SPECIAL ARTICLE A Cautionary Tale of Innovation in Refractive Surgery George O. Waring III, MD, FRCOphth Hyperopia is difficult to treat surgically. Nearly a dozen techniques strive toward safe, effective outcomes. All techniques are bedeviled by 3 challenges: (1) the need for accurate centration over the pupil of a steeper central optical zone approximately 5 mm in diameter, (2) the potential de- creased visual acuity caused by image minification,1,2 and (3) the creation of corneal contours with physiologic characteristics able to minimize optical aberrations and regression of initial refractive effect. Excimer laser photorefractive keratec- Three other methods for surgical cor- tomy3 and laser in situ keratomileusis rection of hyperopia have experienced (LASIK)4 for hyperopia are being devel- transient popularity until the high rate of oped with ongoing changes in ablation al- complications and lack of predictability gorithms. Thermal keratoplasty with a caused them to slink quietly away: hot- pulsed holmium:YAG laser has been needle thermal keratoplasty, which pro- plagued by marked regression of effect duced focal corneal necrosis and an un- when attempting to treat more than ap- stable refraction; hexagonal keratotomy, proximately +1.00 diopter (D) of hyper- which commonly produced irregular astig- opia.5,6 An infrared continuous-wave di- matism; and hyperopic automated lamel- ode laser (wavelength of 1.9 µm) is also lar keratoplasty (also known as H-ALK; being studied in clinical trials.7 Plus- more properly, deep lamellar keratotomy power phakic intraocular lenses face the for hyperopia), which produced progres- design challenge of being placed in hy- sive corneal ectasia in some eyes. peropic eyes that often have shallower an- terior chambers and less surgical space for THE LIMITS OF KERATOMILEUSIS implantation than myopic eyes8; contin- Deep lamellar keratotomy for hyperopia ued design changes and careful clinical tri- (“hyperopic ALK”) attempts to create con- als will probably yield safe and useful pha- trolled steepening of the central cornea by kic intraocular lenses for hyperopia within cutting with a microkeratome a lamellar the next few years. Clear lens extraction disc that is approximately 75% of corneal for high hyperopia with placement of 1 or thickness. This disc is left in place with- 2 intraocular lenses in the capsular bag or out sutures. The remaining thinner pos- ciliary sulcus has successfully treated these terior cornea displaces forward with in- patients who otherwise have consider- creased curvature to reduce hyperopia. The able visual disability.9 Intracorneal lenses procedure was done from approximately have a 50-year history of unsuccessful 1993 to 1996 on an estimated 60 000 eyes. clinical trials in the treatment of hyper- 12 10 In 1998, Lyle and Jin reported 67 con- opia and aphakia, but new materials and secutive eyes treated with hyperopic ALK surgical techniques and designs, such as with a 67% follow-up at 1 year. They ob- radially placed intracorneal polymethyl served that although the procedure steep- methacrylate stents and keratophakia with ened the central cornea and improved un- new synthetic lenticules, keep future pos- corrected visual acuity, instability of sibilities interesting. Intracorneal lenses refraction ensued with a mean myopic shift with small diameter and high index of re- of 0.50 D between 3 months and 1 year fraction have been studied in a small num- and of 1.00 D between 1 and 2 years. In ber of eyes, but decreased quality of vi- addition, 26% of the eyes developed sion led to termination of the research 11 enough steepening and irregular astigma- without publication of results. tism to be dubbed “iatrogenic keratoco- From the Department of Ophthalmology, Emory University School of Medicine, nus,” and 16.4% had enough ectasia to re- Atlanta, Ga. quire penetrating keratoplasty. ARCH OPHTHALMOL / VOL 117, AUG 1999 1069 ©1999 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 10/01/2021 The concept of creating a con- timates the minimal thickness of the late viscoelastic and wound heal- trolled corneal ectasia in refractive bed for long-term stability to be 200 ing processes that occur over time. surgery is not new. Radial kera- to 250 µm. This figure is based on The distribution of the stress- totomy for myopia intends to pro- the observation that for some 30 bearing layers of the cornea has been duce a controlled ectasia in which years, cryolathe and nonfreeze ker- studied, but not definitively charac- the corneal steepening occurs para- atomileusis were done in normal terized. MacPhee and colleagues23 re- centrally with concomitant central corneas that had a presumed cen- ported that all corneal layers equally flattening13; hexagonal keratotomy tral thickness of approximately 500 bear stress. The finite element simu- for hyperopia attempted to create to 550 µm, by creating an approxi- lations of Hanna and colleagues22 central corneal steepening14; and hy- mately 300-µm thick disc that then suggested that more stress was borne peropic ALK, with a similar biome- received a refractive cut and was re- by the posterior layers than the an- chanical concept, aims to produce placed with sutures, leaving a cor- terior layers. Seiler and colleagues24 a controlled central steepening. The neal bed 200 to 250 µm thick. Car- demonstrated that Bowman layer it- “control” is determined by the bio- men Barraquer19 reported that of self does not contribute more biome- mechanical properties of the indi- 1606 eyes that received cryolathe chanical support to the cornea than vidual cornea and is set by the length myopic keratomileusis (with a disc does the cellular stroma. This theo- and depth of the incisions, whether thickness of approximately 300 µm) retical information supports the con- radial, hexagonal, or lamellar. A la- and were followed up for approxi- cept that lamellar keratotomy made mellar keratotomy is also done dur- mately 2 decades, 45 (2.8%) devel- too deeply in the cornea will pro- ing keratomileusis (including oped corneal ectasia. She also re- duce corneal ectasia. LASIK) to create a flap or disc that ported the changes in average The major reason that the mini- reveals the stroma for refractive cor- corneal power (and radius of cur- mal residual thickness required for rection. The changes in corneal vature) over time: before surgery, corneal stability after keratomileu- shape after radial, hexagonal, and la- 43.75 D (7.7 mm); at 30 days, 35.50 sis remains unknown is that there mellar keratotomy occur in 3 time D (9.5 mm); at 90 days, 36.75 D (9.2 has been no reliable way to mea- frames: (1) acutely, within hours, re- mm); at 3 to 12 months, 37.12 D (9.1 sure the thicknesses of individual sulting in a substantial—usually de- mm); at 5 years, 37.87 D ( 8.9 mm); corneal layers postoperatively. The sirable—change in refraction and at 10 years, 39.25 D (8.6 mm); and ultrasound biomicroscope that uses visual acuity; (2) short-term fluctua- at 21 years, 39.25 D (8.6 mm). The a 50-MHz transducer can give beau- tion, such as diurnal variation after causes for this mild progressive cor- tiful images of portions of the cor- radial keratotomy15; and (3) gradu- neal steepening of approximately 2 nea (4-mm wide), but the 20-µm ally over a few years, as occurs in D from 1 to 20 years after keratomi- precision of this system25 is insuffi- progressive ectasia after hyperopic leusis are unknown, but possibili- cient to adequately evaluate cor- ALK and in the hyperopic shift ties include weakened biomechani- neal refractive surgery. A newer that occurs after radial keratotomy cal support in the corneal bed, digital 50-MHz high-frequency ul- with incisions that extend to the underlying disease such as undetec- trasound system described by Rein- limbus.16,17 ted keratoconus, variations in sur- stein and colleagues26 can resolve The use of lamellar kera- gical technique that created a thin- and measure the epithelium, the stro- totomy in hyperopic ALK and in ker- ner corneal bed than intended, mal component of the corneal flap, atomileusis raises a fundamental chronic trauma to the cornea such and the thickness of the residual stro- question about corneal biomechan- as chronic eye rubbing, and epithe- mal bed in keratomileusis with a ics: What is the minimal thickness lial hyperplasia. Unfortunately, there precision of 1.3 µm (Figure). These of the cornea that can preserve cor- is no published long-term fol- measurements may create a better neal shape for the lifetime of the pa- low-up on a consecutive series of understanding of the limits of pro- tient without development of pro- eyes with a thorough statistical cedures that remove corneal tissue gressive corneal steepening or frank analysis of changes in corneal cur- to change refractive shape, such as ectasia? No one knows. An unsu- vature after keratomileusis. photorefractive keratectomy and tured corneal disc or flap that is cre- Jose´ Barraquer20,21 stated that a LASIK. Such postoperative measure- ated during keratomileusis contrib- residual corneal thickness of 300 µm ments in human eyes are neces- utes minimally to the biomechanical should remain to prevent ectasia. sary, because we do not know at the strength of the cornea, because its Hanna and colleagues22 used finite time of surgery the actual thick- attachments to the limbus have been element modeling of the cornea to ness of the corneal flap or the severed. Thus, the corneal strength demonstrate that resection of a layer actual ablation depth within the and shape are determined by the bio- of cornea approximately 50% its stromal bed; manufacturers give mechanical properties of the re- thickness (250 µm, 7.5-mm diam- guidance to the surgeon as to the sidual corneal bed, which are prob- eter) doubled the stresses at the cen- predicted thickness of the corneal ably different for each individual.