The Limits of LASIK
One surgeon shares his criteria for deciding which patients are good candidates -- and which are not.
By Jonathan M. Davidorf, M.D., West Hills, Calif.

Today, LASIK is a common surgery. However, with so many patients willing to undergo the procedure and so much competition for those patients, it's important to remain alert for the situations in which LASIK is not the right treatment.

Some of the warning signs, of course, are obvious. Others, however, are only gradually becoming apparent as our experience with LASIK grows. LASIK is, after all, irreversible. It's up to us to make sure we guide poor LASIK candidates to other, more appropriate options.

In this article, I'll share some of what I've learned about this issue during years of performing LASIK. Hopefully, my experience will help you to avoid a wrong turn while traveling down the sometimes bumpy refractive surgery road.

Refractive error

When a patient's refractive error is beyond certain limits, outcomes begin to lose efficacy, predictability, stability and safety. For that reason, I prefer to perform LASIK on patients who have less than 12 diopters of myopia or 4 diopters of hyperopia. This doesn't mean that LASIK cannot be performed on patients whose refractive errors exceed these limits; patients simply need to understand that if they fall outside of the LASIK comfort zone, the likelihood of an excellent, problem-free outcome begins to diminish.

For patients who fall outside of these limits, I prefer to offer intraocular refractive surgery. For pre-presbyopic patients, I lean toward the use of phakic intraocular lenses; for presbyopic patients, I prefer clear lens exchange surgery. (Because of the risk of retinal detachment with a clear lens exchange procedure in an extreme myope, I recommend phakic IOL surgery until cataractous changes -- albeit sometimes mild -- have been identified.)

Although I have experience with several excimer laser systems, I currently treat my patients using either the VISX Star S2 or the Alcon Summit Autonomous LADARVision laser. With these lasers I can treat most levels of myopia and hyperopia, with or without astigmatism. In addition, the Alcon Summit Autonomous laser has been approved for the treatment of mixed astigmatism.

Corneal thickness

One of the most feared potential complications following LASIK is iatrogenic corneal ectasia. To avoid this, most surgeons try to leave a minimum posterior stromal bed of 250 µ or more. However, we have no conclusive data to support this limit as a safe corneal thickness following LASIK. For that reason, I don't like to make a habit of testing the limits.

When a cornea has a thickness of less than 500 µ, I always consider performing PRK, even if the refractive error is small. Although PRK patients experience a frustrating early postoperative course, most are willing to accept this if you explain that visual outcomes are similar and PRK may be safer than LASIK in their specific situation.

Some surgeons avoid PRK because of the issue of haze following the surgery. However:

The risk of visually significant haze following PRK was less than 5% in the FDA clinical trials for the VISX and Summit lasers. (The risk was even lower in the clinical trials for the Nidek, Alcon Summit Autonomous and Bausch & Lomb scanning lasers, suggesting that scanning laser technology may help lower the incidence of corneal haze following PRK.)

• With a smaller correction, visually significant haze is particularly uncommon.
• When haze does develop, it can usually be managed efficaciously using treatments such as laser scraping with mitomycin C applications.
• Some recent evidence indicates that the modification of PRK known as laser epithelial keratomileusis (LASEK), which is a stromal ablation beneath an epithelial flap, may further reduce the risk of corneal haze. (For more information on LASEK, see "A Look at LASEK")

Topography/keratometry

LASIK may not be an appropriate surgery for patients with certain corneal contours and conditions.

Keratoconus. Patients with keratoconus and unstable refractive errors are not good candidates for LASIK. Even forme fruste keratoconus patients shouldn't be considered for LASIK -- especially patients with thin corneal pachymetry measurements. LASIK could produce an unstable situation that did not exist preoperatively.

Because PRK doesn't involve the creation of a corneal flap, the corneal architecture is not destabilized. In my experience, forme fruste keratoconus patients do well with PRK, as long as the refractive error has been stable. However, it's important to remember that our aim is to correct regular refractive astigmatism. The correction of irregular astigmatism in these patients is not widely available, and the results have been mixed.

Irregular astigmatism. Some patients, such as those with corneal scars, demonstrate irregular astigmatism without the topographic pattern of keratoconus. If these patients can obtain good vision using spectacles, you should be able to reproduce the correction using LASIK (as long as all other variables are within the normal range). However, if a patient doesn't achieve satisfactory vision with spectacle correction, LASIK is not a good option. (Given the promising reports of topographic and wavefront-guided ablations, we can remain optimistic that help for these difficult cases is not far off.)

Flat or steep keratometry reading. Flat K's may increase the likelihood of small-diameter flaps. This can become problematic when treating hyperopic patients where a large diameter ablation is needed. Conversely, steep K's may increase the likelihood of buttonhole formation during keratectomy.

I generally do not reject patients because of excessively flat or excessively steep keratometry readings. However, I do make note of it so that I can be certain that intraocular pressure during the microkeratome pass is adequately elevated. This helps to ensure that the flap diameter is large enough when the patient has a flat cornea, and that the risk of buttonhole formation is minimized when the patient has a steep cornea. (Another alternative is the Pulsion Femtosecond laser microkeratome from Intralase, which may help produce a more predictable flap under these conditions.)

When performing surgery on a hyperope, I attempt to leave the postoperative keratometry reading under 50 diopters. If the keratometry reading is greater than this, patients seem to have an increased risk of dry-eye symptoms and decreased quality of vision. (You can estimate postoperative keratometry values by adding the dioptric value of the planned spherical equivalent refractive correction to the patient's preoperative keratometry values.)

Pupil size

I don't believe that patients should be rejected for LASIK solely on the basis of their pupil size. Although patients with higher corrections and large pupils do have a greater risk of experiencing night vision problems following LASIK, pupil size is not the only contributing factor. (I've seen some patients with below-average pupil diameters and low to moderate corrections who nevertheless have experienced nighttime vision problems following LASIK.)

I've performed LASIK on patients whose pupil size was larger than the ablation diameter without causing scotopic vision problems. Of course, it's imperative that we measure pupil size and conduct an in-depth discussion of the risk of night vision problems with our patients, particularly those with larger diameter pupils and higher corrections. Patients must understand that they may not be able to drive at night if they develop these symptoms, or that a rigid gas permeable contact lens may be required postoperatively to improve symptoms.

Despite these risks, some patients will choose to proceed because of high priority daytime activities they want to perform without glasses or contact lenses. The key is that the patient must make an informed decision about his own eyes.

Co-morbid conditions

The health of the eye is another obvious consideration when deciding whether LASIK is an appropriate option:

Cataract. I usually recommend clear lens exchange surgery for patients with early cataracts (even incipient cataracts that are not visually significant). Occasionally, I make an exception for a patient with a very early cataract if it's a longstanding cortical opacity and the refractive error is small.

Performing LASIK on a patient with an incipient cataract increases his risk of night vision problems and glare symptoms following surgery. Management of this type of patient will be difficult because it's not easy to determine whether the symptoms are being caused by the LASIK procedure or the cataract.

Diabetes Mellitus. Patients with active diabetic retinopathy should not be treated with LASIK. However, it's not unreasonable to perform LASIK in well-controlled diabetics who don't have diabetic retinopathy. These patients must understand, prior to proceeding, that we currently have no good data concerning LASIK in diabetics.

Glaucoma. Patients with unstable glaucoma or visual field damage should not be considered candidates for LASIK. However, I have successfully performed LASIK on patients with ocular hypertension or borderline glaucoma who demonstrated no visual field loss and had well-controlled intraocular pressure. Unfortunately, no good studies on this subject have been done, and patients need to be informed of this. (In fact, some evidence indicates that LASIK can cause subtle nerve fiber layer damage. Also, especially after higher myopic treatments, intraocular pressure tends to read lower. This can affect post-op management of ocular hypertension and glaucoma.

Dry eye. This can be an important consideration, particularly with hyperopes, who are often older than myopic patients and tend to have drier eyes at baseline. Patients with keratitis sicca must understand that their symptoms may become more pronounced following LASIK. If the ocular surface is not free of superficial punctate staining preoperatively, I prefer to rehabilitate the surface before proceeding with LASIK. In more pronounced situations, I insert punctal plugs.

I'm looking forward to the availability of conductive keratoplasty (CK), which should provide hyperopic patients with an alternative that may have a lower risk of dry eye.

Herpes Simplex Keratitis. LASIK can be performed successfully on patients with a history of herpes simplex keratitis. However, both ultraviolet light (such as that produced by an excimer laser) and trauma (such as lamellar keratectomy) can trigger herpes reactivation. If a patient is willing to take this risk, place him on both oral and topical antivirals preoperatively. Withhold steroid drops unless significant inflammatory response (i.e. diffuse lamellar keratitis) surfaces.

Other systemic disease. Although the excimer laser has not been approved for use in patients with connective tissue disease, it's been my experience that many of these patients can be treated with LASIK. I screen them by making sure they don't display an abnormal wound healing response to other injuries, such as cuts and bruises.

Many of these patients have a significant dry eye problem, and this may become the central issue during follow-up care.

Previous surgery. Patients who have undergone penetrating keratoplasty or cataract surgery may undergo LASIK. However, if the patient has had a trabeculectomy, PRK would be a better procedure, although the possible need for a lengthy postoperative corticosteroid eye drop regimen must be considered.

Age

LASIK should not be performed as elective surgery on any patient under the age of 18. Mounting evidence, however, suggests that LASIK may be an option for some younger patients with anisometropic amblyopia, particularly patients under the age of 10 for whom conventional amblyopia therapy has failed.

I'm not aware of any upper age limit for performing LASIK. I have performed LASIK on patients up to the age of 85.

Making the right choice

Fortunately, LASIK is not the only option we can offer our refractive surgery patients. The danger is not that poor LASIK candidates won't have an option for decreasing their glasses or contact lens dependency. The danger is that someone will choose to perform LASIK when it's not appropriate, simply because it's the easy thing to do.

By sharing our experience, and watching for signs that a patient is not a good candidate for LASIK, we can ensure that both our patients and we are happy at the end of the day.

Dr. Jonathan Davidorf is medical director of the Davidorf Eye Group and clinical instructor of ophthalmology at the Jules Stein Eye Institute in Los Angeles.