Douglas D. Koch

Calculating the surgically induced refractive change following ocular surgery

ABSTRACT Calculating the surgically induced refractive change following ocular surgery is important for evaluating the results of keratorefractive procedures, smaller incisions and various wound closures for cataract surgery, and the effect of suturing techniques and suture removal following corneal transplant surgery. We present a ten‐step method of calculating the spherical‐ and cylindrical‐induced refractive change in a manner suitable for a programmable calculator or personal computer. Several applications are given including (1) adding the overrefraction to the spectacle correction, (2) determining the surgically induced refractive change from the preoperative and postoperative refractions, (3) determining the surgically induced refractive change from the K‐readings, (4) rotating axes, (5) determining the power at meridians oblique to the principal meridians of a spherocylinder, (6) determining the coupling ratio, and (7) averaging axes. Standard methods for calculating and reporting aggregate results are also given.

Optical aberrations of the human anterior cornea

Purpose: (1) To investigate the distribution of anterior corneal higher‐order aberrations (HOAs, 3rd to 6th orders) in the population; (2) to evaluate the symmetry of anterior corneal aberrations between right and left eyes of individuals; and (3) to study the variations in anterior corneal aberrations with aging. Setting: Cullen Eye Institute, Baylor College of Medicine, Houston, Texas, USA. Methods: Using the CTView program (Sarver and Associates, Inc.), corneal HOAs were computed from the central 6.0 mm zone of the corneal topographic maps (Humphrey Atlas, Carl Zeiss, Inc.) of 228 eyes of 134 subjects (mean age 50 years ± 17 (SD); range 20 to 79 years; manifest spherical equivalent between –3.0 and +3.0 diopters [D]). Correlation analysis was performed to investigate the aberration symmetry between the right and left eyes and to assess the association between anterior corneal HOAs and age. Results: There was wide individual variability in aberrations with ranges of individual Zernike terms from −0.579 to +0.572 &mgr;m. The mean coefficient of the 4th‐order spherical aberration (SA) (Z40) was 0.280 ± 0.086 &mgr;m and was positive in all corneas. The mean root‐mean‐square (RMS) values were 0.479 ± 0.124 &mgr;m for HOA, 0.281 ± 0.086 &mgr;m for SA (Z40and Z60), and 0.248 ± 0.135 &mgr;m for coma (Z3−1, Z31, Z5−1, and Z51). Moderate to high correlations were found between right and left eyes for HOA, SA, and coma (Pearson r = 0.565, 0.751, and 0.565, respectively; all P<.001); 8 of the 9 Zernike terms in the 3rd and 4th orders were significantly correlated between eyes (Bonferroni correction P′<.05/22). Higher‐order aberration RMS and coma RMS increased with aging (r = 0.434 and 0.290, respectively; both P<.001); SA RMS was not correlated with change in age (r = 0.034). Conclusion: Anterior corneal wavefront aberrations varied greatly among subjects and increased slightly with aging. All corneas had positive 4th‐order SAs, and the values did not change with aging. A moderate to high degree of mirror symmetry existed between right and left eyes.

Contribution of posterior corneal astigmatism to total corneal astigmatism.

PURPOSE: To determine the contribution of posterior corneal astigmatism to total corneal astigmatism and the error in estimating total corneal astigmatism from anterior corneal measurements only using a dual‐Scheimpflug analyzer. SETTING: Cullen Eye Institute, Baylor College of Medicine, Houston, Texas, USA. DESIGN: Case series. METHODS: Total corneal astigmatism was calculated using ray tracing, corneal astigmatism from simulated keratometry, anterior corneal astigmatism, and posterior corneal astigmatism, and the changes with age were analyzed. Vector analysis was used to assess the error produced by estimating total corneal astigmatism from anterior corneal measurements only. RESULTS: The study analyzed 715 corneas of 435 consecutive patients. The mean magnitude of posterior corneal astigmatism was −0.30 diopter (D). The steep corneal meridian was aligned vertically (60 to 120 degrees) in 51.9% of eyes for the anterior surface and in 86.6% for the posterior surface. With increasing age, the steep anterior corneal meridian tended to change from vertical to horizontal, while the steep posterior corneal meridian did not change. The magnitudes of anterior and posterior corneal astigmatism were correlated when the steeper anterior meridian was aligned vertically but not when it was aligned horizontally. Anterior corneal measurements underestimated total corneal astigmatism by 0.22 @ 180 and exceeded 0.50 D in 5% of eyes. CONCLUSIONS: Ignoring posterior corneal astigmatism may yield incorrect estimation of total corneal astigmatism. Selecting toric intraocular lenses based on anterior corneal measurements could lead to overcorrection in eyes that have with‐the‐rule astigmatism and undercorrection in eyes that have against‐the‐rule astigmatism. Financial Disclosure: The authors received research support from Ziemer Group. In addition, Dr. Koch has a financial interest with Alcon Laboratories, Inc., Abbott Medical Optics, Inc., Calhoun Vision, Inc., NuLens, and Optimedica Corp.

Ocular higher-order aberrations in individuals screened for refractive surgery

Purpose: To explore the distribution of ocular higher‐order aberrations (HOAs, 3rd to 6th orders) in the population, evaluate the symmetry of ocular aberrations between right and left eyes in each subject using a Hartmann‐Shack wavefront sensor, and study the differences in aberration as a function of age. Setting: Cullen Eye Institute, Baylor College of Medicine, Houston, Texas, USA. Methods: Ocular HOAs were examined across a 6.0 mm pupil in 532 eyes of 306 subjects (mean age 41 years ± 10 [SD] [range 20 to 71 years]; mean WaveScan spherical equivalent −3.39 ± 2.84 diopters [D] [range −11.56 to 7.60 D]) using the WaveScan system (Visx, Inc.). Zernike coefficients and root‐mean‐square (RMS) values of HOAs, spherical aberration (SA, Z40 and Z60), and coma (Z3−1, Z31, Z5−1, and Z51) were analyzed. Correlation analysis was performed to assess the association between ocular HOAs and age and investigate the aberration symmetry between right and left eyes. Results: For individual terms, the highest mean absolute values were for 4th‐order SA (Z40), 3rd‐order coma, and trefoil terms. The mean RMS values of HOA, SA, and coma were 0.305 ± 0.095 &mgr;m, 0.128 ± 0.074 &mgr;m, and 0.170 ± 0.089 &mgr;m, respectively. Moderate to high correlations were found between the right and left eyes for HOA, SA, and coma (Pearson correlation coefficient = 0.601, 0.776, and 0.511, respectively; all P<.001). Thirteen of the 22 Zernike terms (59%) were significantly correlated across eyes (Bonferroni correction, P′<.05/22). Higher‐order aberrations, SA, and coma were weakly correlated with age (r = 0.317, 0.273, and 0.176, respectively; all P<.002). Conclusion: Wavefront aberrations varied widely among subjects and increased slightly with age. A moderate to high degree of mirror symmetry existed between right and left eyes.

Evaluating and reporting astigmatism for individual and aggregate data

Purpose: To demonstrate the proper method for evaluating and reporting astigmatism for individual and aggregate data. Setting: University of Texas Medical School and Cullen Eye Institute, Baylor College of Medicine, Houston, Texas, USA. Methods: The surgically induced refractive change (SIRC) was determined for three data sets of patients who have had keratorefractive (photorefractive keratectomy) or cataract surgery. To make changes in refraction comparable, vertex distances for the refractions and keratometric index of refraction were considered. Doubledangle plots and single‐angle plots were then used to display the data. Polar values (cylinder and axis) were converted to a Cartesian (x and y) coordinate system to determine the mean value of the induced astigmatism for each data set. Results: Doubled‐angle plots clearly demonstrated the trends of induced astigmatism for each data set, and the mean value for induced astigmatism agreed exactly with the intuitive appearance of the plot. Conclusions: Converting astigmatism data to a Cartesian coordinate system allowed the correct computation of descriptive statistics such as mean values, standard deviations, and correlation coefficients. Using doubled‐angle plots to display the data provides the investigator with the best method of recognizing trends in the data.

Format for reporting refractive surgical data

W ith the proliferation of refractive surgical procedures, it is essential that refractive surgical data be reported in a clear and consistent manner. 1,2 Regrettably, current data presentation is somewhat haphazard, and key data may be incompletely or unclearly described. To assist readers and authors alike, the editors of this journal wish to establish a standard format for data presentation that will be incorporated into the Instructions for Authors. The adoption of a standard format should also serve as a guide for developing a study design that will capture pertinent data in a systematic manner. We describe the important structural and reporting elements of refractive surgical articles. Essentially all these elements are mandatory, although some could be omitted if not pertinent to the particular study.

A comparative analysis of five methods of determining corneal refractive power in eyes that have undergone myopic laser in situ keratomileusis

OBJECTIVE To evaluate the accuracy of computerized videokeratography, keratometry, and the Gaussian optics formula for measuring corneal refractive power in patients after myopic laser in situ keratomileusis (LASIK). DESIGN Noncomparative case series. PARTICIPANTS One hundred eyes of 63 patients (mean age, 45.0 +/- 10.9 [standard deviation] years) who underwent LASIK were included in the study. METHODS Using the clinical history method as the standard, we evaluated the accuracy of values of corneal refractive power derived from computerized videokeratography (the EffRP value of the EyeSys Corneal Analysis System, which averages corneal refractive power over the central 3 mm), keratometry (K), the Gaussian optics formula (GauRP), and values of EffRP and keratometry as modified according to the amount of LASIK-induced refractive change. MAIN OUTCOME MEASURES Correlation of measured corneal power values to those obtained using clinical history method (HisRP). RESULTS Although the values for HisRP were significantly correlated with postoperative EffRP and K values and with GauRP, postoperative EffRP and K values were higher than HisRP (0.87 +/- 0.68 diopters [D] and 1.16 +/- 1.10 D, respectively), and GauRP were lower than HisRP (0.44 +/- 0.66 D) (P < 0.001 for all three comparisons). The differences between HisRP and both postoperative EffRP and K values increased significantly with the amount of myopic correction. The most accurate results were obtained by modifying the postoperative values of EffRP according to the amount of LASIK-induced refractive change; 70% of these values were within +/- 0.5 D and 94% within +/- 1 D of HisRP values. CONCLUSIONS Using the clinical history method as the standard, we found that the most accurate method for determining corneal refractive power in post-LASIK eyes was to adjust the postoperative corneal measurement according to the amount of LASIK-induced refractive change.

Diffuse lamellar keratitis associated with epithelial defects after laser in situ keratomileusis

Purpose: To describe the association between the presence of epithelial defects and the development of diffuse lamellar keratitis (DLK), “Sands of the Sahara” syndrome, following laser in situ keratomileusis (LASIK). Setting: Cullen Eye Institute, Department of Ophthalmology, Baylor College of Medicine, Houston, Texas, USA. Methods: In this retrospective study, the postoperative clinical course and surgical outcomes in 735 eyes of 358 consecutive patients who had myopic LASIK between December 1998 and August 1999 were reviewed. Of the 735 procedures, 680 were primary LASIK procedures and 55 were retreatments performed by lifting the existing flaps. The incidence and severity of DLK and the relationship of DLK to epithelial defects were tabulated. Results: Diffuse lamellar keratitis developed in 9 of the 16 eyes that had postoperative epithelial defects. Seventeen eyes without epithelial defects also developed DLK. The presence of an epithelial defect increased an individuals risk of developing DLK 24 times (95% confidence interval, 13 to 45). In 8 eyes, the inflammation resolved following treatment with intense topical corticosteroids. One eye had irrigation under the flap because of dense central inflammation; the final outcome was mild inferior corneal steepening. All eyes recovered preoperative best spectacle‐corrected visual acuities. Conclusion: Patients who have epithelial defects of any size following LASIK are at significantly increased risk of developing DLK.

Correcting astigmatism with toric intraocular lenses: Effect of posterior corneal astigmatism

Purpose To evaluate the impact of posterior corneal astigmatism on outcomes with toric intraocular lenses (IOLs). Setting Cullen Eye Institute, Baylor College of Medicine, Houston, Texas, USA. Design Case series. Methods Corneal astigmatism was measured using 5 devices before and 3 weeks after cataract surgery. Toric IOL alignment was recorded at surgery and at the slitlamp 3 weeks postoperatively. The actual corneal astigmatism was calculated based on refractive astigmatism 3 weeks postoperatively and the effective toric power calculated with the Holladay 2 formula. The prediction error was calculated as the difference between the astigmatism measured by each device and the actual corneal astigmatism. Vector analysis was used in all calculations. Results With the IOLMaster, Lenstar, Atlas, manual keratometer, and Galilei (combined Placido–dual Scheimpflug analyzer), the mean prediction errors (D) were, respectively, 0.59 @ 89.7, 0.48 @ 91.2, 0.51 @ 78.7, 0.62 @ 97.2, and 0.57 @ 93.9 for with‐the‐rule (WTR) astigmatism (60 to 120 degrees), and 0.17 @ 86.2, 0.23 @ 77.7, 0.23 @ 91.4, 0.41 @ 58.4, and 0.12 @ 7.3 for against‐the‐rule (ATR) astigmatism (0 to 30 degrees and 150 to 180 degrees). In the WTR eyes, there were significant WTR prediction errors (0.5 to 0.6 diopters [D]) by all devices. In ATR eyes, WTR prediction errors were 0.2 to 0.3 D by all devices except the Placido–dual Scheimpflug analyzer (all P<.05 with Bonferroni correction). Conclusions Corneal astigmatism was overestimated in WTR by all devices and underestimated in ATR by all except the Placido–dual Scheimpflug analyzer. A new toric IOL nomogram is proposed. Financial Disclosure Drs. Koch, Weikert, and Wang received research support from Ziemer USA, Inc. Dr. Koch has a financial interest with Alcon Laboratories, Inc., Abbott Medical Optics, Inc., Optimedica Corp., and Ziemer USA, Inc. No other author has a financial or proprietary interest in any material or method mentioned.

Retrospective comparison of techniques to prevent secondary cataract formation after posterior chamber intraocular lens implantation in infants and children

Purpose: To determine the effect of various methods of managing the posterior capsule and anterior vitreous on the rate of posterior capsule opacification in pediatric eyes implanted with posterior chamber intraocular lenses (PC IOLs). Setting: Cullen Eye Institute, Baylor College of Medicine, Houston, Texas, USA. Methods: We reviewed the charts of 20 eyes of 15 children (aged 1.5 to 2 years) who had primary cataract surgery with PC IOL implantation during the past 5 years. The posterior capsule and anterior vitreous were managed in a variety of ways: In 5 eyes, the posterior capsule was left intact; in 15 eyes, a posterior continuous curvilinear capsulorhexis (PCCC) was performed‐6 with and 9 without anterior vitrectomy; in 8 eyes, posterior optic capture was performed‐3 with and 5 without vitrectomy. The follow‐up ranged from 1 to 4.5 years (mean 2 years). Results: Visually significant secondary cataract developed in the five eyes with intact posterior capsules and in the four eyes that had PCCC without vitrectomy and without posterior optic capture (i.e., the optic was left in the capsular bag). The optical axis remained clear in the six eyes that had PC IOL implantation with vitrectomy (with or without posterior optic capture). Initially, all eyes that had optic capture without vitrectomy also remained clear, but after 6 months, four of five developed opacification. Conclusion: In this series, PCCC with anterior vitrectomy was the only effective method of preventing or delaying secondary cataract formation in infants and children.