Patient Management
Epithelial Damage During LASIK: The pH Factor
A surgeon uncovers a key cause of a perplexing LASIK complication.
By Brian R. Will, M.D., Vancouver, Wash.

One of the challenges LASIK surgeons often face is the premature demise of the epithelial layer of the cornea during LASIK surgery. Many surgeons may consider epithelial "sloughing" or an occasional epithelial defect to be relatively insignificant. However, it's becoming increasingly evident that such side-effects of surgery can have significant postoperative implications -- especially if the visual axis is involved.

Ramifications of epithelial damage include:

• immediate issues such as ocular pain, blurred vision, increased difficulty in re-approximating the LASIK flap to the bed, and increased risk of inflammation and infection
• intermediate problems, including ocular irritation and dry eye, recurrent erosions, epithelial ingrowth, increased frequency of flap microstriae, slow visual recovery, unpredictable refractive endpoint, unstable intermediate refractions, difficult-to-manage visual function during the postoperative period, and the potential for flap "melt"
• long-term issues such as increased frequency of enhancement surgery, recurrent erosions, frequent progression to symptomatic basement membrane disease, refractive instability, and a tendency toward loss of best-corrected visual acuity.

Numerous possible causes for epithelial damage during LASIK have been suggested; most can be avoided or minimized. (See "Epithelial Damage: The Usual Suspects") However, despite seemingly perfect surgical technique and rigorous preoperative vigilance, epithelial trauma remains a significant complication in as many as 5% of LASIK cases.

In search of the elusive cause

In an attempt to eliminate, or at least diminish, the incidence of epithelial defects and "sloughing" in our clinical practice, we minimized exposure to topical medications, removed every conceivable microkeratome-related causal factor, screened thoroughly for pre-existing pathology and treated pre-existing dry eye aggressively. This resulted in a substantial improvement in the health of the epithelium following LASIK. However, the problem persisted.

We were particularly distressed to observe that although epithelial defects only occurred sporadically, in several in-stances they occurred in clusters. We might operate for weeks with no significant epithelial problems. Then, for no apparent reason, 10 to 20% of our cases would require a bandage contact lens for ocular rehabilitation. We found that we couldn't easily explain this in terms of the typical risk factors.

Another observation seemed equally problematic: During these episodes of "bad epithelial days" we could sometimes turn the tide by changing the bottle of proparacaine hydrochloride that we were using. However, in other instances, changing the proparacaine source had no clear effect.

We became determined to find the cause of the problem.

The mystery deepens

The arrival of our LADARVision excimer system from Alcon Summit Autonomous added to the mystery. Prior to this, we didn't introduce anesthetic drops into the patient's eye until seconds before the microkeratome pass. But the LADARVision system allowed us to compensate for cyclorotation by marking the horizontal axis preoperatively. We found it was most efficient to do this several minutes before the patient arrived in the operating room, which meant that we had to instill proparacaine drops several minutes preoperatively instead of several seconds.

In addition to the epithelium being exposed to the proparacaine for a longer period, these eyes also underwent a sequential assault over a 30-minute period as they were subjected to topical tropicamide, Cyclogyl, Neo-Synefrine and/or Mydfrin -- along with their preservatives. We expected to observe more frequent epithelial problems using this regimen.

Once again, our observations surprised us. We sometimes observed a higher incidence of epithelial defects while using proparacaine alone than when we bombarded the ocular surface with a litany of pharmaceutical agents. And this litany included a substantially longer exposure time to the exact same proparacaine preparation!

Testing a hypothesis

Finally, we retreated to a more elaborate but less clinical approach to our dilemma. The manufacturer's product specifications for proparacaine indicated that some variation in pH and/or osmolality could occur. In fact, the pH ranged from 4.0 to 6.0. Keeping in mind that the pH scale is a logarithmic function, this represents a source of substantial variability between individual bottles and manufacturers' lots.

Given this variability, we hypothesized that the clustering of epithelial defect cases was caused by variations in the pH and/or osmolality of the pharmacologic agents. (See "A Theoretical Mechanism for Epithelial Damage") We further hypothesized that the impact of these variations on any given patient would depend upon three additional factors:

• the relative permeability of the epithelial barrier
• the buffering capacity of the ocular tear film
• the susceptibility of the basement membrane to the adverse effects of pH, osmolality or other chemical insult.

The first two factors would most likely be compromised in eyes demonstrating dry eye pathology. The third factor would be most pertinent in eyes with a predisposition to a weak basement membrane complex.

What the data showed

To test our hypothesis, we conducted a study: We monitored the pH of our eye drops and then attempted to correlate those measurements with our clinical observations. This time the correlation was clear: As the acidity of the anesthetic agent increased, so did the frequency of epithelial problems. In the immediate postoperative slit lamp exam:

• the epithelium appeared less healthy
• the cornea was less "wetable," suggesting damage to the microvilli/hydrophilic mucous interface on the corneal surface
• more mucous had become "stringy" debris in the tear film. 

In addition, increased acidity was accompanied by:

• A more severe damage to the basement membrane when epithelial defects were present
• A higher incidence of recurrent erosions and basement membrane disease during the postoperative phase
• A a longer recovery time as the epithelial layer re-established a homeostatic corneal surface
• A increased patient complaints of drops stinging.

When the pH of the anesthetic approached a more neutral, less acidic point, the converse was observed.

We also found that the mydriatic agents were co-contributors to the potential demise of the epithelial matrix. Diagnostic formulations of tropicamide, Cyclogyl, Neo-Synephrine and Mydfrin are all dispensed with pH's ranging from 3.5 to 5.5.

Reaping the rewards

As a result of our findings, we now test and balance the pH of all pharmaceutical agents that we instill into the eye preoperatively, including the mydriatics. Although more data needs to be collected, our experience suggests that a pH of 5.5 ± 0.2 is an ideal target for maximizing drug effect and permeability while still maintaining an "epithelium friendly" ocular environment.

Controlling the pH factor has nearly eradicated epithelium-related complications from our clinical practice -- along with a large measure of frustration experienced by patients and clinical staff alike. Improvements that we've noted include:

• fewer patient complaints that the drops sting
• better "wetability" of the corneal surface both pre- and postoperatively
• improved maintenance of the hydrophilic mucous matrix on the ocular surface in the immediate post-op period
• a substantial decrease in the incidence and severity of epithelial defects and epithelial "sloughing"
• the ability to perform uneventful LASIK on patients with basement membrane disease (we've had no cases of recurrent erosions or basement membrane disease to date)
• a significant reduction of epithelial ingrowth.

In addition, we've found that:

• Topical anesthetics or mydriatics can be instilled in the eye with near impunity. This has freed us from the requirement that eye drops must be instilled minutes or seconds before the surgery begins -- a real asset, particularly if a surgical delay results from laser maintenance or patient-related concerns.
• In those rare cases where epithelial defects do occur, healing time is reduced to hours or days, instead of weeks.
• Perhaps the most unexpected but welcome observation is that we've been able to reduce the wait time for flap adherence to the corneal bed from 2 minutes to 30 seconds. This has markedly improved our clinical efficiency and operating room turnaround time -- and it makes my day go a lot faster.

Despite its success, LASIK can still be improved. Now that the connection between pH and epithelial damage during LASIK is becoming clear, you can largely eliminate another negative variable from surgery by making sure that the drops you instill in patients' eyes are within the ideal pH range. That means better outcomes, fewer complications, happier patients -- and less frustration for you and your staff.

Brian R. Will, M.D., is president and chief executive officer of Will Vision & Laser Centers in Vancouver, Wash. He's certified by both the American Board of Ophthalmology and the American Board of Eye Surgeons, and he's served as a project director for the International Institute for Advanced Laser Surgery.