Photorefractive Keratectomy (PRK) for Myopia Correction

Photorefractive keratectomy (PRK) consists of the application of energy of the ultraviolet range generated by an argon fluoride (ArF) excimer laser to the anterior corneal stroma to change its curvature and, thus, to correct a refractive error. The physical process of remodeling the corneal stroma by ultraviolet (193 nm wavelength) high-energy photons is known as photoablation.

An image depicting human eye anatomy can be seen below.

Human eye anatomy.

During the 1980s, several applications of the 193-nm ArF excimer laser were investigated, including its use on human corneas for the correction of refractive errors. In 1988, Munnerlyn, Kroons, and Marshall reported an algorithm relating diameter and depth of the ablation to the required dioptric change.

McDonald performed the first excimer PRK for the correction of myopia on a normally sighted human eye in the United States. That same year, the Food and Drug Administration (FDA) organized a phase 3 trial, the PRK study (which ended in 1996), to demonstrate the safety, predictability, and stability of PRK for the treatment of myopia. At the end of this trial, 2 ophthalmic companies, VISX and Summit, were allowed to manufacture excimer lasers for widespread use in the United States. Since then, Nidek also has obtained approval for the manufacture of excimer lasers in the United States, and several hundred thousand patients have undergone this procedure throughout the world. The first excimer lasers used to perform PRK in the late 1980s have been improved significantly in terms of size, efficiency, and accuracy.

Several epidemiological studies, including the Beaver Dam population-based survey taken in the United States, show a prevalence of myopia greater than 0.5 diopters (D), ranging from 43.0% in people aged 43-54 years to 14.4% in individuals older than 75 years.

The mechanism of ablation of the excimer laser appears to be photochemical in nature and is known as photochemical ablation or ablative photodecomposition. This highly localized tissue interaction is based on the fact that each photon produced by the ArF excimer laser has 6.4 eV of energy, enough to break covalent bonds.[1]

The intramolecular bonds of exposed organic macromolecules are broken when a large number of high-energy 193-nm photons are absorbed in a short time. The resulting fragments rapidly expand and are ejected from the exposed surface at supersonic velocities. This mechanism explains why only the irradiated organic materials are affected, whereas the adjacent areas are not affected.

The return of corneal innervation up to 5 years after PRK was measured. Corneal subbasal nerve density does not recover to near preoperative densities until 2 years after PRK, as compared to 5 years after laser in situ keratomileusis (LASIK).[2, 3]

A study comparing transepithelial PRK and laser surgery found that both offer effective correction of myopia at 1 year, but LASIK seemed to result in less discomfort and less intense wound healing in the early postoperative period.[4]

Clinical indications for PRK include the following:

• Myopia (-1.0 D to -6.0 D) - Higher corrections are associated with a greater risk of corneal haze formation; therefore, LASIK is generally the preferred procedure.
• Astigmatism (0.75 D to 3.0 D) - Higher corrections are associated with regression of the effect; therefore, LASIK is the preferred procedure.
• Hyperopia (+1.0 D to +4.0 D) - Haze and regression of the PRK effect have made LASIK the preferred procedure for most of these patients.
• Patients with documented evidence of a change in manifest refraction of less than or equal to 0.5 D (both cylinder and sphere components) per year for at least 1 year prior to the date of preoperative examination
• Patients aged 18-20 years for the reduction or elimination of myopia of less than or equal to -6.0 D spherical equivalent at the corneal plane
• Patients aged 21 years for the reduction or elimination of myopia from 0 D to -6.0 D spherical myopia at the spectacle plane with up to -3.0 D of astigmatism
• Patients aged 21 years or older with naturally occurring hyperopia from +1.0 D to +4.0 D spherical equivalent, with no more than 1.0 D of refractive astigmatism
• Correction of refractive errors following other ocular surgery, including cataract surgery - PRK has been performed in patients who had previous radial keratotomy (RK) surgery or penetrating keratoplasties, but, in those cases, LASIK appears to be preferable. A significant risk of corneal haze formation exists if PRK is performed on an eye with any previous corneal surgery, so LASIK is generally the procedure of choice because of its minimal haze risk
• PRK in corneas previously treated with LASIK
• Treatment of anisometropic amblyopia in children

PRK ablation of the anterior stroma takes place after removing the epithelium, which is approximately 40-50 µm in thickness. The Bowman layer is destroyed in the process of PRK with no known deleterious consequences. A residual stromal thickness of at least 250 µm after PRK is necessary to prevent future corneal ectasia. A residual stromal thickness of 400 µm or more is preferred. The epithelium can be removed via mechanical, laser, or chemical means.

Contraindications include collagen vascular, autoimmune, or immunodeficiency diseases; pregnancy or breastfeeding; keratoconus; medications, such as Accutane (isotretinoin) or Cordarone (amiodarone hydrochloride); and a history of keloid formation. A recent report on the outcome of PRK in African Americans, including those with a known history of dermatologic keloid formation, revealed that a history of keloid formation does not appear to have an adverse effect on the outcome. These results question whether known dermatologic keloid formation should be a contraindication to photorefractive keratectomy.

All patients who are surgical candidates for photorefractive keratectomy (PRK) should undergo preoperative screening with videokeratography (corneal topography) and corneal pachymetry. A high percentage of candidates for keratorefractive surgery have clinical or subclinical keratoconus. Preoperative corneal topography must be radially symmetric and free of irregular astigmatism; however, topographic-guided treatment of irregular astigmatism can be performed and is gaining in popularity. Patients suspected of having keratoconus are detected most easily with videokeratography using a 1.5-D interval scale by looking for a local area of corneal steepening.

Pseudokeratoconus is a form of localized corneal steepening caused by an artifact. The most common cause of pseudokeratoconus is found in contact lens wearers who have worn decentered contact lenses.

Contact lens warpage is a phenomenon resulting in destabilization of a patient's refraction. This effect may last for several weeks after contact lens use has been discontinued.

Quantitative descriptors of corneal optical performance, such as the Surface Regularity Index (SRI), can be helpful to quantify the smoothness of the ablation postoperatively.

The following histologic excimer-induced changes in corneal morphology have been reported:

• Epithelium: Initial reepithelialization occurs within the first 3-5 days. Over the following 6-18 months, the epithelium thickens primarily at the deepest part of the ablation site. No clear explanation exists for this phenomenon of epithelial hyperplasia. It is most likely secondary to the epithelium "smoothing out" the ocular surface in conjunction with the mechanical smoothing from the eyelids with each blink. ^[5] Morphologic and immunohistochemical studies demonstrate normal epithelial attachment complexes as evidenced by the presence of type III collagen (anchoring fibrils), beta 4-integrin (epithelial hemidesmosomes), and type IV collagen (basement membrane).
• Stroma: Within the first 24 hours after excimer PRK, stromal wound healing begins, as inflammatory cells invade the corneal stroma from the tear film. For at least 3 weeks following PRK, activated fibrocytes repopulate the treated area. These cells are responsible for the formation of new collagen and proteoglycan matrix.
• No significant changes have been found in the corneal endothelium or the Descemet membrane following excimer PRK as long as 250 μm of stroma remain.

Following photorefractive keratectomy (PRK), topical antibiotics should be used with a therapeutic contact lens until reepithelialization is complete. Furthermore, a weak corticosteroid, such as fluorometholone 0.1%, is frequently used to avoid excessive collagen deposition.

Topical intraoperative application of 0.02% mitomycin C can reduce haze formation in highly myopic eyes undergoing PRK.[6, 7, 8]

Preoperative evaluation before PRK includes the following:

• Refractive history (including history of contact lens wear)
• Keratometry
• Uncorrected and best-corrected visual acuity
• Manifest and cycloplegic refraction
• Videokeratography and keratometry
• Slit lamp examination
• Tonometry
• Dilated funduscopy
• Assessment of a patient's ability to tolerate the procedure under topical anesthesia, to fixate steadily, to lie flat without difficulty, to understand the nature of the procedure, and to give an informed consent

After placement of antibiotic and anesthetic drops in the eye, the eyelid is held open using a speculum. When using the mechanical epithelial removal technique, a 6.0- or 6.5-mm marker can be used, centered over the entrance pupil to mark the area where the epithelium is to be removed. The epithelium can be removed using a blunt spatula or a small blade, such as a Beaver 64.

After epithelial removal, the surface of the cornea must be wiped with a nonfragmenting sponge that has been soaked with balanced sterile saline and then squeezed so it is moist but not saturated. During the procedure, focusing properly on the stromal surface and avoiding excessive illumination to allow patient fixation are important. A Thornton ring or similar instrument can be used to gently help the patient maintain fixation without distorting the corneal surface.

The excimer laser beam is centered on the entrance pupil and focused on the anterior stromal surface, and the laser treatment is applied. Excimer lasers frequently have tracking devices to help properly center treatment.

Placing a therapeutic contact lens for an average of 3 days (and not uncommonly for up to 5 days) is helpful, and application of a combination of antibiotic/steroid drops is routine. Alternatively, a pressure patch following application of an antibiotic/steroid ointment is well tolerated. Analgesic and antianxiety oral medications can be used during the first 24 hours postoperatively.

Patients should be monitored daily after PRK until the reepithelialization is complete. At this time, the therapeutic contact lens can be removed, and the patient can be monitored 1 week later. The remaining follow-up interval times can double until the vision stabilizes.

For excellent patient education resources, visit eMedicineHealth's Eye and Vision Center. Also, see eMedicineHealth's patient education article Vision Correction Surgery.

Complications following PRK include the following:

• Undercorrections, overcorrections, and induced astigmatism: As the most common complications that occur following PRK, they may be caused by inadequate centration or focusing at the time of surgery or an inadequate surgical plan. Management of these complications often includes a retreatment using either a second PRK procedure or LASIK after the initial result has stabilized.
• Regression: This complication may occur postoperatively, particularly in young patients or those with high corrections for up to 6 months following PRK. Initial treatment with topical steroids may halt regression. If a regression results in visually significant changes, a PRK or LASIK retreatment can be performed once refractive stabilization has been documented if residual tissue is adequate.
• Reticular stromal haze: Why some patients develop this pattern on the anterior stroma is unclear. It has been associated with the use of postoperative nonsteroidal anti-inflammatory drugs (NSAIDs). In many cases, it does not affect a patient's visual acuity. Initial management of corneal haze usually consists of waiting until the opacity fades spontaneously. Treatment of persistent moderate-to-severe haze includes the use of topical steroids and/or a second laser application (eg, PRK, phototherapeutic keratectomy [PTK]). Only haze that is causing a reduction in visual acuity should be considered for retreatment. An association exists between significant regression and the formation of severe reticular haze.
• Increase in intraocular pressure (IOP) associated with the use of postoperative steroids: In all cases, IOP reverts to normal values on discontinuation of topical steroids. No reports of persistent increased IOP exist in the literature.
• Steroid-induced herpes simplex keratitis: This complication may be associated with the use of postoperative steroids or possibly from the stress of the excimer laser. In patients with a known history of herpes simplex viral keratitis, the use of preoperative and postoperative topical/oral acyclovir or trifluorothymidine drops is warranted, and some feel this is a relative contraindication.
• Posterior subcapsular cataracts: These have been reported infrequently in patients who used dexamethasone phosphate postoperatively.
• Superficial keratopathy: This has been reported in association with topical NSAIDs or the preservatives used in other drops.
• Complications related to the use of therapeutic contact lenses, including infectious keratitis: Treatment of this complication requires corneal scrapings for culture and sensitivity, as well as the use of fortified topical antibiotics.
• Infectious and sterile paracentral corneal infiltrates: These infiltrates can be treated with fortified topical antibiotics and topical steroids, respectively. Atypical organisms have been reported to cause infectious cases in addition to usual organisms.
• Delayed epithelial healing: This can occur in some patients, requiring a longer period of time with therapeutic contact lens to allow epithelium regeneration.
• Recurrent corneal erosions: This complication may occur in the presence of basement dystrophy unrecognized during the preoperative evaluation. The treatment may include PTK and the use of hypertonic saline solution.
• Decentration of the ablation zone: This may occur following inadequate centration on the entrance pupil at the time of laser application. The treatment may include a retreatment guided by corneal topography and/or 3D corneal topography (elevation map).
• Central islands: Excimer lasers with a flattop energy beam profile have the potential of producing central islands. Four main theories exist as to the cause of central islands, as follows: focal central epithelial hyperplasia, the vortex plume theory, degradation of laser optics, and the acoustic shock wave theory. However, central island formation is most likely multifactorial. Modern excimer lasers have built-in software to neutralize the possible appearance of central islands. This software acts by centrally applying additional pulses.
• Corneal melting and perforation: This has been reported in patients with connective tissue disease. A penetrating keratoplasty may be necessary in these cases to restore vision and ocular integrity.

For predictability of outcome, investigators collected the following data from patients who were operated on with a VISX excimer laser in a prospective, nonrandomized, unmasked, multicenter PRK clinical study and followed them for at least 2 years:

• In myopic PRK, refractive stability achieved at 1 year was maintained up to 12 years with no evidence of hyperopic shift, diurnal fluctuation, or late regression in the long term. Corneal haze decreased with time, with complete recovery of BSCVA. Night halos remained a significant problem in a subset of patients due to the small ablation zone size.
• After hyperopic PRK, refractive stability achieved at 1 year was maintained up to 7.5 years with no evidence of hyperopic shift, diurnal fluctuation, or late regression. Peripheral corneal haze decreased with time but was still evident in a number of eyes at the last follow-up visit.
• PRK for severe anisometropic amblyopia in children resulted in long-term stable reduction in refractive error and improvement in visual acuity and stereopsis, with negligible persistent corneal haze.
• The safety and efficacy of PRK after LASIK show good reduction of refractive error and improvement of UCVA and BSCVA. A significant undercorrection of astigmatism was attributed to surgically induced astigmatism. Further studies are necessary to determine the long-term safety and stability of outcomes.

Even though several reports demonstrate that long-term visual outcome of patients treated with PRK versus LASIK is equivalent for mild-to-moderate myopia, PRK has become a second choice procedure for most refractive surgeons.

When using the excimer laser, LASIK has become the preferred technique because of the lack of a significant amount of discomfort, the faster rate of postoperative visual rehabilitation, and the greater amount of stability.

However, PRK remains useful when treating patients whose corneas are too thin to perform LASIK and leaves a 250-µm stromal bed after the ablation is performed. PRK also is the procedure of choice when treating a refractive error associated with an uneven corneal surface or a superficial leukoma.

New tools for mapping the cornea, including wavefront technology, will provide more accurate information that can be linked to the excimer laser and allow a customized ablation.[9] However, one study found that wavefront-guided PRK offered no advantage over non – wavefront-guided PRK in predictability or safety.[10] Similarly, technical improvements in the design of excimer machines, such as eye-tracking devices, have minimized potential complications, such as decentration, following PRK.