Limiting stromal dehydration during refractive procedures

April 15, 2013

Faster treatment times will help reduce variable hydration during ablation


In this surgical perspective, Sonia H. Yoo, MD, explains how her patients have much more precise refractive outcomes, and part of that can be attributed to the decreased stromal hydration variability.


By Sonia H. Yoo, MD; Special to Ophthalmology Times

One of the variables during LASIK is the amount of stromal hydration that exists during excimer ablation. Lifting the flap immediately causes the stroma to begin to dry.

If the stroma becomes rapidly dehydrated, the laser may ablate more tissue than intended, causing overcorrection of refractive errors. If the stroma is too wet, there is a chance of undercorrection of refractive errors.

Other potential unexpected refractive outcomes can include astigmatism and central islands if there are fluctuations or transient changes in corneal hydration during the surgery.1,2 While we were getting good outcomes with early-generation lasers, there was room for improvement.

When refractive surgeons shifted from microkeratome-based flaps to femtosecond laser-created flaps, we found more uniformly dry beds. (Microkeratomes tended to leave beds more hydrated.) The femtosecond-created flaps were more uniform and this led to tighter surgical nomograms, which, in turn, led to fewer unintended refractive outcomes. In today’s demanding society, unexpected outcomes or results that patients perceive as less than perfect necessitate re-treatments.

The good news is the latest generation of femtosecond lasers with their flying spots and rapid treatment times are much faster than earlier versions. As a result, patients’ stroma is exposed to conditions that can cause transient changes in hydration for a shorter period. In my hands, patients have much more precise refractive outcomes, and part of that can be attributed to the decreased stromal hydration variability.

Relative hydration

A patient’s relative corneal hydration will affect the “laser take”-the ability of the laser’s effect on the corneal tissue. For example, one laser spot used to treat the corneal stroma will have a greater effect on a very dry cornea than that same laser spot would have on a wet cornea. What we’ve found in general is that microkeratomes tended to leave relatively wet beds and femtosecond lasers leave drier beds.

It helps to think of the cornea as a sponge. Once the epithelium and Bowman’s layer are lifted, we’ve raised the flap and have just bare stroma exposed to the elements of the laser suite. Wetting the cornea will make it act like a sponge so even the smallest amount of introduced hydration will quickly spread and affect the treatment outcomes. There is also the potential for random hydration when the surgeon introduces liquid to the surface. Portions of the stromal bed might be naturally hydrated while others might be dry; when a surgeon introduces liquid to the surface to alleviate hydration concerns the liquid could potentially spread unevenly. That could potentially be worse than if the cornea were uniformly hydrated or dry.

I’ve had cases where saline or surface tears can get onto the stroma, creating a non-uniform hydration. When that happens, I will hydrate the entire stromal bed to ensure consistent hydration. As surgeons, we should strive to have the driest surfaces, but an all-wet surface will still yield better results than one that is half-wet/half-dry.

In my opinion, every surgeon should be paying more attention to the state of the cornea’s hydration throughout the treatment. When (or if) surgeons wipe the surface, the ability to detect areas of hydration versus areas of dryness is not as easy.

Going back to the sponge analogy, pouring water directly on top of a sponge makes it easy to see where it’s been hydrated. But if you then take a paper towel and wipe the surface of that sponge, it will be difficult to determine what’s wet and what is not. I can’t stress enough how these hydration issues might affect outcomes with over- or undercorrections.

Current outcome measures

Surgeons still use Snellen visual acuity as a primary endpoint, and while a patient may be 20/20, he or she may complain of poor quality of vision. For instance, in the office the patient may read 20/20, but in the real world, the patient may experience fluctuating vision, or have difficulty with night driving. At this time, we don’t have great metrics to measure quality of vision. Some of the newer technologies to measure contrast sensitivity and light scatter may be able to address some of that in the future.

In my opinion, Snellen vision is a bit of a crude way to evaluate the kinds of refractive changes we might be getting with the better lasers. The newer, faster laser platforms that are giving surgeons more uniform ablations and helping to minimize variable stromal bed hydration may yield improvements in quality of vision that may not be measureable with Snellen testing alone.

Until we have a metric that will allow ophthalmologists to uniformly measure quality of vision, however, my advice is to monitor what we can-namely, nomograms, initial refractive settings, and the patient’s stromal hydration during surgery. Doing that will yield fewer re-treatments and happier patients.


1.         Fisher BT, Masiello KA, Goldstein MH, Hahn DW. Assessment of transient changes in corneal hydration using confocal Raman spectroscopy. Cornea. May 2003;22:363-370.

2.         Wilkinson PS, Davis EA, Hardten DR. LASIK. In: Yanoff M, Duker JS, eds. Ophthalmology. 2nd ed. Philadelphia: Mosby Elsevier; 2009:145-158.


Sonia H. Yoo, MD, is professor of ophthalmology at Bascom Palmer Eye Institute, University of Miami Miller School of Medicine. Dr. Yoo may be reached at 305/326-6322; She did not indicate a financial interest in the subject matter.