Eyeworld

EW CATARACT 27 April 2015 by Michelle Dalton EyeWorld Contributing Writer specific formulas," Dr. Olsen said. "Taking the refraction itself can be subject to error. The question of bias also applies to the 'exact' field of clinical optics." He recom- mends aiming for ±0.25 D error when taking manifest refractions as most modern formulas are "well below 0.5 D error on average," he said. 7.The formula should have an ac- curate, unbiased ELP prediction. Accurate ELP predictions are considered "the most important" source of potential error. "The goal of the 'optimization' procedure may be to find the average ELP in a sample of cases," he said. "But if there is a sufficiently large sample size, this analysis may also be extended to study the influence of different predictors, such as anterior chamber depth, lens thickness, corneal diameter, gender, age, and refraction, among others," he said. EW References 1. Olsen T. Use of fellow eye data in the calcu- lation of intraocular lens power for the second eye. Ophthalmology. 2011;118(9):1710–1715. 2. Olsen T, Funding M. Ray-tracing analysis of intraocular lens power in situ. J Cataract Refract Surg. 2012;38(4):641–647. Editors' note: Dr. Olsen has financial interests with IOL Innovations and Haag-Streit (Koniz, Switzerland). Contact information Olsen: tkolsen@dadlnet.dk errors tend to produce a hyperopic surprise in post-refractive eyes. 4. The formula should allow the input of "true, physical data," Dr. Olsen said. Presuming a surgeon knows the exact curvatures of the front and the back surface of the cornea, "the formula should be able to translate this information into the true corneal power (and corneal cylinder)," he said. Once surgeons know the exact position of the implant, "the formula should give an accurate answer on the refraction to be expected with that eye. The latter option can be used to improve the prediction of the fellow eye, 1 and it can be reversed to back-calculate the power of the implant given the actual refractive outcome 2 ," he said. 5. The formula should be accurate when calculating toric cylinder. "I believe many of the so-called 'toric calculators' out there are repeating the errors of the old IOL power formula by assuming that the cylinder of the anterior surface of the cornea translates into a constant cylinder of the toric im- plant," Dr. Olsen said, adding that science has now shown "more variables should be taken into account, such as the posterior cor- neal astigmatism and the position of the IOL in the eye." 6. Formulas should be evaluated using optimum conditions. "What that means is using the correct IOL constants with the 2. The formula should be accurate in long eyes. "The eye may be larger than normal in many ways—in all di- mensions, or only in the posterior segment, which may be the case in excessive myopia," he said. The long eye is "interesting because the corneal power becomes the dominant source of error." For ex- ample, if a very long eye requires an IOL power close to zero, "it does not matter where the IOL is located within the eye so if the formula produces an error, it is due to an error in the assumed corneal power or in the assumed axial length," Dr. Olsen said. 3. The formula should be accurate in post-refractive eyes, some- thing that has remained a challenge for surgeons. "The good news is that the error of the conventional thin lens formulas has been identified as 2-fold: 1) A 'K-reading' based on the anterior corneal curvature, which does not give a representa- tive measure of the true corneal power when the normal relation- ship between the anterior and the posterior curvature of the cornea has been changed as a result of the LASIK procedure; 2) When the formula uses the 'K-reading' as a predictor of the ELP (most con- ventional thin lens formulas do), this formula will make an error in an eye with a broken anterior segment anatomy like the post- LASIK eye," Dr. Olsen said. Both Advice from an expert M ost anterior segment surgeons would agree that IOL calculations are far from perfect. EyeWorld spoke to Thomas Olsen, MD, University Eye Clinic, Aarhus, Denmark, author of the Olsen formula, and the origina- tor of the PhacoOptics PC software (IOL Innovations, Aarhus, Denmark) using the newly described C-con- stant approach for the prediction of the IOL position. EyeWorld asked him what went into developing his fourth gener- ation formula; others include the Holladay II (and its related software, Holladay Consulting, Bellaire, Texas) and Preussner formulas. Dr. Olsen said there are "7 main considerations" that must be present when considering an IOL calculator: 1. The formula should be accurate in short eyes. "A short eye is a very difficult eye for the simple fact that a small error in any of the measurements has a higher impact on the refrac- tive outcome than in a long eye," he said. Additionally, any errors in effective lens position (ELP) will have "a dramatic influence on the outcome. As the higher powered IOL thickens, the differ- ence between the first and second principal plane widens, and this will itself modify the effective lens plane," he explained. 7 key parameters in IOL calculations Off-centered continued from page 24 style is recommended. Meanwhile, surgeons who use phaco prechop or femtosecond surgery may want to consider the straight or Akahoshi style. "Initially, these off-centered tips were developed for torsional phaco by OZil," Dr. Akahoshi said. "How- ever, I found these tips are very effective for only longitudinal phaco by phaco machines from other companies such as Bausch + Lomb [Bridgewater, N.J.] and Abbott Medi- cal Optics [Abbott Park, Ill.]." Simply by changing the tip he has found that it makes for a safer and more efficient procedure. After testing tips with the Centurion Vision System (Alcon), Dr. Akahoshi has found the Harmonyx to be safer and more efficient than the Alcon original balanced tip. "I believe the square off-cen- tered tip is the most efficient and safest phaco tip in the world," he said. "In the future when femtosec- ond laser cataract surgery becomes more popular, the Akahoshi-style square off-centered Harmonyx should be the tip of choice." Dr. Akahoshi discussed off-cen- tered phaco tip shapes in a presenta- tion at the 2014 ASCRS•ASOA Sym- posium & Congress in Boston and why these could be more effective than the conventional phaco tip. By making the phaco tip square, he said cutting efficiency can be maximized. Mikio Inamura, MD, Yo- kohama, Japan, is also using the Harmonyx tip. The square angle of this tip provides destructive power against a dense cataract, he said. "[The] Harmonyx tip has less damage to the wound, especially to corneal endothelial cells," he said. But it still provides strong power to deal with dense cataracts. Dr. Inamura said the Harmonyx tip is easy to handle and may be more efficient in aspiration than the Kelman tip. Generally, Dr. Inamura divides the nucleus into 4 to 8 pieces before phaco when using the Harmonyx tip, and he prefers the prechop technique, although with dense cataracts, the phaco chop technique may be the way to go. For about 4 years, he has been using the Harmonyx tip and has found that this method has helped to reduce CDE values during his surgery and provided for safe and effective phaco. EW Editors' note: Drs. Akahoshi and Inamura have no financial interests related to their comments. Contact information Akahoshi: eye@phaco.jp Inamura: inamura1771@gmail.com

• Cover