Although the term "refractive cataract surgery" is redundant, it is a helpful reminder that we must strive to attain the same precision in refractive outcomes for our cataract patients as we do for our patients undergoing laser vision correction and phakic IOLs. A realistic goal is within 0.50 D of the intended target, and a benchmark study has shown that 55% of patients attain this mark while 85% are within 1.0 D.
Achieving the best refractive results is a multifactorial process. Technologic advances are certainly helping us reach this goal, but in order to consistently deliver outstanding surgical outcomes, we must understand the components that contribute to IOL power calculation, validate and optimize these various components, detect and treat ocular surface disease, be cognizant of surgically induced astigmatism, and standardize our surgical technique for wound and capsulorhexis construction.
A refractive "surprise" can cause serious problems as evidenced by the fact that wrong IOL power is a common reason for ophthalmic malpractice claims. About one third of ophthalmology claims are for cataract surgery, and 25% of these are IOL-related, mostly due to incorrect IOL power. Therefore, to minimize such an occurrence, it is important to pay careful attention to the factors used for IOL power calculation.
The most significant factors that affect IOL power include axial length (AL), average keratometry (K), and effective lens position (ELP). ELP is preoperatively estimated by IOL formulas and anterior chamber depth (ACD) values. This works well for 3rd generation theoretic formulas in eyes with average AL and K readings, but not in eyes with high or low readings. The reason is that these formulas make the incorrect assumptions that 1) the anterior and posterior segments are proportional, and 2) the K and ACD are related. The ELP for a posterior chamber IOL is predicted to be in the plane of the zonules, but the actual IOL resting position is determined by the configuration of the capsulorhexis. The ideal capsulorhexis for current posterior chamber IOLs is round, well-centered, and slightly smaller in diameter than the IOL optic, so that it symmetrically overlaps and contains the optic in the capsular bag. A large or an asymmetric capsulorhexis will not overlap or will only partially overlap the IOL optic, and subsequent fibrosis and contraction of the capsule may cause the IOL to rest anterior to its predicted location or to tilt inducing a myopic shift or astigmatism, respectively. A 1 mm error in AL measurement results in about a 3 D error in postop refraction, a 1 D error in K measurement produces a 1 D refractive error, and a 0.5 mm change in the IOL position induces 1 D of refractive change. We can reduce these sources of IOL calculation errors by optimizing and validating AL and K measurements with the best technology and careful attention to technique, as well as properly creating the capsulorhexis. The other factors that contribute to IOL power calculations are the IOL formula, retinal thickness at the fovea, and IOL manufacturing tolerances.
A significant range of error can occur in predicted IOL power depending upon the accuracy of each of the aforementioned factors. Dr. Warren Hill writes and lectures extensively on this subject. His superb presentation "Highly Accurate IOL Calculations" (available on the Zeiss website: Click here to watch) reviews each component in detail and includes an exercise in which he demonstrates the effect on the final refractive outcome by optimizing each component. The resulting mean absolute error values for a normal eye range from 0.71 D (for applanation biometry, manual keratometry, variable capsulorhexis, nonoptimized 3rd generation IOL formula; i.e., none of the 6 components optimized) to 0.24 D (for IOLMaster measurements, perfect capsulorhexis, optimized 4th generation IOL formula; all of the components optimized). Unfortunately, optimizing only 1 or 2 of the components provides very little improvement in final IOL error (0.61-0.68 D) since this value is calculated from the square root of the sum of the squares for each component.
The use of optical coherence devices such as the IOLMaster and Lenstar 900 in conjunction with optimized 4th generation IOL formulas enables us to minimize the error in most of the components necessary for IOL calculations. This includes the retinal thickness variation around the fovea which is a factor for A-scan ultrasound, but not for these instruments which use a narrow beam that measures AL to the RPE. The capsulorhexis is a surgeon dependent manual procedure and is therefore not perfect every time; however, technology such as femtosecond lasers that cut a precise capsulotomy has the potential to further increase the accuracy of our outcomes by guaranteeing the proper position of the IOL. Three companies (LenSx, LensAR, and OptiMedica) are developing femtosecond laser devices for cataract surgery. Not only do these instruments cut the capsule, but also they are able to fragment the lens nucleus and make corneal incisions. The current IOL manufacturing tolerances of most IOLs (0.50 D increments) only contributes 0.17 D mean absolute error to the equation; however, this will improve slightly as lenses become available in 0.25 D increments.
In addition, tear film and corneal irregularities (i.e, ocular surface disease, anterior basement membrane dystrophy) can have a significant impact on keratometry measurements. Preoperatively, these conditions must be detected and managed appropriately to improve the accuracy of IOL calculations. We must remember that "garbage in equals garbage out", and also that our measurements are less accurate for extreme AL and K values. Furthermore, refractive cataract surgery requires that surgeons be aware of the surgically induced astigmatism they create because this affects the overall result. If we pay attention to and optimize all these variables, then it is possible to achieve results within 0.25 D of the targeted outcome.