Issues in Custom LASIK: Pupil Dilation
Technology experts make the case for either the natural or pharmacologic approach.

You can't treat what you don't measure
BY RONALD KRUEGER, M.D.

It's a well-recognized fact that aberration level in the eye is dependent on pupil size. If you measure a patient's wavefront when his pupil is open wide, you see substantially more aberrations than if you measure his wavefront when his pupil is constricted. Because much more deviation from the norm exists in the periphery of the visual system, the larger the pupil, the greater the higher-order aberrations.

The way to address this issue is to pharmacologically dilate the pupil. If a patient's pupil dilates to 6 mm in darkened conditions and you capture wavefront data across that 6 mm, approximately one-half mm of that information may be compromised due to clipping of data in the peripheral lenslets of the Shack-Hartmann array, so you may have less than 6 mm of reliable data. But if you pharmacologically dilate to 7 mm, as is done with the Alcon LADARVision System, you will gather at least 6.5 mm of reliable wavefront data, allowing a true custom treatment of aberrations within the 6.5-mm optical zone of a CustomCornea treatment.

Why We Shouldn't Estimate

Some custom refractive laser platforms don't require pharmacologic pupil dilation for wavefront measurement. They rely on measurements of dark-adapted pupils and they estimate peripheral data according to a Munnerlyn formula. A recent study by Roger Steinert, M.D., illustrates the problem with this approach.

On average, the calculated higher-order aberrations were 62% higher than those measured (p<0.005) despite there not being much difference between the measured and calculated values for spherical equivalent refractive error. Almost 90% of the eyes exhibited a difference of less than +/- 0.25D. And in terms of higher-order aberrations, the differences ranged from 40% lower to 300% higher than the measured values. This indicates that not only are calculated values different than measured values, but that variability from patient to patient is high, making treatment of calculated aberrations risky. Furthermore, because the periphery is of greatest interest when it comes to reducing problems with post-op night vision due to the prevalence of peripheral aberrations, it's precisely the area where we don't want to rely on estimated data.

Dilation Does Not Affect Aberrations

Another point of debate surrounding the issue of pupil dilation is whether artificially dilating affects the wavefront aberrations we're measuring. According to the results of a study presented at the 2002 ARVO meeting, the answer is no. Ten patients (20 eyes) were enrolled in the study. Their mean age was 33. Eight eyes had previously undergone refractive surgery; six eyes were myopic up to -6.50D sphere and -1D cylinder; and six eyes were hyperopic up to +0.75D of sphere and -0.50D cylinder. For each eye, Kevin Liedel, et al. measured the pupil center/limbus geometry, dark-adapted the pupils, and obtained wavefront measurements using the LADARWave aberrometer. They then dilated the eyes with 1% Tropicamide, waited 30 minutes, and obtained a second round of measurements. Seventeen of the 20 pupils dilated naturally to 5 mm or larger, and for those eyes the researchers compared 5-mm diameter wavefront measurements for naturally dilated versus drug-dilated pupils.

Only the defocus term showed a nonzero difference between the two measurement conditions, and the difference never exceeded 0.33D. No difference appeared in vertical coma or spherical aberration. The researchers therefore concluded that pharmacologic dilation has an insignificant impact on wavefront aberrations, particularly the higher-order terms.

Another aim of the study was to assess the impact of shifts in the pupil center (from undilated to naturally dilated to drug-dilated) on the wavefront exam. Wavefronts were reconstructed from Hartmann-Shack data using either the daytime pupil center or the dilated pupil center as the origin for Zernike polynomial description. The researchers found no significant differences between the two methods.

Furthermore, with the CustomCornea platform, prior to pupil dilation, the wavefront device is used to image the geometric relationship between the daytime pupil and the limbus. This information is then used to center the wavefront measurement and subsequent treatment on the natural line of sight. Therefore, dilation has no adverse effect on centration when using the LADARVision System.

What's Best for Patients

Based on what I've explained here, it's clear to me that in order to acquire sufficient wavefront data, we need to pharmacologically dilate the pupil. Otherwise, we may not be capturing the most wavefront data possible that we need to provide our patients with the best possible custom refractive surgery procedure, one that is truly customized.

Dr. Krueger is medical director of the department of refractive surgery at The Cleveland Clinic Foundation Cole Eye Institute.  He receives travel and research support from Alcon.

Natural dilation is both sufficient and necessary
BY MARC ODRICH, M.D.

Wavefront measurement and treatment require dilation of the patient's pupil. The question is whether it should be dilated physiologically or pharmacologically. Unlike other wavefront ablation platforms, the VISX platform does not require pharmacological dilation. It requires capture of at least 5 mm worth of wavefront data, and in most cases captures data out to 6 mm, depending on the amount of the patient's physiological pupil dilation.

Wavefront data capture out to 7 mm will be available early this year with the next WaveScan software upgrade.

There are two main reasons that physiological pupil dilation is preferred over pharmacological dilation. First, and most important, is that pharmacological dilation can change both the lower- and higher-order aberrations of the eye. Every dilating agent has at least a small effect on the cycloplegic state. Ciliary muscle tone is compromised, which changes the shape and the measurement of the actual aberrations. It has also been made clear in the past by Dr. Borish and others that pharmacologic pupil dilation changes cylinder.

I recently conducted a simple study to illustrate this concept. I first obtained manifest refractions on a group of 31 patients. I then obtained WaveScan measurements (3.07 software) with only physiological dilation. Finally, I pharmacologically dilated these patients with Neosynephrine 2.5% for 20 minutes and captured wavefront measurements on the same eye of each patient. All WaveScan measurements included 6 mm of captured data. The range of myopia was -0.50D to -6D in these eyes.

Data from this study revealed that the lower- and higher-order aberration levels decreased in these eyes between the physiologically and pharmacologically dilated states. Outcomes from this study are reported here:

Dilation Shifts the Pupil Center

The second reason why dilating with pharmacologic agents is not ideal is that they have uneven interactions with the iris sphincter, which means they can cause some degree of shifting of the pupil center. It is crucial that the pupil center of the wavefront measurement match the pupil center of the laser treatment. Any mismatch will displace the treatment, causing unintended results.

Several studies illustrate this. One was performed by Coorpender, Klyce and McDonald and published in the May 1999 issue of the Journal of Cataract & Refractice Surgery. They reported that nasal and superior displacement of the pupil centroid occur with pharmacologic dilation. One-third of the eyes they studied displayed a centroid shift of at least 0.5 mm off-center compared with the undilated state. This can often result in a decentered treatment.

In addition to those two key points, note that patients have difficulty fixating on a point source of light under the laser while their pupils are dilated, which could lead to an off-angled ablation. It has also been shown that cyclorotational movement increases in dilated eyes.

Here is what I consider to be the bottom line: Pupil size has absolutely no relationship to patient satisfaction or other outcomes metrics with VISX wavefront ablations. In the VISX wavefront ablation clinical trial, 59 eyes had pupils that dilated to 7.5 mm and beyond, some to as large as 9.5 mm. They were all treated with 6-mm optical zones blended out to 8 mm. They were among our happiest patients, as the accompanying slides show. OM

Dr. Odrich is an assistant professor of ophthalmology at Columbia University in New York and the medical director for VISX.

VISX FDA Study