Eyeworld

OCT 2016

EyeWorld is the official news magazine of the American Society of Cataract & Refractive Surgery.

Issue link: https://digital.eyeworld.org/i/733437

Contents of this Issue

Navigation

Page 110 of 186

EW RESIDENTS 108 October 2016 by Sophie Cai, MD, and Michael Boland, MD, PhD, Wilmer Eye Institute, Johns Hopkins University School of Medicine Review of "Outcome of adjustment for posterior corneal curvature in toric intraocular lens calculation and selection" and "Prospective one-year evaluation of toric intraocular lens in eyes with with-the-rule, against-the-rule and oblique astigmatism" T otal corneal power in intraocular lens (IOL) calculations has histori- cally been estimated by measuring anterior corneal power alone—an approximation based on the likely oversimplified assumption 1 of a particular relation- ship between anterior and posterior corneal curvature. Recent literature suggests a need for greater attention to the contribution of posterior corneal astigmatism to postoperative cataract surgery outcomes, espe- cially in eyes implanted with toric IOLs. 2–5 A number of studies report a trend toward astigmatic overcor- rection in eyes with preoperative with-the-rule (WTR) astigmatism and undercorrection in eyes with preoperative against-the-rule (ATR) astigmatism. 6–10 A proposed mech- anism for this phenomenon is the tendency of the posterior cornea to have more ATR astigmatism than previously recognized, supported both by clinical observation and also by imaging research demonstrating greater pericentral corneal thick- ness in the vertical than horizontal meridian. 8,11,12 While the ideal solution would involve direct measurement of total or posterior corneal astigmatism, 13 a clinically validated gold standard for such measurement has yet to be established. 9,14–17 Several revised toric IOL formulae have shown promise in reducing postoperative residual astigmatism, including the Baylor nomogram, 16 Barrett toric calcula- tor, 18,19 and Abulafia-Koch formula. 20 In a retrospective study 7 of 143 consecutive eyes that underwent mi- croincision phacoemulsification us- ing a 1.9-mm temporal clear corneal incision on the horizontal meridian, Goggin et al. applied a geometric mean model to calculate separate co- efficients of adjustment for WTR and ATR eyes that can be multiplied by preoperative astigmatism to improve accuracy of toric IOL selection. WTR astigmatism was defined by a steep axis within 30 degrees of the vertical meridian, and ATR astigmatism was defined by a steep axis within 30 degrees of the horizontal meridian. In a follow-up study in the October issue of the Journal of Cataract & Re- fractive Surgery, "Outcome of adjust- ment for posterior corneal curvature in toric intraocular lens calculation and selection," Goggin et al. present the postoperative astigmatic out- comes of application of this "Goggin nomogram" to 31 consecutive eyes treated with the same microincision phacoemulsification technique as was used to develop their nomo- gram. All included eyes had preoper- ative keratometric cylinder power of at least 1 D and no more than 2 D of cylinder power in the selected toric IOL. Exclusion criteria were: corneal disease, history of previous kera- torefractive procedures, potential visual acuity worse than 20/40, and preoperative oblique astigmatism (defined by a steep axis not fitting the previously defined WTR or ATR categories). As a comparison group, the authors identified 65 histor- ical control eyes that underwent phacoemulsification by the same surgical technique but were implant- ed with toric IOLs selected using un- adjusted preoperative keratometry. The authors report that appli- cation of their Goggin nomogram resulted in a statistically significant (P=0.004) reduction in magnitude of error of predicted IOL cylinder pow- er from a median of 0.45 D (95% CI 0.33–0.58) among historical controls to 0.23 D (95% CI 0.13–0.35 D). To- ric IOLs were well-aligned with the targeted axis among eyes to which the Goggin nomogram was applied (mean misalignment of 0.8 degrees counterclockwise) and historical controls (mean misalignment of 2.5 degrees clockwise). The authors also report favorable postoperative visual acuity outcomes among eyes to which the Goggin nomogram adjustment was applied, but do not compare these with the analogous outcomes in historical controls. These results suggest that among eyes with a relatively low degree of preoperative astigmatism treated with a specific phacoemulsification surgical technique, adjustments to traditional estimation of total corneal power can improve accuracy of toric IOL selection and targeted postoperative astigmatism. While the halving of error of predicted toric IOL astigmatic effect is certainly encouraging, it may be too early to call this a "statistically and clinically highly significant" result. In particular, more evi- dence may be needed that a 0.25 D decrease in astigmatism is of substantial clinical consequence to patients, considering that elsewhere 7 the authors have considered an incision-induced change in corneal power of 0.19 D to be below the threshold of clinical refraction and thus clinically insignificant. It would be helpful if the authors showed that postoperative uncorrected and/ or corrected distance visual acuity were significantly improved by use of the Goggin nomogram. In addition, some limitations to the study design must be kept in mind. First, although the authors express some hesitation in their discussion about the ethical advis- ability of a prospective, random- ized, controlled trial to confirm the findings of their retrospective study, prospective studies in larger patient cohorts would in fact offer valuable follow-up to their work. Second, while the authors provide power calculations to support the adequacy of their relatively small sample sizes, it is not entirely clear how they picked the parameters used in these calculations, and in particular why they selected for an imbalance between the number of eyes to which the Goggin nomo- gram was applied and the number of control eyes. Third, the authors do not specify statistical adjustments for the potential bias introduced by including both eyes from some patients in their analysis. Fourth, the decision to exclude eyes with high or oblique astigmatism was based on findings from a single previous study suggesting that such eyes are less susceptible to systematic error in targeted postoperative astigma- tism. 7 Future studies will be import- ant to evaluate the validity of this conclusion. Fifth, the authors state that their nomogram "is designed to account for the likely effect of posterior corneal astigmatism which alters total […] corneal astigmatism in a predictable manner," whereas Koch et al. 8 have previously demon- strated clinically significant variabil- ity in the magnitude of posterior astigmatism regardless of the steep axis of anterior corneal astigmatism, concluding that posterior astigma- tism must be measured, rather than simply estimated. Even if a predict- able relationship between anterior and posterior corneal astigmatism could be reasonably assumed, it The role of the posterior cornea in postoperative astigmatism Michael Boland, MD, PhD, residency program director, Wilmer Eye Institute, Johns Hopkins University School of Medicine Sophie Cai, MD, first-year resident, Wilmer Eye Institute, Johns Hopkins University School of Medicine How should we adjust toric IOL power selection for pa- tients with ATR vs. WTR astigmatism? I asked the Wilmer residents to review two papers on this subject in the October issue of JCRS. –David F. Chang, MD, EyeWorld journal club editor EyeWorld journal club

Articles in this issue

Archives of this issue

view archives of Eyeworld - OCT 2016