Eyeworld Daily News

EW SHOW DAILY 34 ASCRS Symposia Monday, May 8, 2017 He recommended doing a personal SIA, with the use of centroid values, or using an SIA value of 0.12 D. Douglas Koch, MD, Houston, divides preoperative measurements into two categories: Scheimpflug and OCT swept source devices. The devices are back calibrated to be able to use IOL constants, and the limit of agreement among devices is 95%. He explained that the Pentacam (Oculus, Wetzlar, Germany), Galilei G6 (Ziemer Ophthalmic Systems, Port, Switzerland) and IOLMaster (Carl Zeiss Meditec, Jena, Germa- ny), OA-2000 (Tomey, Erlangen, Germany), and IOL Master700 and LENSTAR (Haag-Streit, Mason, Ohio) all facilitate the analysis of raw data and help measure through dense cataracts. In macular disease, however, the IOLMaster 700 cannot be used to replace OCT. In complex corneas, measuring the back surface of the cornea is essential, he said, and topography for total corneal power is important. Biometry is no longer a consid- erable source of error, said Thomas Olsen, MD, Aarhus, Denmark, but keratometry requires high precision to be any good. Measuring corneal power with the Zemax (Seattle) gives an exact ray tracing analysis of the cornea. It also analyzes the corne- al-IOL system, without the bias of axial length or IOL power. Warren Hill, MD, Mesa, Arizona, closed the session by predicting a bright future based on his ongoing investigations with neural networks, whose high accuracy power calculations with 12,000 patients will be available later this year. EW Editors' note: Drs. Abulafia, Barrett, Hill, Koch, and Olsen have related financial interests. Drs. Holladay and Weikert have no financial interests related to their comments. by Stefanie Petrou Binder, MD, EyeWorld Contributing Writer Problems with true K and central K come into the mix, said Graham Barrett, MD, Perth, Aus- tralia, who thinks that there really is no "best method." However, the refractive history of the patient is useful for power calculations with improved formulae, and wavefront aberrometry and OCT-based systems are promising for the future and continue to be refined. Corneal astigmatism, calcu- lation methods, and surgically induced astigmatism (SIA) are addi- tional sources of error in toric IOL calculations, said Adi Abulafia, MD, Jerusalem, Israel. When determining corneal astigmatism power, it is important to validate the steep meridian, as well as the power difference between meridians. Standard keratomet- ric and topography machines are inaccurate for this task. Solutions include mathematical models based on anterior corneal measurements and direct measurement of the pos- terior corneal curvature, which he thinks should be the gold standard. in specificity, including more and more factors to reduce prediction error. Dr. Holladay's own prediction is that surgeons will be able to im- prove outcomes to within a standard deviation of ±0.3 D from the current norm value of ±0.439 D. Trying to improve IOL power prediction comes at a time when patients have high expectations, and want pain-free, complication-free surgery with instant visual recovery, minimal downtime, all at the lowest possible cost, according to Mitchell Weikert, MD, Houston. Although the literature shows that overall, sur- geons achieve 70% of refractions to within 0.5 D and 93% to within 1.0 D, today's well-read patient expects more. Dr. Weikert explained that us- ing the right formula is a good place to start. Formulas were less reliable with extremes in axial length, ne- cessitating different approaches for short and long eyes. He suggested the Holladay 1, Hoffer Q, and SRK/T for short eyes and more significant- ly, the Holladay 1 and Haigis for adjusted axial length for long eyes. N ew approaches to IOL power calculations can help surgeons achieve better refractive out- comes, according to an expert panel. The problem begins with differences between actual and predicted refraction, according to Jack Holladay, MD, Houston, who kicked off the session. The predicted error is a combination of 16 dif- ferent variables, 95% of which can be explained by the effective lens position, refraction, and the axial length. Formulae used to determine IOL power include the thin IOL vergence formula that includes effective lens position, corneal power (K), axial length, and the A-constant; thick IOL vergence formula that in addition takes the shape of the lens into account; ray tracing that accounts for all surfaces (although we technically do not know the optics of the entire system); and neural network that computes all possible combinations for effective lens position. These systems increase IOL power calculations: State-of-the-art Drs. Barrett, Holladay, Packard, and Koch discuss state-of-the-art IOL power calculation strategies.

• Cover