Eyeworld

EW CATARACT 48 September 2015 by Erin L. Boyle EyeWorld Contributing Writer a normal eye will introduce the new errors of radius error, keratometer index error, and IOL formula error. The radius error can happen following refractive laser surgery for myopia, he said, if the keratometer finds the radius of the corneal curva- ture too far off the optical axis. The keratometer index error can happen because the "ratio of anterior to posterior corneal radius on which the keratometer calibra- tion is based is altered by refractive surgery." "The IOL formula error is linked to the usage of corneal power as a predictor for the ELP: The new K does not represent the internal geometry of the eye any longer, thus producing an incorrect ELP," Dr. Haigis said. Other factors According to Dr. Hoffer, based on results today, axial length and preop- erative ACD are probably the best parameters to use as predictors for the axial length and K reading, we'll get two benefits. We'll get a better prediction of the ELP and we'll get a better outcome in terms of our pre- diction error," Dr. Holladay said. He said the benchmark today is 75% within plus or minus half a diopter, although some very com- pulsive surgeons achieve 85% and intraoperative aberrometry can reach 95%. Reasons for error Dr. Haigis explained the reasons for error when calculating IOL power: "Neglecting empirical formulas like the SRK II, which today—for a multitude of reasons—should not be used, IOL calculation essentially de- pends on the variables axial length, corneal power, and effective lens position," he said. "These variables are loaded with measurement or pre- diction errors, which combine in a Gaussian error propagation to create an overall error." He said that treating an eye fol- lowing corneal refractive surgery as but one that would "call for interdis- ciplinary cooperation." Jack T. Holladay, MD, MSEE, FACS, clinical professor of ophthal- mology, Baylor College of Medicine, Houston, said that ELP matters most in lensectomy cases. He said about a third of the error is from the posi- tion of the lens, another third is the result of the refraction, and the last third is the result of both the axial length and the corneal (K) reading, with the axial length the most im- portant of the two. Newer formulas like the Holladay 2, Olsen 2, and the Barrett 2 use up to 7 variables to determine ELP: axial length, K, ACD, lens thickness, horizontal white-to- white, age, and pre-cataract refrac- tion. "The more we know about the anatomy of the eye, the better we can predict the ELP," he said. "It turns out that the biggest reasons for our prediction errors are primarily related to the axial length, the K reading and the refraction that we do postoperatively. Knowing that, it means that if we improve Lens position is an important factor for postop refractive success O ne of the most common sources of error in calcu- lating IOL power remains the effective lens position (ELP), which can have a major impact on IOL refractive pow- er accuracy following implantation. "ELP prediction is still the holy grail of IOL calculation. One of the goals today is to come up with algo- rithms that allow a smaller scatter in the prediction errors, thus increasing the number of patients with only small deviations from target refrac- tion," said Wolfgang Haigis, PhD, professor, University Eye Hospital, Wurzburg, Germany. Currently, the parameters that can be accurately measured in normal eyes are the axial length by optical biometry or immersion A-scan, the corneal power by ker- atometry, and the anterior chamber depth (ACD) by optical pachymetry or dual Scheimpflug photography, said Kenneth J. Hoffer, MD, clinical professor of ophthalmology, UCLA. "The biggest problem in normal unoperated eyes is the determina- tion of the exact healed postoper- ative axial position of the IOL (or ELP)," Dr. Hoffer said. "With eyes that have undergone corneal refrac- tive surgery, the biggest problem is determining the exact power of the cornea, which is also compounded by the ELP." Dr. Haigis, author of the Haigis formula, said current measurement instrumentation and algorithms al- low good estimation of the ELP, but there are still variations. "There is a physiological 'base noise' producing high standard deviations, which cause refractive outcomes to follow a broader dis- tribution than intended," he said. "There are indications that femto laser technology can reduce this 'noise' by allowing a precise and reproducible rhexis. It may also be speculated that individual wound healing processes may influence the final ELP." Dr. Hoffer added that more recent studies do not show a benefit of femto laser in this regard. He said the solution to this issue would not be physiological optics Effective lens position prediction still sought The ELP is the equivalent lens with no thickness and is shown for an IOL with an A-constant of 118.5, which is equivalent to being 5.25 mm posterior to the corneal vertex. Source: Jack T. Holladay, MD, MSEE, FACS

• Cover