgery. (3) Autorefractometry can be imprecise with irregular corneas. The use of optical measurement of axial length as well as advanced IOL formulas should improve IOL calculation and reduce the percentage of IOL exchanges. However, in a large series of myopic RLEs,4 final mean SE was ÿ1.22 D, very near our result in keratoconic eyes (ÿ1.31 D). In our series, final mean defocus equivalent was 1.94 D, which can be at least partially explained by a preoperative mean astigmatism of 1.86 D. As observed in our previous study,3 the effect of tunnel incision on keratoconic corneas is similar to that on normal corneas14; therefore, the use of toric IOL is justified in RLE in keratoconus.11 The main bias affecting our studies is the limited and heterogeneous (in terms of age and SE) sample. This explains the low incidence of vitreoretinal complications, which are considered the main drawback of RLE and can occur after years, especially in more myopic eyes.15 Several techniques have been tried to correct ametropias in keratoconus. Photorefractive keratectomy is said to not only correct myopia and improve UCVA but also to stop keratoconus progression.7 The mechanism is unclear. In fact, corneal thinning may increase the corneal protrusion.2,8 The rationale for additive corneal surgery (namely, intracorneal rings) is the reinforcement and regularization of corneal structure, leading in most cases to an improvement in astigmatism, BSCVA, and UCVA.4–6 However, the correction of the spherical error is highly unpredictable4 and is usually more evident in cases of corneal-induced myopia. Despite the difficulties in obtaining accurate preoperative refraction, phakic IOLs are more predictable in keratoconic eyes with myopia3 and a toric model has been successfully used in 3 patients with keratoconus.16 Phakic IOLs have the advantage of an almost complete reversibility, but their use is not advised in patients of presbyopic age because of progressive convexity of the crystalline lens, which can induce cataract and pigment dispersion, especially with anterior chamber models.17 Therefore, RLE is indicated for presbyopic patients with higher spherical errors. The astigmatism can be addressed by a toric IOL or by intracorneal rings (in this case, ring implantation should precede clear lens extraction for a better IOL calculation). Toric IOLs in keratoconus have, as the main disadvantage, difficulty in determining axis and power of preoperative astigmatism. Rotational stability should not be an issue, especially with loop haptics.18
When RLE is considered, keratoconus stability is important to maintain long-term refractive results (because progression would lead to increased corneal myopia) and to reduce the risk for future keratoplasty (causing a hyperopic shift by means of a flatter cornea). In both cases, further adjustment could be achieved by implantation of a second (piggyback) IOL in the ciliary sulcus. Prolonged contact lens use can reduce corneal endothelial cell density19; therefore, patient selection must be accurate because intraoperative or postoperative IOL exchange may further affect the endothelium. In conclusion, RLE predictably corrected myopia associated with early-stage keratoconus. Intraoperative autorefractometry is recommended because IOL power calculation by US biometry can be inaccurate in almost one third of eyes. The use of more advanced biometric techniques as well as toric IOLs are required to improve predictability and reduce final defocus equivalent. However, IOL calculation precision is also affected by imprecise K-readings in keratoconus, and most IOL exchanges occurred with more irregular corneas. Therefore, this technique should be limited to early stages of keratoconus. Residual astigmatism should be addressed by secondary corneal procedures or (primarily) by the use of toric IOLs. REFERENCES 1. Feder RS. Keratoconus and pellucid marginal degeneration. In: Brightbill FS, ed, Corneal Surgery; Theory, Technique, and Tissue, 3rd ed. St Louis, MO, Mosby, 1999; 205–213 2. Randleman JB, Russel B, Ward MA, et al. Risk factors and prognosis for corneal ectasia after LASIK. Ophthalmology 2003; 110:267–275 3. Leccisotti A, Fields SV. Angle-supported phakic intraocular lenses in eyes with keratoconus and myopia. J Cataract Refract Surg 2003; 29: 1530–1536 4. Kwitko S, Severo NS. Ferrara intracorneal ring segments for keratoconus. J Cataract Refract Surg 2004; 30:812–820 5. Alio´ JL, Artola A, Hassanein A, et al. One or 2 Intacs segments for the correction of keratoconus. J Cataract Refract Surg 2005; 31:943–953 6. Boxer Wachler BS, Christie JP, Chandra NS, et al. Intacs for keratoconus. Ophthalmology 2003; 110:1031–1040; errata, 1475 7. Kasparova EA, Kasparov AA. Six-year experience with excimer laser surgery for primary keratoconus in Russia. J Refract Surg 2003; 19:S250–S254 8. Kremer I, Shochot Y, Kaplan A, Blumenthal M. Three year results of photoastigmatic refractive keratectomy for mild and atypical keratoconus. J Cataract Refract Surg 1998; 24:1581–1588 9. Ferna´ndez-Vega L, Alfonso JF, Villacampa T. Clear lens extraction for the correction of high myopia. Ophthalmology 2003; 110:2349–2354 10. Olsen T. Sources of error in intraocular lens power calculation. J Cataract Refract Surg 1992; 18:125–129 11. Sauder G, Jonas JB. Treatment of keratoconus by toric foldable intraocular lenses. Eur J Ophthalmol 2003; 13:577–579 12. Celikkol L, Ahn D, Celikkol G, Feldman ST. Calculating intraocular lens power in eyes with keratoconus using videokeratography. J Cataract Refract Surg 1996; 22:497–500 13. Jaffe NS, Clayman HM. The pathophysiology of corneal astigmatism after cataract extraction. Trans Am Acad Ophthalmol Otolaryngol 1975; 79:OP615–OP630
J CATARACT REFRACT SURG - VOL 32, MAY 2006
745
RLE IN KERATOCONUS
14. Steinert 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. Ophthalmology 1991; 98:417–423; discussion by DD Koch, 423–424; correction 1997; 104:1370 15. Colin J, Robinet A, Cochener B. Retinal detachment after clear lens extraction for high myopia: seven-year follow-up. Ophthalmology 1999; 106:2281–2284; discussion by M Stirpe, 2285 16. Budo C, Bartels MC, van Rij G. Implantation of Artisan toric phakic intraocular lenses for the correction of astigmatism and spherical errors in patients with keratoconus. J Refract Surg 2005; 21:218–222
746
17. Baikoff G, Bourgeon G, Jodai HJ, et al. Pigment dispersion and Artisan phakic intraocular lenses; crystalline lens rise as a safety criterion. J Cataract Refract Surg 2005; 31:674–680 18. Warlo I, Krummenauer F, Dick HB. Rotationsstabilita¨t monofokaler Intraokularlinsen mit C-Haptik versus Z-Haptik nach Kataraktchirurgie; Prospektiver randomisierter Vergleich. Ophthalmologe 2005; 102: 987–992 19. Esgin H, Erda N. Corneal endothelial polymegethism and pleomorphism induced by daily-wear rigid gas-permeable contact lenses. CLAO J 2002; 28:40–43
J CATARACT REFRACT SURG - VOL 32, MAY 2006
Refractive lens exchange in keratoconus Antonio Leccisotti, MD, PhD
PURPOSE: To evaluate the visual results and complications of refractive lens exchange to correct myopia associated with early-stage keratoconus. SETTING: Private practice, Siena, Italy. METHODS: This prospective noncomparative interventional series comprised 34 consecutive eyes of 20 patients with stages I to II keratoconus. Mean patient age was 56.7 years G 10.4 (SD). Preoperative mean spherical equivalent (SE) was –11.0 G 4.65 diopters (D), (range –5.75 to –22). Ultrasound biometry was performed using videokeratographic central K-readings and the Holladay 2 formula. An intraoperative handheld autorefractor was used to check the power of implanted intraocular lenses. RESULTS: Intraocular lens exchange due to inaccurate power occurred in 11 eyes (32%; 9 eyes intraoperative, 2 eyes postoperative). At 12 months, mean SE was –1.31 G1.08 D and mean defocus equivalent was 1.94 G 1.57 D. Twenty-two eyes (65%) were within G2 D of defocus equivalent, 16 eyes (47%) were within G1 D, and 3 eyes (9%) were within G0.5. Mean surgically induced astigmatism (vector analysis) was 0.54 G 0.43 D. Preoperative mean best spectacle-corrected visual acuity (BSCVA) was 0.55 G 0.20, and postoperative mean BSCVA was 0.76 G 0.23; the difference was statistically significant (P<.05; 95% confidence interval, 0.19 to 0.25). Postoperative mean uncorrected visual acuity was 0.48 G 0.25. The safety index was 1.38, and the efficacy index was 0.87. Complications were posterior vitreous detachment (9%) and dysphotopsia phenomena (15%). Corneal endothelial cell density at 12 months decreased by 6.3%. CONCLUSION: Refractive lens exchange in keratoconic eyes predictably corrected myopia. However, ultrasound biometry was inaccurate in almost one third of eyes. Intraoperative autorefractometry is recommended to improve refractive outcome. J Cataract Refract Surg 2006; 32:742–746 Q 2006 ASCRS and ESCRS
Keratoconus can be associated with axial myopia or can induce myopia by increased corneal curvature.1 Laser in situ keratomileusis in keratoconus may cause corneal ectasia,2 and surgical correction of the spherical error associated with keratoconus remains a challenge. We previously evaluated phakic intraocular lenses (IOLs) for this purpose,3 but these lenses have several anatomical limitations and are generally not implanted in patients older than 45 years. Other techniques, such as intracorneal rings, can improve visual acuity and reduce corneal curvature in keratoconus,4,5 but correction of the
Accepted for publication August 14, 2005. From the Ophthalmic Surgery Unit, Casa di Cura Rugani, Siena, Italy. The author has no financial or proprietary interest in any material or method mentioned. Reprint requests to Antonio Leccisotti, MD, PhD, Piazza 5 Bersaglieri, 2 53100 Siena, Italy. E-mail: [email protected]. Q 2006 ASCRS and ESCRS Published by Elsevier Inc.
742
spherical error is limited.6 Photorefractive keratectomy has been used by some researchers to treat myopia associated with keratoconus,7 but thinning of keratoconic corneas raises long-term safety concerns.8 Refractive lens exchange (RLE), also called clear lens extraction, consists of phacoemulsification of the clear crystalline lens and in-the-bag implantation of an appropriately powered IOL. It is generally used to correct large spherical errors in patients in the presbyopic age range because it causes loss of accommodation.9 Myopia associated with keratoconus is not considered an indication for RLE because of difficult IOL power calculation.10 In a Medline search of RLE in keratoconus, only a single case report was found,11 together with the suggestion (based on 2 cases) to use videokeratographic K-readings in biometry in keratoconus12 (October 11, 2005; keywords: keratoconus, intraocular lens, and lens exchange). We therefore evaluated the efficacy and safety of RLE for the correction of myopic spherical error associated with early-stage keratoconus. 0886-3350/06/$-see front matter doi:10.1016/j.jcrs.2006.01.063
RLE IN KERATOCONUS
PATIENTS AND METHODS Patients with keratoconus having RLE to correct myopia by 1 surgeon (A.L.) between September 2002 and July 2004 were evaluated. Inclusion criteria were age greater than 45 years, keratoconus stages I to II in each operated eye, subjective contact lens intolerance (discomfort), myopic refraction, measurable preoperative subjective refraction, best spectacle-corrected visual acuity (BSCVA) 0.2 or better, refractive stability for at least 3 years, and minimum follow-up of 12 months. Exclusion criteria were corneal scars, previous corneal surgery, untreated rhegmatogenous retinal degeneration, monocularity, endothelial cell density less than 1800 cells/mm2 (evaluated by specular microscopy), glaucoma or ocular hypertension, diabetic retinopathy, ocular inflammatory diseases, and crystalline lens opacities affecting preoperative BSCVA. Informed consent was obtained from all patients. This stated that only the spherical component would be corrected, that further enhancement by excimer laser was not possible, and that spectacles would be needed after the operation to correct residual errors. The study was approved by the ethics committee of our institute. Keratoconus was diagnosed by combination of computed videokeratography, K-readings, and corneal ultrasound (US) pachymetry. Pachymetry was used to reveal differences between superior and inferior thickness (I–S difference) of more than 100 mm. The keratoconus stage was assessed by the system included in the computed videokeratographer keratograph (Oculus), as in a previous study.3 The stages were as follows: Stage I: Minimum corneal curvature (MCC), 7.5 to 6.5 mm; index of surface variance (ISV Z deviation of corneal radium from average), 30 to 55; presumed BSCVA, 20/25 to 20/15. Stage II: MCC: 6.9 to 5.3 mm; ISV, 55 to 90; BSCVA, 20/60 to 20/20. Stage III: MCC, 6.6 to 4.8 mm; ISV, 90 to 150; BSCVA, 20/125 to 20/30. Stage IV: MCC, !5 mm; ISV, O150; BSCVA, worse than 20/100. Contact lens use had been discontinued for 2 months before corneal topography. Manifest and cycloplegic refraction, slitlamp microscopy of undilated and dilated anterior segment, fundus with Goldmann mirror, applanation tonometry, and dilated binocular indirect ophthalmoscopy were also performed before surgery. Intraocular lens power was calculated by US biometry using the Allergan Humphrey 820 and Holladay 2 formula, targeting a residual myopic astigmatism. Keratometry readings were obtained by axial topographic maps. The dioptric power of the steepest meridian over the central 3 mm of the cornea was considered K1, and the dioptric power of the flattest meridian over the central 3 mm of the cornea was considered K2. Each meridian power was calculated by averaging its 2 semimeridian values, always considering the central 3 mm. Sixty and 30 minutes before surgery, ciprofloxacin (Oftacilox) and tropicamide 1% eyedrops were instilled. Surgery was performed on an outpatient basis with topical anesthesia (unpreserved lidocaine 1%). Five minutes before surgery, povidone– iodine (Betadine) diluted at 5% was instilled in the conjunctival sac. A sclerocorneal 3.2 mm 3.0 mm tunnel was made, the anterior chamber was filled with an ophthalmic viscosurgical device (OVD) (sodium hyaluronate 1.0% [Provisc]), continuous curvilinear capsulorhexis was performed by forceps, and, after hydrodissection, the lens nucleus was emulsified and the cortex aspirated.
A single-piece foldable acrylic IOL (Stabibag [Rockmerd BV]) with a 5.5 mm square-edged optic was inserted in the capsular bag by an injector, OVD was aspirated, and a single 10-0 nylon radial suture was used to close the wound. With the patient still on the operating table and the eyelid speculum on, autorefractometry was performed with a handheld instrument (Nikon Retinomax 2). If a spherical error greater than 1.5 diopters (D) was found, the IOL power was immediately exchanged. Approximately C1.5 D was added to the original IOL power for each diopter of residual hyperopia and ÿ1 D was added for each diopter of residual myopia. Exchange was performed by filling the anterior chamber with OVD, lifting the IOL in the anterior chamber by a hook, cutting the IOL in 2 with Vannas scissors, removing the IOL parts, injecting the new lens into the capsular bag, and aspirating the OVD again. Ciprofloxacin and dexamethasone 0.1% eyedrops (Visumetazone) were used 4 times a day for 15 days. Follow-up visits were at 1 and 7 days and 1, 3, 6, 12, 18, and 24 months. Evaluation included uncorrected visual acuity (UCVA), BSCVA, corneal topography, endothelial cell count, slitlamp evaluation, and fundoscopy. Statistical analysis was performed by StatView (Abacus Concepts, Inc.), and P values less than 0.05 were considered statistically significant; 95% confidence intervals (CI) were calculated by CIA Software (BMJ). Predictability was measured as attempted versus achieved spherical equivalent (SE) and as the percentage of eyes within G0.5 D, G1.0 D, and G2.0 D of defocus equivalent. Surgically induced astigmatism was evaluated by vector analysis.13 Safety was evaluated by assessing the percentage of eyes losing lines of BSCVA and by safety index (postoperative BSCVA/ preoperative BSCVA). Efficacy was evaluated by assessing postoperative UCVA and by efficacy index (postoperative UCVA/preoperative BSCVA). Visual acuity was measured and all calculations performed in logMAR units. Data were converted into decimals for presentation. RESULTS
Thirty-four eyes of 20 patients were included in the study. Mean patient age was 56.7 years G10.4 (SD) (range 45 to 76 years). Eight of 20 patients (40%) were male. Twenty eyes had keratoconus stage I, and 14 eyes had stage II. Mean follow-up was 17.4 G 5.1 months (range 12 to 24 months). Surgery was uneventful in all cases. Intraoperative in-the-bag IOL exchange after autorefractometry was performed in 9 eyes (26%) of 6 patients (4 eyes with keratoconus stage I and 5 eyes with stage II). In 2 eyes (6%) of 2 other patients (both stage II), a second operation was required 2 weeks after the first to exchange the IOL because a ÿ5 D and a ÿ6 D postoperative SE had resulted. Intraoperative autorefractometry had not revealed these large errors. Preoperative mean SE was ÿ11.0 G 4.65 D (range ÿ5.75 to ÿ22) and ÿ1.31 G 1.08 D (range, ÿ0.25 to ÿ4.5) at 12 months. The difference was statistically significant (P!.05 with Student t test and 95% CI, ÿ8.26 to ÿ11.18 D) (Figure 1). Preoperative mean defocus equivalent was 12.0 G 4.64 D (range 6.50 to 23 D) and 1.94 G 1.57 D (range 0.25 to 5.5 D) at 12 months. The difference was statistically
J CATARACT REFRACT SURG - VOL 32, MAY 2006
743
RLE IN KERATOCONUS
40 35
% of eyes
30 25 20 15 10 5 0 <-1
-1
0
1
2
3
>3
BSCVA Snellen lines gained/lost
Figure 2. Changes in BSCVA 12 months after myopic refractive lens exchange in 34 keratoconic eyes.
significant (P!.05 with Student t test; 95% CI, 8.63 to 11.46 D). Twenty-two of 34 eyes (65%) were within G2.0 D of defocus equivalent, 16 of 34 (47%) were within G1.0 D, and 3 of 34 eyes (9%) were within G0.5 D. Preoperative mean refractive cylinder was 1.86 G 1.39 D (range 0.5 to 5) and at 12 months, 1.22 G 1.37 D (range 0 to 4). The difference was not statistically significant with Student t test but was significant with 95% CI (0.4 to 0.9 D). Mean surgically induced astigmatism, calculated by vector analysis, was 0.54 G 0.43 D (range 0 to 1.80 D). Preoperative mean BSCVA was 0.55 G 0.23 (range 0.2 to 0.9) G 0.20 and 0.76 G 0.23 (range 0.4 to 1.0) at 12 months. The difference was statistically significant (P!.05 with Student t test; 95% CI, 0.19 to 0.25) (Figure 2). Mean UCVA at 12 months was 0.48 G 0.25 (range 0.1 to 0.9) (Figure 3). The safety index was 1.38, and the efficacy index was 0.87. Stability was evaluated by paired comparison of mean defocus equivalent at 3 months (1.78 D) and at 12 months (1.95 D). The variation was not statistically significant (95% CI, ÿ0.51 to 0.20 D). Posterior vitreous detachment occurred in 3 patients (9%) after 1, 6, and 12 months, respectively. Posterior capsule opacification occurred in 6 eyes (17.6%) of 4 patients. Neodymium:YAG laser capsulotomy was performed in all cases with full functional recovery. Dysphotopsia phenomena were reported by 3 of 20 patients (15%), consisting especially of light reflections
744
DISCUSSION
Although the final refractive error was acceptable (mean SE ÿ1.31 D), this result required a high number (almost 1 of every 3 eyes) of intraoperative (26%) or postoperative (6%) IOL exchanges due to inaccurate
40 35 30
% of eyes
Figure 1. Scattergram of attempted versus achieved correction (SE) 12 months after myopic refractive lens exchange in 34 keratoconic eyes with 1-diopter interval bars.
caused by lateral light sources in mesopic conditions; however, night activities were not impaired, as patients favorably compared dysphotopsia with halos generated by rigid gas-permeable contact lenses. Postoperative corneal endothelial cell density at 12 months was available in 21 of 34 eyes. In these eyes, mean preoperative density was 2652 G 421 cells/mm2 and mean postoperative density was 2487 G 327 cells/mm2, with a 6.3% reduction. However, this reduction was not statistically significant (95% CI, ÿ70 to 401 cells/mm2).
25 20
Pre-op BSCVA
15
Post-op UCVA
10 5 0 <0,1
0,2
0,4
0,6
0,8
1
Visual Acuity
Figure 3. Efficacy of myopic RLE in 34 keratoconic eyes 12 months after surgery. Black columns, preoperative BSCVA and white columns, postoperative UCVA.
J CATARACT REFRACT SURG - VOL 32, MAY 2006
RLE IN KERATOCONUS
calculation of the IOL power, more commonly in eyes with keratoconus stage II. Intraoperative autorefractometry improved predictability, although in 2 eyes it was totally inaccurate. This can be explained by the following considerations: (1) The IOL position at the end of surgery was not definitive. (2) Patient fixation into the autorefractor was affected by prolonged retinal illumination during surgery. (3) Autorefractometry can be imprecise with irregular corneas. The use of optical measurement of axial length as well as advanced IOL formulas should improve IOL calculation and reduce the percentage of IOL exchanges. However, in a large series of myopic RLEs,4 final mean SE was ÿ1.22 D, very near our result in keratoconic eyes (ÿ1.31 D). In our series, final mean defocus equivalent was 1.94 D, which can be at least partially explained by a preoperative mean astigmatism of 1.86 D. As observed in our previous study,3 the effect of tunnel incision on keratoconic corneas is similar to that on normal corneas14; therefore, the use of toric IOL is justified in RLE in keratoconus.11 The main bias affecting our studies is the limited and heterogeneous (in terms of age and SE) sample. This explains the low incidence of vitreoretinal complications, which are considered the main drawback of RLE and can occur after years, especially in more myopic eyes.15 Several techniques have been tried to correct ametropias in keratoconus. Photorefractive keratectomy is said to not only correct myopia and improve UCVA but also to stop keratoconus progression.7 The mechanism is unclear. In fact, corneal thinning may increase the corneal protrusion.2,8 The rationale for additive corneal surgery (namely, intracorneal rings) is the reinforcement and regularization of corneal structure, leading in most cases to an improvement in astigmatism, BSCVA, and UCVA.4–6 However, the correction of the spherical error is highly unpredictable4 and is usually more evident in cases of corneal-induced myopia. Despite the difficulties in obtaining accurate preoperative refraction, phakic IOLs are more predictable in keratoconic eyes with myopia3 and a toric model has been successfully used in 3 patients with keratoconus.16 Phakic IOLs have the advantage of an almost complete reversibility, but their use is not advised in patients of presbyopic age because of progressive convexity of the crystalline lens, which can induce cataract and pigment dispersion, especially with anterior chamber models.17 Therefore, RLE is indicated for presbyopic patients with higher spherical errors. The astigmatism can be addressed by a toric IOL or by intracorneal rings (in this case, ring implantation should precede clear lens extraction for a better IOL calculation). Toric IOLs in keratoconus have, as the main disadvantage, difficulty in determining axis and power of preoperative astigmatism. Rotational stability should not be an issue, especially with loop haptics.18
When RLE is considered, keratoconus stability is important to maintain long-term refractive results (because progression would lead to increased corneal myopia) and to reduce the risk for future keratoplasty (causing a hyperopic shift by means of a flatter cornea). In both cases, further adjustment could be achieved by implantation of a second (piggyback) IOL in the ciliary sulcus. Prolonged contact lens use can reduce corneal endothelial cell density19; therefore, patient selection must be accurate because intraoperative or postoperative IOL exchange may further affect the endothelium. In conclusion, RLE predictably corrected myopia associated with early-stage keratoconus. Intraoperative autorefractometry is recommended because IOL power calculation by US biometry can be inaccurate in almost one third of eyes. The use of more advanced biometric techniques as well as toric IOLs are required to improve predictability and reduce final defocus equivalent. However, IOL calculation precision is also affected by imprecise K-readings in keratoconus, and most IOL exchanges occurred with more irregular corneas. Therefore, this technique should be limited to early stages of keratoconus. Residual astigmatism should be addressed by secondary corneal procedures or (primarily) by the use of toric IOLs. REFERENCES 1. Feder RS. Keratoconus and pellucid marginal degeneration. In: Brightbill FS, ed, Corneal Surgery; Theory, Technique, and Tissue, 3rd ed. St Louis, MO, Mosby, 1999; 205–213 2. Randleman JB, Russel B, Ward MA, et al. Risk factors and prognosis for corneal ectasia after LASIK. Ophthalmology 2003; 110:267–275 3. Leccisotti A, Fields SV. Angle-supported phakic intraocular lenses in eyes with keratoconus and myopia. J Cataract Refract Surg 2003; 29: 1530–1536 4. Kwitko S, Severo NS. Ferrara intracorneal ring segments for keratoconus. J Cataract Refract Surg 2004; 30:812–820 5. Alio´ JL, Artola A, Hassanein A, et al. One or 2 Intacs segments for the correction of keratoconus. J Cataract Refract Surg 2005; 31:943–953 6. Boxer Wachler BS, Christie JP, Chandra NS, et al. Intacs for keratoconus. Ophthalmology 2003; 110:1031–1040; errata, 1475 7. Kasparova EA, Kasparov AA. Six-year experience with excimer laser surgery for primary keratoconus in Russia. J Refract Surg 2003; 19:S250–S254 8. Kremer I, Shochot Y, Kaplan A, Blumenthal M. Three year results of photoastigmatic refractive keratectomy for mild and atypical keratoconus. J Cataract Refract Surg 1998; 24:1581–1588 9. Ferna´ndez-Vega L, Alfonso JF, Villacampa T. Clear lens extraction for the correction of high myopia. Ophthalmology 2003; 110:2349–2354 10. Olsen T. Sources of error in intraocular lens power calculation. J Cataract Refract Surg 1992; 18:125–129 11. Sauder G, Jonas JB. Treatment of keratoconus by toric foldable intraocular lenses. Eur J Ophthalmol 2003; 13:577–579 12. Celikkol L, Ahn D, Celikkol G, Feldman ST. Calculating intraocular lens power in eyes with keratoconus using videokeratography. J Cataract Refract Surg 1996; 22:497–500 13. Jaffe NS, Clayman HM. The pathophysiology of corneal astigmatism after cataract extraction. Trans Am Acad Ophthalmol Otolaryngol 1975; 79:OP615–OP630
J CATARACT REFRACT SURG - VOL 32, MAY 2006
745
RLE IN KERATOCONUS
14. Steinert 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. Ophthalmology 1991; 98:417–423; discussion by DD Koch, 423–424; correction 1997; 104:1370 15. Colin J, Robinet A, Cochener B. Retinal detachment after clear lens extraction for high myopia: seven-year follow-up. Ophthalmology 1999; 106:2281–2284; discussion by M Stirpe, 2285 16. Budo C, Bartels MC, van Rij G. Implantation of Artisan toric phakic intraocular lenses for the correction of astigmatism and spherical errors in patients with keratoconus. J Refract Surg 2005; 21:218–222
746
17. Baikoff G, Bourgeon G, Jodai HJ, et al. Pigment dispersion and Artisan phakic intraocular lenses; crystalline lens rise as a safety criterion. J Cataract Refract Surg 2005; 31:674–680 18. Warlo I, Krummenauer F, Dick HB. Rotationsstabilita¨t monofokaler Intraokularlinsen mit C-Haptik versus Z-Haptik nach Kataraktchirurgie; Prospektiver randomisierter Vergleich. Ophthalmologe 2005; 102: 987–992 19. Esgin H, Erda N. Corneal endothelial polymegethism and pleomorphism induced by daily-wear rigid gas-permeable contact lenses. CLAO J 2002; 28:40–43
J CATARACT REFRACT SURG - VOL 32, MAY 2006