The Endothelium and IOL Implantation

The recent FDA approval of several phakic refractive IOLs provides an excellent backdrop for a discussion of the corneal endothelium and IOL implantation. It may be startling to realize that an accurate understanding of the endothelium—including its clinical limitations, frailty, and resilience as well as its tolerance of IOL implantation—only dates to the late 1960s to early 1970s. Those pioneering surgeons who defied conventional wisdom to pursue the dream of pseudophakia deserve a great deal of credit. Certainly, the status of phakic IOLs today is analogous to that of pseudophakic IOLs in the 1970s, except that the time required for the development and acceptance of the former will undoubtedly be far shorter.

MY BACKGROUND
The information and opinions contained in this article are based upon my observations and experience during the past 30 years as an active IOL and corneal surgeon. As a resident in 1975, I gained a great deal of experience with ACIOLs that had both rigid and flexible haptics (closed and open), capsular-fixated IOLs that cause iris-sphincter erosion, and PCIOLs made of silicone or PMMA that had haptics named after innovators such as Steven Shearing, MD, and Robert Sinskey, MD. I performed hundreds of penetrating keratoplasties on eyes with all of these IOL types and bullous keratopathy—procedures that usually entailed replacing one IOL with another. Much has occurred in the anterior chamber during the past 25 years.

ANATOMY AND FUNCTION
The corneal endothelium is a monolayer of cells derived from the neural crest that maintains corneal deturgesence by means of an osmotic gradient created through the action of a sodium/potassium pump. The endothelial cells do not divide but do increase in size in order to cover areas of injury or cell loss. Endothelial cells number about 2,800 per square millimeter in youth, but this amount declines with age to approximately 2,000 at 70 years. In addition, there is no good correlation between endothelial function and endothelial cell number, and a nonhexagonal shape typically indicates some endothelial cell distress.

As a result of sampling error, an endothelial cell count is commonly considered to be accurate to approximately ±5%. For this reason, using a variance of less than 5% to describe a trend creates misinformation. Conclusions reached based upon cell counts of 2,200 ±60 per square millimeter, for example, are invalid. I have seen clear, thin corneas with cell counts of 600 cells per square millimeter and edematous corneas with cell counts of 1,200 cells per square millimeter. Serial pachymetry performed at the same location on the cornea is probably a better measure of endothelial cell function than an absolute endothelial cell count, because the latter does not necessarily correlate well with endothelial function.

Factors causing endothelial cell loss include an individual's genetic predisposition (eg, endothelial dystrophy), direct contact with an IOL, excessive irrigation at the time of surgery, and severe postoperative inflammation. During phacoemulsification, I believe that it is the location and duration of irrigation that causes endothelial cell loss, whereas ultrasound, which is dampened by the aqueous, has no significant effect on the endothelium. Through the years, I have performed and witnessed many phaco cases with 4 to 6 minutes of ultrasound time that have resulted in clear corneas the next day. Of course, direct contact with the endothelium by the phaco tip or any other object, including an IOL, can cause endothelial cell loss.

SURPRISE
Which factor has the greatest effect upon endothelial cell loss during cataract extraction and/or IOL implantation? Assuming excellent surgical technique during cataract extraction, IOL implantation, and the removal of viscoelastic from the anterior chamber, the loss of endothelial cells is determined primarily by the length of the corneal incision.

In 1982, I heard a lecture by Thomas Neuhann, MD, the well-respected ophthalmic surgeon from Munich, Germany. He referred to a comparative study of intracapsular cataract extraction, extracapsular cataract extraction, and phacoemulsification. The investigators found that, the longer the corneal incision was, the greater the amount of cell loss was and the more the residual cells had to increase in surface area and migrate in order to repair the area of injury. Conversely, the endothelial cells on the opposite side of the cornea from the incision site were only minimally affected.

As stated earlier, endothelial cell counts are useful only if the same site is measured each time. If the measurement of the endothelial cell count prior to IOL implantation is performed slightly above the central cornea and the postoperative count is taken below the central cornea (assuming a standard wound at the 12-o'clock position), then, statistically, it is rather easy to show erroneously that removing a cataract and implanting an IOL cause an increase in endothelial cells! Using the center of the miotic pupil as a landmark might be a reasonable approach for improving the consistency of the location for endothelial cell counts—a critical parameter for accurate studies.

If the effects of wound length as well as of traumatic cataract removal and IOL implantation are removed from the equation, it becomes clear that the rate of endothelial cell loss is not affected significantly by a well-designed, well-placed ACIOL, iris-fixated IOL, or phakic/pseudophakic PCIOL. The long-term loss of endothelial cells after surgery will then most closely parallel the normal aging process.

Overall, the literature reports 10% cell loss after phacoemulsification and IOL implantation with a 3-mm limbal incision and no significant, long-term progression.1-5

Phakic IOLs
The material covered thus far is important with respect to phakic IOLs. A small incision is desirable with regard to both the endothelium and to astigmatism, but a well-designed phakic IOL must be well tolerated for life by the corneal endothelium. The corneal endothelium used in corneal transplants routinely functions for at least 100 years (the life span of the donor plus the time the cornea functions in the recipient), despite the trauma of corneal transplant surgery. Interestingly, the success of corneal transplantation is not correlated to the age of the corneal donor.

CONCLUSION
With respect to the corneal endothelium, the goal of a well-designed phakic IOL should be to cause no long-term problems. Creating small corneal incisions and minimizing trauma during IOL implantation surgery are clearly beneficial to the endothelium.

A test using a vital dye that can demonstrate endothelial cell function and reserve is sorely needed. In my experience, endothelial cell counts have little predictive usefulness, similar to how healthy eyes may score low and symptomatic eyes may appear normal on a Schirmer test. The fundamental finding relating to endothelial function is whether the cornea is clear and thin or cloudy and edematous. Until improved testing of endothelial function is available, the best indicators of corneal function will continue to be (1) serial endothelial counts and serial pachymetry performed at the same location on the cornea and (2) a careful slit-lamp examination by an astute observer.

Lee T. Nordan, MD, is a technology consultant for Vision Membrane Technologies, Inc., in Carlsbad, California. Dr. Nordan may be reached at (760) 431-1846; laserltn@aol.com.
1. Edelhauser HG, Sanders DR, Azar R, Lamielle H; ICL in Treatment of Myopia Study Group. Corneal endothelial assessment after ICL implantation. J Cataract Refract Surg. 2004;30:576-583.
2. Bourne RR, Manassian DC, Dart JK, et al. Effect of cataract surgery on the corneal endothelium: modern phacoemulsification compared with extracapsular cataract surgery. Ophthalmology. 2004;111:679-685.
3. Beltrame G, Salvetat ML, Driussi G, Chizzolini M. Effect of incision size and site on corneal endothelial changes in cataract surgery. J Cataract Refract Surg. 2002;28:118-125.
4. Lai LJ, Chen YF, Wu S, et al. Endothelial cell loss induced by phacoemulsification occurs through apoptosis. Chang Gung Med J. 2001;24:621-627.
5. Lesiewska-Junk H, Kaluzny J, Malukiewicz-Wisniewska G. Long-term evaluation of endothelial cell loss after phacoemulsification. Eur J Ophthalmol. 2002;12:30-33.