In an ironic turn of events, plummeting fees for refractive surgery have coincided with skyrocketing patient expectations. As fees in some areas of the country begin to approach the prices of specialty contact lenses, patients expect their vision to be as good as with contact lenses the moment they get up from under the laser.

They have trouble understanding that postoperative healing is necessary. Therefore, it's our job to explain that even as improved corneal ablation profiles promise faster visual recovery and aberration-free vision, refractive surgery will still remain a process rather than an event. And that process really begins when the patient walks away from the laser and his corneas begin to heal.

It's also incumbent upon us to gain a better understanding of the wound-healing process so we can manage patients' expectations better. Ultimately, we should be able to modulate the process to help patients achieve superior outcomes. In this article, I'll review what we know about corneal wound healing and highlight new information that may change or confirm our preferred treatment plans.

Preoperative lactogerrin(LFN0 values vs. postoperative refraction at 3 months after LASIK. Low preoperative LFN values correlate with myopic refration postoperatively. High LFN values correlate with hyperopic refraction postoperatively.

Healing after PRK

Laser-assisted in situ keratomileusis (LASIK) is now the most common method of surgical vision correction. However, our understanding of the wound-healing process following excimer laser surgery began years ago with photorefractive keratectomy (PRK). PRK is still the preferred procedure in certain cases, particularly when corneas are too thin for LASIK, or when the patient has epithelial basement membrane dystrophy and is at high risk for epithelial ingrowth and slow visual recovery.

Following PRK, both the corneal epithelium and stroma heal vigorously. Immediately following keratectomy, the injured epithelial cells release inflammatory mediators and chemotactic factors, such as interleukins and growth factors TGFb, bFGF and EGF.

These cytokines attract inflammatory cells, such as polymorphonuclear leukocytes (PMNs) and monocytes. The inflammatory cells migrate to the wound site via tears and clean the wound of debris. The inflammatory cells release more cytokines that stimulate epithelial migration and proliferation, facilitating closure of the epithelial defect.

In the stroma immediately after PRK, keratocytes underlying the ablation site disappear. Cell death occurs by apoptosis. The extent of the cell death depends on the method of epithelial removal prior to the laser ablation. Transepithelial PRK is associated with less keratocyte apoptosis than mechanical epithelial debridement. Transepithelial PRK is also associated with less corneal haze postoperatively.

Based on this information, it's been hypothesized that the extent of keratocyte apoptosis influences the extent of stromal wound healing. Dying keratocytes release inflammatory mediators that stimulate stromal healing. It was, therefore, further proposed that if we can prevent keratocyte apoptosis, we can eliminate excessive stromal healing and haze after PRK.

On the other hand, we can also hypothesize that keratocyte apoptosis is not the cause of stromal healing. But rather that it occurs coincidentally with stromal healing. The injured epithelium and the inflammatory cells at the wound site release inflammatory mediators. When epithelial injury is minimal, as in transepithelial PRK, fewer inflammatory mediators are released and less stromal wound healing response results. There is also less keratocyte apoptosis. If this second scenario proves true, to prevent excessive stromal healing following PRK, we'd aim therapy at reducing epithelial trauma and blocking inflammatory mediators rather than trying to prevent keratocyte apoptosis.

Why healing after LASIK is different

Corneal healing at the outer edge of the LASIK flap is identical to healing after PRK. This finding supports the hypothesis that when injured epithelium comes in contact with stroma without the protection of the Bowman's layer, a specific wound-healing cascade begins that results in corneal haze at the wound site. Over time, corneal remodeling causes the haze to fade and the flap edge becomes barely detectable.

Just as in PRK, the keratocytes disappear initially from the area of the cornea adjacent to the keratectomy site. Interestingly, keratocytes also disappear from the flap, even though the excimer ablation is performed only on the stromal bed. Perhaps the inflammatory mediators from the disrupted epithelium at the flap edge are dispersed across stroma by a microkeratome or a spatula and induce keratocyte apoptosis.

As in PRK, activated keratocytes begin to reappear within several days after LASIK. But they're not as numerous or as activated as after PRK, and unlike PRK, their activity normalizes within several weeks. These findings are consistent with minimal stromal remodeling, collagen deposition, and fewer problems with haze after LASIK compared with PRK.

Dealing with the downsides of the healing process

Postoperative wound healing has welcome and unwelcome results. In many instances, corneal healing helps to improve the outcome of refractive surgery. The edge of the LASIK flap heals so well that a month after the procedure, only severe trauma or surgical intervention can displace it.

Stromal remodeling after PRK and epithelial remodeling after LASIK cause most central islands and other corneal irregularities to disappear with time. Corneal sensation and normal tear film are restored 6 to 8 months after surgery. Glare and halos decrease with time as the profile of the ablation edge gets smoother through corneal healing and remodeling. Spherical aberration for a 3-mm pupil resolves at 6 weeks after LASIK and 6 months after PRK as the corneas heal and become less oblate.

But, as you know, corneal healing can also hinder refractive surgery results, despite advances in microkeratome and excimer laser technology to improve ablation profiles. Let's take a closer look at some unwanted effects and the latest thinking on the best ways to prevent and treat them.

�         Regression. Regression is not uncommon following PRK or LASIK, although stromal remodeling is minimal after LASIK. Based on recent evidence from studies performed with very high-frequency digital ultrasound, regression after LASIK is associated with epithelial hyperplasia. It's hypothesized that epithelial hyperplasia may be due to both hyperactive basal epithelium and decreased exfoliation of the superficial epithelium in ablated cornea depressed by the eyelids during blinking.
Regression rarely recurs after an enhancement, even though the ablated area becomes depressed again. Possibly the basal epithelium becomes less active as time passes after original keratectomy, or perhaps the second correction is simply too minimal to induce significant activity of the basal epithelium. If the activity of the basal epithelium indeed normalizes at 3 months after the original LASIK, partial ablation of superficial epithelium following LASIK regression may produce a stable result.
The mechanism of regression after PRK is different. Regression is associated with regrowth of the ablated stroma. Epithelial thickness is normal.
Because the mechanisms of regression are different in LASIK and PRK, perhaps the preventative measures should be different as well. Prevention of regression following LASIK should aim at preventing epithelial hyperplasia by normalizing activity of the basal epithelium. Prevention of regression following PRK should aim at controlling stromal wound-healing response. In either case, the inflammatory mediators responsible for the epithelial and stromal wound-healing response need to be identified and controlled.
We should also keep in mind that ablation zones with steep edge contours and less than 6-mm diameters are associated with more epithelial hyperplasia and regression of refractive outcomes than larger ablation zones with smoother edge contours. Smooth ablation profiles induce less stromal healing response and are associated with less corneal haze and regression clinically.
Interestingly, the preoperative level of lactoferrin (LFN) in the tear film appears to correlate with refractive outcome following LASIK. (See "LFN vs. Post-Op Refraction" )
LFN is a protein secreted by the lacrimal gland. Levels of LFN in the tear film correlate directly with the amount of tears produced. Preliminary studies suggest that low LFN levels are associated with a myopic refraction following LASIK. LFN controls cell growth. If there is not enough LFN, cell growth may be accelerated, leading to epithelial hyperplasia. Alternatively, LFN may simply be a marker for low tear volume. When tear volume is low, inflammatory mediators, such as epidermal growth factor, may be concentrated and result in epithelial hyperplasia.
Regardless of the pathogenesis, preoperative LFN levels may be useful in identifying patients who are likely to experience myopic regression postoperatively -- and possibly can be used as a guide to either treat them to increase their tear volume or modify the ablation nomograms. An LFN assay test is easily performed on a tear sample collected with a precalibrated capillary tube at the slit lamp.

�         Post-PRK haze. Another unwelcome side effect of corneal wound healing is haze after PRK. Haze is a result of the stromal wound-healing response.
Within several days after PRK, keratocytes repopulate the stroma underlying the ablation site. The new keratocytes are active, reflective, and produce new collagen and other extracellular matrix components. Recent studies with confocal microscopy indicate that haze is associated with this increased reflectivity from high numbers of activated Keratocytes. 
Keratocytes may remain active for as long as 12 months after PRK. This residual long-term activity of the keratocytes may explain why corneal haze may appear months after PRK, following exposure to ultraviolet light, for example.
Haze is typically treated with phototherapeutic keratectomy (PTK). But multiple modalities have been tried to prevent corneal haze after PRK, including transepithelial PRK, cooling the cornea prior to the procedure, and pharmacologic means, such as minoxidil (Rogaine) heparin, and latanoprost (Xalatan). Some ophthalmologists have found that controlling corneal moisture with artificial tears, punctal occlusion and moisture-retention goggles can be important in haze prevention.
In some studies, antibodies to the TGFb inflammatory mediator of wound healing have been found to prevent haze formation. Most recently, superficial keratectomy combined with 0.02% Mitomycin C applied for 2 minutes intraoperatively has been successful.

�         Uneven healing. The cornea may also heal unevenly following either PRK or LASIK. Uneven healing may induce irregular astigmatism, glare, and coma-like aberrations. Patients with sub-clinical epithelial basement dystrophy (EBMD) preoperatively may develop epithelial disturbances that may impair their visual acuity and cause recurrent erosions postoperatively. With time and lubrication, the corneal surface will become smoother. Hypertonic saline ointment at bedtime may also help patients with EBMD.

Beware of using the term "20-minute miracle"

Corneal healing after refractive surgery is a complicated and multifactorial process. Advanced diagnostic modalities have helped us to better understand this process, while improvements in surgical techniques have reduced corneal reactivity. New treatment options are emerging to help control wound healing.

Nevertheless, healing after refractive surgery will always remain a process. The phrase "20-minute miracle" simply refers to how long it takes to perform the procedure.

New Tools Help Unravel the Mystery

Though we still have much to learn about post-refractive surgery wound healing, we do know it's a highly complex process whose course can be determined by a variety of factors.

Decades of research have attempted to unravel this mystery. In the past, studies have been limited by our inability to analyze wound-healing components in patients in vivo. Animal studies may not be directly applicable to patients because of the differences in corneal structure between most animal species and humans. Post-mortem studies are inadequate because they're static, and wound healing is a dynamic process taking place in a living organism.

With the advent of new diagnostic technologies, such as in vivo confocal microscopy and very high-frequency digital ultrasound, we can now analyze components of the postoperative wound-healing process in patients in vivo. We can now map each layer of the healing cornea with up to 1.3 microns of accuracy. This capability affords us a much more revealing look at what happens to the cornea after refractive surgery.

Ella G. Faktorovich, M.D., is director of the Pacific Vision Institute in San Francisco. A fellowship-trained refractive and corneal surgeon, she has been performing LASIK in San Francisco for 2 years. Corneal wound healing is one of her main interests.