Highly accurate IOL power calculations result from optimizing a collection of interconnected nuances. The keratometry technique, method of axial length measurement, IOL power calculation formula, optimized lens constant, and configuration of the capsulorhexis—all individually influence the final refractive outcome. For this reason, focusing on a single item such as the axial length measurement or the IOL power calculation formula is usually insufficient to ensure consistent accuracy over a wide anatomical range. The surgeon must consider the process as a whole while simultaneously optimizing each component.
KERATOMETRY
Ophthalmologists and their technicians often accept without question corneal power measurements by keratometry or simulated keratometry, but not all measurements have the same level of accuracy or reproducibility. It should be remembered that keratometry errors have a 1:1 correlation with postoperative refractive errors at the spectacle plane. For example, if the keratometry reading is off by 0.50 D, the result will be a 0.50-D postoperative refractive error at the spectacle plane, even if all other aspects of the IOL power calculation and surgery are perfect. Add in other small errors such as variable corneal compression induced by applanation A-scan biometry or the use of an older 2-variable formula in axial hyperopia, and a 1.00-D deviation from the target refraction is not difficult to imagine.
To maximize keratometry accuracy, first, make the decision to use a single instrument for all pre- and postoperative measurements in order to limit the number of variables. For manual measurements, switching to a Javal-Schiotz–style keratometer will to help improve accuracy. Autokeratometry is quick and easy, but it typically requires multiple measurements to confirm accuracy. The simulated keratometry feature of many topographers is an excellent way to objectively determine the axis of astigmatism, but it can sometimes be less accurate than careful manual keratometry for measuring the central corneal power.
Second, regularly check your keratometer against a set of standard calibration spheres and consider keeping a logbook of these evaluations (Figure 1). Third, if the results for any patient vary, ask a second staff member to confirm the measurements to ensure accuracy. Finally, if the keratometry mires are unreliable or distorted, obtaining a topographic axial map may help uncover something unsuspected such as a forme fruste of keratoconus.
AXIAL LENGTH MEASUREMENTS
One of the most common reasons for an incorrect IOL power is an error in the axial length measurement. The familiar and trusted 10-MHz applanation A-scan biometry is probably no longer accurate enough to consistently satisfy contemporary patients' expectations. The reason is that measurements by the applanation technique produce a falsely short axial length and sometimes widely different results due to varying degrees of corneal compression and axial alignment.
Immersion A-scan biometry is unquestionably a more reliable method. This technique causes no corneal compression, and, when used in conjunction with a Prager shell, measurements can be of very high quality and quite reproducible. Even in the hands of the most skilled biometrist, however, immersion A-scan biometry is still limited by the fact that it is based on the resolution of a 10-MHz sound wave (Figure 2).
At present, optical coherence biometry using the IOLMaster (Carl Zeiss Meditec AG, Jena, Germany) is unquestionably the most accurate way to measure axial length prior to cataract surgery. Optical coherence biometry's use of a short-wavelength light source (instead of a longer-wavelength sound beam) increases axial length measurement accuracy by fivefold when compared with ultrasound (Figure 3).1
For challenging axial length measurements (eg, in eyes containing silicone oil, extremely short nanophthalmic eyes, or extremely long myopic eyes with posterior staphylomata), the accuracy of optical coherence biometry is unparalleled. The one disadvantage of the technique is that it is an optical method. Axial opacities such as a corneal scar, dense posterior subcapsular plaque, or vitreous hemorrhage may decrease the signal-to-noise ratio to the point that reliable measurements are not possible. In the typical North American ophthalmology practice, optical coherence biometry is unable to measure between 5% and 15% of patients, and immersion ultrasound is required.
IOL POWER CALCULATION FORMULASLimitations
The main limitation of all IOL power calculation formulas pertains to their ability to accurately predict preoperatively where the IOL will be located postoperatively in relation to the cornea. As described by Jack Holladay, MD, of Bellaire, Texas, this distance from the secondary principal plane of the cornea to the thin lens equivalent of the IOL is known as the effective lens position (ELPo).2
Commonly used 2-variable formulas such as SRK/T predict the IOL's postoperative position based on the eye's axial length and keratometry readings. To produce this prediction, these formulas must make a number of broad assumptions. In general, most 2-variable formulas assume that short eyes produce a shallower ELPo and longer eyes will result in a deeper ELPo. They also assume that flat K readings will result in a more shallow ELPo and steeper Ks will result in a deeper ELPo. The anterior and posterior segments of the human eye are often not proportional, however.3 This is the main reason why the accuracy of 2-variable formulas decreases at the extremes of axial length and corneal power, especially in the setting of axial hyperopia.2
As long as an eye has parameters close to those of a schematic eye, 2-variable formulas work very well. For example, every modern 2-variable formula will predict essentially the same IOL power for an eye with an axial length of 23.49 mm and K readings of 43.50 D. Repeat the exercise with an axial length of 21.00 mm, however, and their IOL power recommendations quickly diverge. Formulas that base their calculations on more information than axial length and keratometry have an obvious advantage over those that do not.
Best Bets
Currently, the Holladay 2 formula (part of the Holladay IOL Consultant; Holladay Consulting, Inc., Bellaire, TX) is the best “off-the-shelf” tool for improving the accuracy of IOL power calculations for all axial lengths. The Holladay 2 formula employs several additional variables to adjust the recommended IOL power; these include the horizontal corneal diameter, lens thickness, measured anterior chamber depth, and the patient's age and preoperative refraction.
The Haigis formula also represents a significant improvement over popular 2-variable formulas. It uses three IOL and surgeon-specific variables (a0, a1, and a2) in order to set both the position and the shape of an IOL power prediction curve. At present, Haigis constants for many popular IOLs are being developed from data submitted by physicians worldwide (visit http://www.augenklinik.uni-wuerzburg.de/eulib/index.htm). The Haigis formula is included as part of the IOLMaster's standard software package.
IOL CONSTANT OPTIMIZATION
Surgeons must personalize the lens constant (Holladay 1 Surgeon Factor; SRK/T A-constant; Holladay 2 or Hoffer Q anterior chamber depth; Haigis a0, a1, and a2) for a given formula in order to make adjustments for a variety of practice-specific variables, including different styles of IOLs, keratometers, and variations in A-scan biometry calibration. Most IOL power calculation programs provide either internal software or specific recommendations for how to go about lens constant optimization. Currently, Wolfgang Haigis, MD, of Würzburg, Germany, and I carry out surgeon-specific optimization of the three Haigis formula variables.
SURGICAL TECHNIQUE
The configuration of the capsulorhexis can affect refractive outcomes if a surgeon is implanting a single-piece acrylic or a three-piece assembled IOL. If the capsulorhexis' diameter is larger than the lens optic, the forces of capsular bag contraction may anteriorly displace the IOL, a situation resulting in an increased effective lens power and more myopia than anticipated (Figure 4).
A simple “rhexis rule” is that the capsulorhexis should be round, centered, and slightly smaller than the optic. In order for the IOL power calculation formula to be most consistent and accurate, the capsular bag should completely contain the IOL. Attention to this detail can help maximize refractive accuracy.
CONCLUSION
Overall suggestions for improving IOL power calculations include (1) minimizing the number of variables, (2) verifying measurements when necessary, (3) relying upon either immersion ultrasound or optical coherence biometry, (4) carefully tracking your refractive outcomes, (5) optimizing the lens constants for each IOL used, and (6) creating a round, centered capsulorhexis that is slightly smaller than the IOL's optic. By following these simple rules, you will be well on your way to maximizing the accuracy of your refractive outcomes after cataract surgery.
Warren E. Hill, MD, FACS, is Medical Director of East Valley Ophthalmology in Mesa, Arizona. He is a consultant for Carl Zeiss Meditec Inc. but holds no financial interest in the products mentioned herein. Dr. Hill may be reached at (480) 981-6111; hill@doctor-hill.com.
1. Vogel A, Dick B, Krummenauer F. Reproducibility of optical biometry using partial coherence interferometry: intraobserver and interobserver reliability. J Cataract Refract Surg. 2001;27:1961-1968.
2. Holladay JT. Standardizing constants for ultrasonic biometry, keratometry, and intraocular lens power calculations. J Cataract Refract Surg. 1997;23:1356-1370.
3. Holladay JT, Gills JP, Leidlen J, Cherchio M. Achieving emmetropia in extremely short eyes with two piggyback posterior chamber intraocular lenses. Ophthalmology. 1996;103:1118-1123.
