- Email: [email protected]
Residual astigmatism after toric intraocular lens implantation: Analysis of data from an online toric intraocular lens back-calculator
ARTICLE
Residual astigmatism after toric intraocular lens implantation: Analysis of data from an online toric intraocular lens back-calculator Brent A. Kramer, BS, John P. Berdahl, MD, David R. Hardten, MD, Richard Potvin, OD
PURPOSE: To evaluate some possible causes for residual astigmatism after toric intraocular lens (IOL) implantation based on an analysis of data from an online toric IOL back-calculator. DESIGN: Retrospective data review. METHODS: An online toric back-calculator was designed to allow users to input preoperative toric planning information along with postoperative IOL orientation and refractive results. These were then used to determine the optimum orientation of the IOL to reduce refractive astigmatism. The collected aggregate data were extracted from this calculator to investigate the associated reasons for residual astigmatic refractive error with toric IOLs. RESULTS: The study analyzed 12 812 records with a mean postoperative refractive astigmatism of 1.89 diopters (D). Refractive astigmatism was significantly higher with higher IOL cylinder power (P < .01) but was not different by IOL manufacturer. Ninety percent of IOLs were not at the ideal orientation, despite 30% being at the preoperative calculated orientation. Misalignment showed a directional bias for some IOLs but not for others. The mean calculated percentage reduction in residual cylinder after reorientation was 50% G 31% (SD), with the magnitude of residual astigmatism after IOL reorientation expected to be 0.50 D or less in 37% of eyes (4835/12 812). Expected outcomes were significantly different by IOL type. CONCLUSIONS: Analysis of data from the online toric back-calculator provided insights into the nature of residual astigmatism after toric IOL implantation. The reasons for residual astigmatism in this data set varied by IOL type. Financial Disclosure: Proprietary or commercial disclosures are listed after the references. J Cataract Refract Surg 2016; 42:1595–1601 Q 2016 ASCRS and ESCRS
A large percentage of patients presenting for cataract surgery are likely to inquire about astigmatism reduction. A study in Germany1 found that a third of 23 239 eyes had 1.00 diopter (D) or more of astigmatism. Two common options to reduce astigmatism at the time of cataract surgery are relaxing incisions, toric intraocular lens (IOL) implantation, or both. A 2016 metaanalysis of 13 randomized clinical trials2 concluded that toric IOLs were more effective at reducing astigmatism than monofocal IOLs, with or without the use of relaxing incisions. Despite the noted efficacy of toric IOLs, up to 28% or 47% of eyes had more than 0.50 D of residual refractive astigmatism and up to
Q 2016 ASCRS and ESCRS Published by Elsevier Inc.
6% or 16% had more than 1.00 D of residual refractive astigmatism depending on the toric IOL model and cylinder power.3,4 Residual astigmatism after toric IOL implantation has also been reported to range from 0.00 to 2.25 D depending on the preoperative astigmatism.5 Although the rates of significant residual refractive astigmatism are relatively low, a better understanding of the factors associated with suboptimum outcomes could be helpful. There are 2 fundamental reasons a toric IOL might not correct the refractive astigmatism in an eye. Either the IOL is not in the appropriate orientation to correct the astigmatism or the IOL has too much or too little
http://dx.doi.org/10.1016/j.jcrs.2016.09.017 0886-3350
1595
1596
ANALYSIS OF RESIDUAL ASTIGMATISM FROM AN ONLINE TORIC IOL BACK-CALCULATOR
cylinder power. Of course, 1 or both of these might occur, and the causes for each might vary. Three causes of residual refractive astigmatism appear most likely. First is related to the precision and accuracy of the measurements of the eye and consideration of contributing effects, such as surgically induced astigmatism (SIA) or posterior corneal astigmatism. Studies6,7 have shown that using 1 keratometer over another or combining 2 keratometry measurements might help improve outcomes. Dry eye can significantly affect the accuracy of keratometry.8 Posterior corneal astigmatism can also be a factor; ignoring posterior corneal astigmatism is believed to overestimate and underestimate the total corneal astigmatism in with-the-rule and against-the-rule anterior astigmatism, respectively.9 Surgically induced astigmatism is another important consideration because actual SIA can be quite variable relative to its mean.5,10,11 Second is the accuracy of orientation of the IOL relative to the intended target, which can be affected by the orientation at the time of surgery and any postoperative rotation of the IOL. One study showed that the average toric IOL misalignment was less than 5 degrees although misalignment can be higher than about 6 degrees in 25% of patients.5 A large metaanalysis of 4 toric IOL models2 found that the mean rotation was less than 5 degrees and an additional procedure was necessary to correct misalignment in only 6 patients of the 554 receiving toric IOLs. Third is the accuracy of the toric IOL calculation. Calculations can be affected by the presumed effective lens position (ELP) because a toric IOL cylinder is based on the IOL plane and nominal corneal plane values are used in the calculators. Intraocular lens
efficacy might also contribute because cylinder at the corneal plane changes with the spherical power of the IOL implanted. Goggin et al.12 were the first to describe the importance of taking into account both the ELP and spherical power when performing toric IOL calculation; however, not all calculators take this into consideration.13 Postoperative data are required to analyze the 3 causes listed above. The ideal orientation of the IOL can be determined from its current (measured) orientation and the residual refractive astigmatism in the eye using vector mathematics. This ideal orientation can then be compared with the intended orientation from the toric IOL calculator and the actual postoperative orientation. Figure 1 shows the relationship of these various factors. The orientations pictured are for illustrative purposes only. The appropriateness of the IOL cylinder power can be determined from the expected residual astigmatism when the IOL in the eye is at the ideal orientation. The data provided when using a toric back-calculator can be especially useful in providing this necessary postoperative information. The purpose of the current analysis was to determine whether the aggregate results from the users of a toric back-calculation website could provide insights into the causes of residual refractive astigmatism after implantation of a toric IOL. MATERIALS AND METHODS An application for approval of this data review was submitted to the University of Iowa Human Subjects Office/Institutional Review Board and resulted in a waiver of this requirement because there was no protected health information in the stored data from the website.
Submitted: April 10, 2016. Final revision submitted: August 6, 2016. Accepted: September 7, 2016. From the University of Iowa Carver College of Medicine (Kramer), Iowa City, Iowa, Vance Thompson Vision (Berdahl), Sioux Falls, South Dakota, Minnesota Eye Consultants (Hardten), Minnetonka, Minnesota, and Science in Vision (Potvin), Akron, New York, USA. Data analysis and the preparation of this manuscript were supported with an investigator-initiated research grant to Ocular Surgical Data, LLC, from Alcon Laboratories, Inc., Fort Worth, Texas USA (IIT: 23134697). Sarah Makari, OD, Science in Vision, provided writing assistance in the preparation of the manuscript. Corresponding author: Richard Potvin, OD, Science in Vision, 6197 Dye Road, Akron, New York 14001, USA. E-mail: rick@ scienceinvision.com.
Figure 1. Intraocular lens orientation data and relevant variables of interest (IOL Z intraocular lens).
J CATARACT REFRACT SURG - VOL 42, NOVEMBER 2016
ANALYSIS OF RESIDUAL ASTIGMATISM FROM AN ONLINE TORIC IOL BACK-CALCULATOR
The analysis here is based on the data collected from a website designed to assist surgeons in reducing or eliminating postoperative refractive astigmatism after toric IOL implantation.A Figure 2 shows the input and output screens of the online back-calculator. Available data for each eye, including the residual refraction (sphere, cylinder, and axis), the cylinder power and orientation of the toric IOL along with details such as the eye (right or left), the intended orientation of the IOL, the actual IOL model and power, and some user identifiers, are input on the screen, as shown in Figure 2, A. Using the current toric IOL orientation and cylinder power as well as the current manifest refraction, the toric back-calculator determines the orientation of the IOL that will minimize the residual refractive astigmatism in the eye. This optimum orientation and the expected residual refractive astigmatism after reorientation are provided to the user, along with a plot showing the expected residual refractive astigmatism associated with the IOL in the eye through 360 degrees of orientation (Figure 2, B). The data set of interest included all calculations collected since inception of the website (mid-2012 to December 31, 2015). Data from several iterations of the online program were collected; early versions of the program did not include details, such as the date of the calculation or the intended orientation. A preliminary data review was performed to reduce the likelihood of erroneous and/or multiple calculations. Additional analytical variables, both numeric and categorical, were calculated from the raw data. Vector mathematics was used where appropriate for astigmatism calculations. Toric usage data in the United States were obtained from Market Scope, LLC.B To examine the reasons for residual astigmatism in the data set, the following categorical criteria were applied: The current orientation of the IOL was considered “ideal” if it was within G5 degrees of the calculated ideal orientation. The current orientation of the IOL was considered “as intended” if it was within G5 degrees of the intended orientation. This is because any difference within G5 degrees is likely within the standard measurement error.14 Where the
1597
expected residual refractive astigmatism (after IOL reorientation) was more than 0.50 D, it can be determined whether this is the result of an overcorrection or undercorrection, based on the nominal direction of the residual refractive cylinder once the IOL is in the ideal orientation. The absolute angle between the ideal orientation and the expected residual astigmatism axis in plus cylinder form was calculated as a value between zero and 90. If this angle was greater than 80 degrees, the eye was overcorrected (the axis was flipped). If the angle was less than 10 degrees, the eye was undercorrected. For any other values, the result was unclear. The difference between the actual orientation and the intended orientation was also calculated, with a negative number indicating an orientation clockwise of that intended. The data extracted from the online website were available in a comma-delimited text file. The data were imported into an Access database (Microsoft Corp.) for data checking, collation, and preliminary analysis. Statistical analyses were performed using Statistica software (version 12, Statsoft, Inc.). Statistical testing was performed using analysis of variance (ANOVA) on continuous variables and appropriate nonparametric tests on categorical data. Statistical significance was set at a P value of 0.05.
RESULTS A total of 35 846 raw data records were available from the website; these calculation requests were from an estimated 3000 surgeons. Range criteria (eg, “eye” was specified as right or left, axes of astigmatism were between 0 degrees and 180 degrees, absolute residual refractive cylinder was 10.00 D or less, absolute residual refractive sphere was 6.00 D or less) were used to identify and qualify valid records. When long series of calculations (O10 on the same day) were found for 1 user, the user was asked (by e-mail message) whether the calculations were for research
Figure 2. Input (A) and output (B) screens of online back-calculator.
J CATARACT REFRACT SURG - VOL 42, NOVEMBER 2016
1598
ANALYSIS OF RESIDUAL ASTIGMATISM FROM AN ONLINE TORIC IOL BACK-CALCULATOR
purposes rather than a “true” calculation; identified research calculations were removed. Data from known site testers were also removed, along with records that were exact duplicates. After this preliminary data qualification, 19 018 data records remained. Of the remaining records, 8793 were identified as unique, being the only calculation on a given day for a given user for a specific IOL. An additional 10 225 records were considered possible duplicates. A test of several algorithms to aggregate these records suggested that the most reliable was to take the final calculation for a given user and IOL combination on a given day as the best; the other calculations were discarded. This yielded 4019 aggregated records. The net result was 12 812 total data records for subsequent analysis. Figure 3 shows the data cleanup and filtering process. Of these records, 10 176 included the intended orientation of the IOL in the eye; early versions of the website did not collect this information. Estimates suggest 890 000 toric IOLs were implanted in the U.S. from mid-2012 to the end of 2015. If a similar number were implanted outside the U.S., the data records here would represent 0.7% (12 812 of 1 780 000) of all implanted toric IOLs. Figure 4 shows the residual refractive cylinder from all validated data records. Seventy percent (8946 of 12 818) of the calculations involved residual refractive astigmatism between 0.50 D and 2.00 D, while 90% of all cases (11 635 of 12 818) were between 0.50 D and 3.00 D. Although not shown, there was a statistically significant difference in the magnitude of residual refractive cylinder by IOL manufacturer (P ! .01); however, the differences were clinically insignificant. The overall average magnitude of residual refractive astigmatism was 1.86 D; the Tecnis toric (Abbot Medical Optics), Acrysof toric (Alcon Laboratories, Inc.), Staar toric (Staar Surgical), Trulign toric (Bausch &
Figure 4. The distribution of residual refractive astigmatism.
Figure 3. Flowchart of the back-calculator raw data cleanup and filtering.
Lomb), and not specified IOL groups were within G0.25 D of this overall average. The Tecnis, Acrysof, and not specified IOL groups accounted for 98% (12 541/12 812) of all entries; these 3 groups had a wider range of IOL cylinder powers at the corneal plane than the Trulign and Staar groups. Intraocular lenses in the former 3 groups were categorized as high cylinder power (O2.50 D at the corneal plane) or low cylinder power (%2.50 D at the corneal plane). Figure 5 shows the results of ANOVA of residual refractive cylinder by IOL model and cylinder power category. There was a statistically and clinically significant difference in the residual refractive astigmatism by cylinder power (P ! .01) but no difference between the IOL groups (P Z .68). An early version of the software did not include the intended axis; thus, only 10 176 available data records were available for analysis. Table 1 shows the
Figure 5. Residual refractive cylinder by IOL model and cylinder power category (IOL Z intraocular lens).
J CATARACT REFRACT SURG - VOL 42, NOVEMBER 2016
1599
ANALYSIS OF RESIDUAL ASTIGMATISM FROM AN ONLINE TORIC IOL BACK-CALCULATOR
Table 1. Intended and ideal IOL orientation.
Table 2. Rates of overcorrection or undercorrection by IOL model.
Oriented as Intended, n (%) IOL/Ideal Orientation? Total Yes No Total Acrysof Yes No Total Tecnis Yes No Total
Yes
No
Total
628 (6.2) 614 (6.0) 1242 (12.2) 2971 (29.2) 5963 (58.6) 8934 (87.8) 3599 (35.4) 6577 (64.6) 10 176 (100) 405 (7.3) 326 (5.9) 1861 (33.5) 2960 (53.3) 2266 (40.8) 3286 (59.2)
731 (13.2) 4821 (86.8) 5552 (100)
58 (2.4) 124 (5.2) 458 (19.0) 1766 (73.4) 516 (21.4) 1890 (78.6)
182 (7.6) 2224 (92.4) 2406 (100)
IOL Z intraocular lens
cross-tabulation of these data for all records, then the subset of Acrysof and Tecnis records. Note that about 90% of IOLs (8934 of 10 176) were in the non-ideal orientation, suggesting that reorientation would be helpful to some degree in all these cases. Note, too, that approximately 30% (2971/10 176) of all IOLs were oriented as intended but were not at an ideal orientation. The percentage of IOLs in this category (oriented as intended Z yes; ideal Z no) was significantly lower for the Tecnis IOLs than for the Acrysof IOLs (P ! .01). It was unclear whether the eye was overcorrected or undercorrected in fewer than 0.7% of cases (51/7977), all with a not specified IOL, so they were removed from the summary in Table 2. Table 2 shows the
Table 3. Direction of difference from intended. Direction of Difference, n (%) Ideal Orientation Total Yes No Total Acrysof Yes No Total Tecnis Yes No Total
CCW
None
CW
Total
8 75 (24.3) 1805 (50.2) 919 (25.5) 3599 (100) 3697 (56.2) 2880 (43.8) 6577 (100) 4572 (44.9) 1805 (17.7) 3799 (37.3) 10 176 (100) 555 (24.5) 1119 (49.4) 592 (26.1) 1674 (50.9) 1612 (49.1) 2229 (40.1) 1119 (20.2) 2204 (39.7)
2266 (100) 3286 (100) 5552 (100)
137 (26.6) 256 (49.6) 123 (23.8) 1236 (65.4) 654 (34.6) 1373 (57.1) 256 (10.6) 777 (32.3)
516 (100) 1890 (100) 2406 (100)
CCW Z counterclockwise; CW Z clockwise
IOL
Total (n)
Over (n)
Under (n)
% Over
3499 1319 84 58 2966 7926
1478 670 21 16 1258 3443
2021 649 63 42 1708 4483
42.2 50.8 25.0 27.6 42.4 43.4
Acrysof Tecnis Trulign Staar Not specified Total
IOL Z intraocular lens; Over Z overcorrected; Under Z undercorrected
percentage of eyes undercorrected or overcorrected by IOL type. There was an overall bias to undercorrection; undercorrection was significantly more likely with other IOL types than with the Tecnis IOL (P ! .01), which was the only IOL for which there appeared to be an equal likelihood of overcorrection or undercorrection. Table 3 shows the direction of the difference between the intended orientation and actual orientation for all IOLs and for the Acrysof and Tecnis IOLs. The row percentages are provided, for both the IOLs within G5 degrees of the intended orientation and for IOLs more than 5 degrees from the intended orientation. In the former group, the directional differences were relatively evenly spread. However, for IOLs with an orientation more than 5 degrees different from the intended, there was a statistically significant difference by IOL, with the Acrysof IOLs showing no particular directional bias and the Tecnis IOLs oriented more often in a relative counterclockwise direction from the intended (P ! .01). The expected outcome from rotating the IOL already in the eye is of primary interest to users of the site. The magnitude of residual astigmatism was expected to be 0.50 D or less after IOL rotation in 37% of eyes (4835/12 812). There was a significant difference in this success metric between IOLs, with a higher percentage of Tecnis IOLs calculated to provide residual astigmatism of 0.50 D or less relative to Acrysof IOLs (Tecnis 45.2% [1088/2407] versus Acrysof 37.0% [2053/5552]) (P ! .01). This was independent of the cylinder power (low or high) of the toric IOLs in question. Without regard to changes in the angle of astigmatism, there was an expected 50% mean percentage reduction in the magnitude of the refractive cylinder through reorientation of the IOLs, with a median reduction of 54%. This percentage reduction in cylinder from reorientation was significantly higher with Tecnis IOLs than with Acrysof IOLs (56.5% G 30.3% [SD] versus 49.5% G 31.9%) (P ! .01).
J CATARACT REFRACT SURG - VOL 42, NOVEMBER 2016
1600
ANALYSIS OF RESIDUAL ASTIGMATISM FROM AN ONLINE TORIC IOL BACK-CALCULATOR
DISCUSSION This large data set available from a toric backcalculation website provided a unique opportunity to assist surgeons in reducing postoperative refractive astigmatism after toric IOL implantation and to analyze the factors associated with this residual astigmatism. The mean residual astigmatism magnitude entered was 1.89 D, with a 90th percentile range of 0.50 to 3.00 D, which is significantly higher than that observed in normal clinical populations.2,5 Knowing the intended orientation of the IOL before surgery and calculating the ideal orientation after surgery (based on the residual refractive astigmatism and the known IOL orientation) allowed us to make relative comparisons. We believe this is the first large-scale effort to analyze postoperative refractive astigmatism after toric IOL implantation. In 30% of cases overall, the IOLs were oriented as intended; however, the intended orientation was not ideal. This suggests there are deficiencies in the preoperative calculation or changes to the cornea at the time of surgery that might affect the calculated orientation. Preoperative effects would include things such as variability in SIA, effects of posterior corneal astigmatism that are not accounted for, and variability in keratometric measurement.5 The relative importance of these factors has not been well quantified. A more detailed analysis of these factors will be the subject of future analyses. To better characterize the factors might require that additional data be collected on the website. The difference between intended and ideal orientation in this data set varied by IOL model, as did the likelihood of overcorrection or undercorrection; this is likely a function of the toric calculator associated with each IOL. We have presumed in the given time period that the toric IOL calculator associated with a given IOL was used to determine cylinder power and orientation so that when implanting Acrysof toric or Tecnis toric IOLs, the surgeon would use the AcrysofC and TecnisD toric calculators, respectively. There are known differences between different toric calculators that could explain some of the differences observed in the current study. For instance, the Acrysof toric calculatorC yields a result that never allows residual refractive astigmatism in the orthogonal meridian (ie, the axis of astigmatism cannot be flipped). This results in a bias toward undercorrection, as seen here and consistent with results in previous studies15 and expectations from a simulation of the effects of the calculator.16 More specifically, a comparison between the Tecnis and Acrysof toric calculators indicates that the Acrysof toric calculatorC suggested a lower cylinder power than the Tecnis toric calculator,D which might explain why the bias for
undercorrection is not present when using the Tecnis toric IOL.15 With new generic calculators now available,13,15 this might not be as true in the future. Recording the toric calculator used for the original IOL power calculation would appear to be an important addition to consider for the website. There was also a significant difference in the nature of misalignment by IOL model, with some IOLs showing a directional bias. It has been suggested that the direction of rotation might be the result of surgical factors specifically related to the capsulorhexis or to IOL factors such as the haptics.17 A subsequent manuscript is planned to explore this issue in more detail. On the basis of the results presented here, using a toric back-calculator for postoperative analysis of residual astigmatism after toric IOL implantation can be recommended. Data show that reorienting a toric IOL would be likely to have some benefit in 90% of cases. In the current data set, the median expected reduction in cylinder magnitude was 54%; 37% of eyes were expected to have less than 0.50 D of cylinder postoperatively. There were, however, several cases in which reorientation appeared insufficient to provide a significant reduction in postoperative refractive cylinder. In recognition of this, a future version of the website is planned that would include suggested IOL exchange parameters. There are limitations to the data set analyzed here. Although input data were screened, duplicate or erroneous records might have been retained. It is expected that the large sample mitigates the effect of these filtering errors. Results are limited to a small percentage (!1.0%) of eyes with a toric IOL. The website is used only when there is significant residual refractive cylinder; as such, the collected data are not representative of the typical results with a toric IOL. There are also other options for toric IOL back-calculation, so not all eyes with residual astigmatism will be captured. However, the data that are captured provide the opportunity to better understand the differences between IOLs and eyes that are associated with residual astigmatism after toric IOL implantation. This preliminary analysis also highlighted additional variables that might be helpful to collect with future versions of the website to better delineate the causative factors of postoperative refractive astigmatism. In conclusion, the online toric back-calculator can be helpful to surgeons when they encounter residual refractive astigmatism after toric IOL implantation and provides a useful analytical data set to help better understand the causes of residual refractive astigmatism. The latter may provide objective support for
J CATARACT REFRACT SURG - VOL 42, NOVEMBER 2016
ANALYSIS OF RESIDUAL ASTIGMATISM FROM AN ONLINE TORIC IOL BACK-CALCULATOR
changes to planning and performing cataract surgery with toric IOLs. WHAT WAS KNOWN Postoperative refractive astigmatism is present in some patients after toric IOL implantation. Rotation of toric IOLs might be used to reduce or eliminate this residual refractive astigmatism.
10.
11.
12.
WHAT THIS PAPER ADDS Use of a toric back-calculator to analyze refractive astigmatism after toric IOL implantation is recommended to predict the effect of IOL reorientation. Analysis of toric back-calculator inputs showed overall, and IOL-specific, factors that are associated with postoperative refractive astigmatism.
13.
14.
15.
REFERENCES €tz WW. Analysis of biometry and prevalence 1. Hoffmann PC, Hu data for corneal astigmatism in 23,239 eyes. J Cataract Refract Surg 2010; 36:1479–1485 2. Kessel L, Andresen J, Tendal B, Erngaard D, Flesner P, Hjortdal J. Toric intraocular lenses in the correction of astigmatism during cataract surgery; a systematic review and metaanalysis. Ophthalmology 2016; 123:275–286. Available at: http://www.aaojournal.org/article/S0161-6420(15)01148-3/pdf. Accessed September 10, 2016 3. Visser N, Bauer NJC, Nuijts RMMA. Toric intraocular lenses: historical overview, patient selection, IOL calculation, surgical techniques, clinical outcomes, and complications. J Cataract Refract Surg 2013; 39:624–637 4. Waltz KL, Featherstone K, Tsai L, Trentacost D. Clinical outcomes of TECNIS toric intraocular lens implantation after cataract removal in patients with corneal astigmatism. Ophthalmology 2015; 122:39–47 5. Hirnschall N, Hoffmann PC, Draschl P, Maedel S, Findl O. Evaluation of factors influencing the remaining astigmatism after toric intraocular lens implantation. J Refract Surg 2014; 30:394–400 6. Potvin R, Gundersen KG, Masket S, Osher RH, Snyder ME, Vann RR, Solomon KD, Hill WE. Prospective multicenter study of toric IOL outcomes when dual zone automated keratometry is used for astigmatism planning. J Refract Surg 2013; 29:804–809 7. Browne AW, Osher RH. Optimizing precision in toric lens selection by combining keratometry techniques. J Refract Surg 2014; 30:67–72 8. Epitropoulos AT, Matossian C, Berdy GJ, Malhotra RP, Potvin R. Effect of tear osmolarity on repeatability of keratometry for cataract surgery planning. J Cataract Refract Surg 2015; 41:1672–1677 9. Zheng T, Chen Z, Lu Y. Influence factors of estimation errors for total corneal astigmatism using keratometric astigmatism in
16.
17.
1601
patients before cataract surgery. J Cataract Refract Surg 2016; 42:84–94 Hill W. Expected effects of surgically induced astigmatism on AcrySof toric intraocular lens results. J Cataract Refract Surg 2008; 34:364–367 Savini G, Næser K. An analysis of the factors influencing the residual refractive astigmatism after cataract surgery with toric intraocular lenses. Invest Ophthalmol Vis Sci 2015; 56:827– 835. Available at: http://iovs.arvojournals.org/article.aspx?articleidZ2212837. Accessed September 10, 2016 Goggin M, Moore S, Esterman A. Outcome of toric intraocular lens implantation after adjusting for anterior chamber depth and intraocular lens sphere equivalent power effects. Arch Ophthalmol 2011; 129:998–1003. correction, 1494. Available at: http://archopht.jamanetwork.com/data/Journals/OPHTH/22540/ ecs05115_998_1003.pdf. Correction Available at: http://archopht.jamanetwork.com/data/Journals/OPHTH/22547/ecx15011_ 1494_1494.pdf. Both Accessed September 10, 2016 Abulafia A, Hill WE, Franchina M, Barrett GD. Comparison of methods to predict residual astigmatism after intraocular lens implantation. J Refract Surg 2015; 31:699–707 Cha D, Kang SY, Kim S-H, Song J-S, Kim H-M. New axismarking method for a toric intraocular lens: mapping method. J Refract Surg 2011; 27:375–379 Park HJ, Lee H, Woo YJ, Kim EK, Seo KY, Kim HY, Kim TI. Comparison of the astigmatic power of toric intraocular lenses using three toric calculators. Yonsei Med J 2015; 56:1097–1105. Available at: https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC4479840/pdf/ymj-56-1097.pdf. Accessed September 10, 2016 Hill W, Potvin R. Monte Carlo simulation of expected outcomes with the AcrySofÒ toric intraocular lens. BMC Ophthalmol 2008; 8:22. Available at: http://www.ncbi.nlm.nih.gov/pmc/articles/ PMC2586009/pdf/1471-2415-8-22.pdf. Accessed September 10, 2016 Shah GD, Praveen MR, Vasavada AR, Vasavada VA, Rampal G, Shastry LR. Rotational stability of a toric intraocular lens: influence of axial length and alignment in the capsular bag. J Cataract Refract Surg 2012; 38:54–59
OTHER CITED MATERIAL A. Berdahl J, Hardten D. Toric Results Analyzer. Available at: www.astigmatismfix.com. Accessed September 10, 2016 B. Toric IOL use reported in the quarterly survey of U.S. cataract surgeons. St. Louis, MO, Market Scope, LLC C. Alcon Surgical, Inc. AcrySofÒ Toric IOL Web Based Calculators. Available at: http://www.acrysoftoriccalculator.com. Accessed September 10, 2016 D. Abbott Medical Optics, Inc. Calculate the Proper Cylinder Power for the TechnisÒ Toric IOL. Available at: http://www.tecnisiol. com/us/physician/calculation. Accessed September 10, 2016
FINANCIAL DISCLOSURES Drs. Berdahl and Hardten are owners of Ocular Surgical Data LLC, makers of astigmatismfix.com. Dr. Berdahl is a consultant to Abbott Medical Optics, Inc., Alcon Laboratories, Inc., and Bausch & Lomb, Inc. Dr. Hardten is a consultant to Abbott Medical Optics, Inc., ESI, and TLC Laser Eye Center, Dr. Potvin is a consultant to Alcon Lab€ te GmbH. Mr. oratories, Inc., Haag-Streit AG, and Oculus Optikgera Kramer has no financial or proprietary interest in any material or method mentioned.
J CATARACT REFRACT SURG - VOL 42, NOVEMBER 2016
Residual astigmatism after toric intraocular lens implantation: Analysis of data from an online toric intraocular lens back-calculator Brent A. Kramer, BS, John P. Berdahl, MD, David R. Hardten, MD, Richard Potvin, OD
PURPOSE: To evaluate some possible causes for residual astigmatism after toric intraocular lens (IOL) implantation based on an analysis of data from an online toric IOL back-calculator. DESIGN: Retrospective data review. METHODS: An online toric back-calculator was designed to allow users to input preoperative toric planning information along with postoperative IOL orientation and refractive results. These were then used to determine the optimum orientation of the IOL to reduce refractive astigmatism. The collected aggregate data were extracted from this calculator to investigate the associated reasons for residual astigmatic refractive error with toric IOLs. RESULTS: The study analyzed 12 812 records with a mean postoperative refractive astigmatism of 1.89 diopters (D). Refractive astigmatism was significantly higher with higher IOL cylinder power (P < .01) but was not different by IOL manufacturer. Ninety percent of IOLs were not at the ideal orientation, despite 30% being at the preoperative calculated orientation. Misalignment showed a directional bias for some IOLs but not for others. The mean calculated percentage reduction in residual cylinder after reorientation was 50% G 31% (SD), with the magnitude of residual astigmatism after IOL reorientation expected to be 0.50 D or less in 37% of eyes (4835/12 812). Expected outcomes were significantly different by IOL type. CONCLUSIONS: Analysis of data from the online toric back-calculator provided insights into the nature of residual astigmatism after toric IOL implantation. The reasons for residual astigmatism in this data set varied by IOL type. Financial Disclosure: Proprietary or commercial disclosures are listed after the references. J Cataract Refract Surg 2016; 42:1595–1601 Q 2016 ASCRS and ESCRS
A large percentage of patients presenting for cataract surgery are likely to inquire about astigmatism reduction. A study in Germany1 found that a third of 23 239 eyes had 1.00 diopter (D) or more of astigmatism. Two common options to reduce astigmatism at the time of cataract surgery are relaxing incisions, toric intraocular lens (IOL) implantation, or both. A 2016 metaanalysis of 13 randomized clinical trials2 concluded that toric IOLs were more effective at reducing astigmatism than monofocal IOLs, with or without the use of relaxing incisions. Despite the noted efficacy of toric IOLs, up to 28% or 47% of eyes had more than 0.50 D of residual refractive astigmatism and up to
Q 2016 ASCRS and ESCRS Published by Elsevier Inc.
6% or 16% had more than 1.00 D of residual refractive astigmatism depending on the toric IOL model and cylinder power.3,4 Residual astigmatism after toric IOL implantation has also been reported to range from 0.00 to 2.25 D depending on the preoperative astigmatism.5 Although the rates of significant residual refractive astigmatism are relatively low, a better understanding of the factors associated with suboptimum outcomes could be helpful. There are 2 fundamental reasons a toric IOL might not correct the refractive astigmatism in an eye. Either the IOL is not in the appropriate orientation to correct the astigmatism or the IOL has too much or too little
http://dx.doi.org/10.1016/j.jcrs.2016.09.017 0886-3350
1595
1596
ANALYSIS OF RESIDUAL ASTIGMATISM FROM AN ONLINE TORIC IOL BACK-CALCULATOR
cylinder power. Of course, 1 or both of these might occur, and the causes for each might vary. Three causes of residual refractive astigmatism appear most likely. First is related to the precision and accuracy of the measurements of the eye and consideration of contributing effects, such as surgically induced astigmatism (SIA) or posterior corneal astigmatism. Studies6,7 have shown that using 1 keratometer over another or combining 2 keratometry measurements might help improve outcomes. Dry eye can significantly affect the accuracy of keratometry.8 Posterior corneal astigmatism can also be a factor; ignoring posterior corneal astigmatism is believed to overestimate and underestimate the total corneal astigmatism in with-the-rule and against-the-rule anterior astigmatism, respectively.9 Surgically induced astigmatism is another important consideration because actual SIA can be quite variable relative to its mean.5,10,11 Second is the accuracy of orientation of the IOL relative to the intended target, which can be affected by the orientation at the time of surgery and any postoperative rotation of the IOL. One study showed that the average toric IOL misalignment was less than 5 degrees although misalignment can be higher than about 6 degrees in 25% of patients.5 A large metaanalysis of 4 toric IOL models2 found that the mean rotation was less than 5 degrees and an additional procedure was necessary to correct misalignment in only 6 patients of the 554 receiving toric IOLs. Third is the accuracy of the toric IOL calculation. Calculations can be affected by the presumed effective lens position (ELP) because a toric IOL cylinder is based on the IOL plane and nominal corneal plane values are used in the calculators. Intraocular lens
efficacy might also contribute because cylinder at the corneal plane changes with the spherical power of the IOL implanted. Goggin et al.12 were the first to describe the importance of taking into account both the ELP and spherical power when performing toric IOL calculation; however, not all calculators take this into consideration.13 Postoperative data are required to analyze the 3 causes listed above. The ideal orientation of the IOL can be determined from its current (measured) orientation and the residual refractive astigmatism in the eye using vector mathematics. This ideal orientation can then be compared with the intended orientation from the toric IOL calculator and the actual postoperative orientation. Figure 1 shows the relationship of these various factors. The orientations pictured are for illustrative purposes only. The appropriateness of the IOL cylinder power can be determined from the expected residual astigmatism when the IOL in the eye is at the ideal orientation. The data provided when using a toric back-calculator can be especially useful in providing this necessary postoperative information. The purpose of the current analysis was to determine whether the aggregate results from the users of a toric back-calculation website could provide insights into the causes of residual refractive astigmatism after implantation of a toric IOL. MATERIALS AND METHODS An application for approval of this data review was submitted to the University of Iowa Human Subjects Office/Institutional Review Board and resulted in a waiver of this requirement because there was no protected health information in the stored data from the website.
Submitted: April 10, 2016. Final revision submitted: August 6, 2016. Accepted: September 7, 2016. From the University of Iowa Carver College of Medicine (Kramer), Iowa City, Iowa, Vance Thompson Vision (Berdahl), Sioux Falls, South Dakota, Minnesota Eye Consultants (Hardten), Minnetonka, Minnesota, and Science in Vision (Potvin), Akron, New York, USA. Data analysis and the preparation of this manuscript were supported with an investigator-initiated research grant to Ocular Surgical Data, LLC, from Alcon Laboratories, Inc., Fort Worth, Texas USA (IIT: 23134697). Sarah Makari, OD, Science in Vision, provided writing assistance in the preparation of the manuscript. Corresponding author: Richard Potvin, OD, Science in Vision, 6197 Dye Road, Akron, New York 14001, USA. E-mail: rick@ scienceinvision.com.
Figure 1. Intraocular lens orientation data and relevant variables of interest (IOL Z intraocular lens).
J CATARACT REFRACT SURG - VOL 42, NOVEMBER 2016
ANALYSIS OF RESIDUAL ASTIGMATISM FROM AN ONLINE TORIC IOL BACK-CALCULATOR
The analysis here is based on the data collected from a website designed to assist surgeons in reducing or eliminating postoperative refractive astigmatism after toric IOL implantation.A Figure 2 shows the input and output screens of the online back-calculator. Available data for each eye, including the residual refraction (sphere, cylinder, and axis), the cylinder power and orientation of the toric IOL along with details such as the eye (right or left), the intended orientation of the IOL, the actual IOL model and power, and some user identifiers, are input on the screen, as shown in Figure 2, A. Using the current toric IOL orientation and cylinder power as well as the current manifest refraction, the toric back-calculator determines the orientation of the IOL that will minimize the residual refractive astigmatism in the eye. This optimum orientation and the expected residual refractive astigmatism after reorientation are provided to the user, along with a plot showing the expected residual refractive astigmatism associated with the IOL in the eye through 360 degrees of orientation (Figure 2, B). The data set of interest included all calculations collected since inception of the website (mid-2012 to December 31, 2015). Data from several iterations of the online program were collected; early versions of the program did not include details, such as the date of the calculation or the intended orientation. A preliminary data review was performed to reduce the likelihood of erroneous and/or multiple calculations. Additional analytical variables, both numeric and categorical, were calculated from the raw data. Vector mathematics was used where appropriate for astigmatism calculations. Toric usage data in the United States were obtained from Market Scope, LLC.B To examine the reasons for residual astigmatism in the data set, the following categorical criteria were applied: The current orientation of the IOL was considered “ideal” if it was within G5 degrees of the calculated ideal orientation. The current orientation of the IOL was considered “as intended” if it was within G5 degrees of the intended orientation. This is because any difference within G5 degrees is likely within the standard measurement error.14 Where the
1597
expected residual refractive astigmatism (after IOL reorientation) was more than 0.50 D, it can be determined whether this is the result of an overcorrection or undercorrection, based on the nominal direction of the residual refractive cylinder once the IOL is in the ideal orientation. The absolute angle between the ideal orientation and the expected residual astigmatism axis in plus cylinder form was calculated as a value between zero and 90. If this angle was greater than 80 degrees, the eye was overcorrected (the axis was flipped). If the angle was less than 10 degrees, the eye was undercorrected. For any other values, the result was unclear. The difference between the actual orientation and the intended orientation was also calculated, with a negative number indicating an orientation clockwise of that intended. The data extracted from the online website were available in a comma-delimited text file. The data were imported into an Access database (Microsoft Corp.) for data checking, collation, and preliminary analysis. Statistical analyses were performed using Statistica software (version 12, Statsoft, Inc.). Statistical testing was performed using analysis of variance (ANOVA) on continuous variables and appropriate nonparametric tests on categorical data. Statistical significance was set at a P value of 0.05.
RESULTS A total of 35 846 raw data records were available from the website; these calculation requests were from an estimated 3000 surgeons. Range criteria (eg, “eye” was specified as right or left, axes of astigmatism were between 0 degrees and 180 degrees, absolute residual refractive cylinder was 10.00 D or less, absolute residual refractive sphere was 6.00 D or less) were used to identify and qualify valid records. When long series of calculations (O10 on the same day) were found for 1 user, the user was asked (by e-mail message) whether the calculations were for research
Figure 2. Input (A) and output (B) screens of online back-calculator.
J CATARACT REFRACT SURG - VOL 42, NOVEMBER 2016
1598
ANALYSIS OF RESIDUAL ASTIGMATISM FROM AN ONLINE TORIC IOL BACK-CALCULATOR
purposes rather than a “true” calculation; identified research calculations were removed. Data from known site testers were also removed, along with records that were exact duplicates. After this preliminary data qualification, 19 018 data records remained. Of the remaining records, 8793 were identified as unique, being the only calculation on a given day for a given user for a specific IOL. An additional 10 225 records were considered possible duplicates. A test of several algorithms to aggregate these records suggested that the most reliable was to take the final calculation for a given user and IOL combination on a given day as the best; the other calculations were discarded. This yielded 4019 aggregated records. The net result was 12 812 total data records for subsequent analysis. Figure 3 shows the data cleanup and filtering process. Of these records, 10 176 included the intended orientation of the IOL in the eye; early versions of the website did not collect this information. Estimates suggest 890 000 toric IOLs were implanted in the U.S. from mid-2012 to the end of 2015. If a similar number were implanted outside the U.S., the data records here would represent 0.7% (12 812 of 1 780 000) of all implanted toric IOLs. Figure 4 shows the residual refractive cylinder from all validated data records. Seventy percent (8946 of 12 818) of the calculations involved residual refractive astigmatism between 0.50 D and 2.00 D, while 90% of all cases (11 635 of 12 818) were between 0.50 D and 3.00 D. Although not shown, there was a statistically significant difference in the magnitude of residual refractive cylinder by IOL manufacturer (P ! .01); however, the differences were clinically insignificant. The overall average magnitude of residual refractive astigmatism was 1.86 D; the Tecnis toric (Abbot Medical Optics), Acrysof toric (Alcon Laboratories, Inc.), Staar toric (Staar Surgical), Trulign toric (Bausch &
Figure 4. The distribution of residual refractive astigmatism.
Figure 3. Flowchart of the back-calculator raw data cleanup and filtering.
Lomb), and not specified IOL groups were within G0.25 D of this overall average. The Tecnis, Acrysof, and not specified IOL groups accounted for 98% (12 541/12 812) of all entries; these 3 groups had a wider range of IOL cylinder powers at the corneal plane than the Trulign and Staar groups. Intraocular lenses in the former 3 groups were categorized as high cylinder power (O2.50 D at the corneal plane) or low cylinder power (%2.50 D at the corneal plane). Figure 5 shows the results of ANOVA of residual refractive cylinder by IOL model and cylinder power category. There was a statistically and clinically significant difference in the residual refractive astigmatism by cylinder power (P ! .01) but no difference between the IOL groups (P Z .68). An early version of the software did not include the intended axis; thus, only 10 176 available data records were available for analysis. Table 1 shows the
Figure 5. Residual refractive cylinder by IOL model and cylinder power category (IOL Z intraocular lens).
J CATARACT REFRACT SURG - VOL 42, NOVEMBER 2016
1599
ANALYSIS OF RESIDUAL ASTIGMATISM FROM AN ONLINE TORIC IOL BACK-CALCULATOR
Table 1. Intended and ideal IOL orientation.
Table 2. Rates of overcorrection or undercorrection by IOL model.
Oriented as Intended, n (%) IOL/Ideal Orientation? Total Yes No Total Acrysof Yes No Total Tecnis Yes No Total
Yes
No
Total
628 (6.2) 614 (6.0) 1242 (12.2) 2971 (29.2) 5963 (58.6) 8934 (87.8) 3599 (35.4) 6577 (64.6) 10 176 (100) 405 (7.3) 326 (5.9) 1861 (33.5) 2960 (53.3) 2266 (40.8) 3286 (59.2)
731 (13.2) 4821 (86.8) 5552 (100)
58 (2.4) 124 (5.2) 458 (19.0) 1766 (73.4) 516 (21.4) 1890 (78.6)
182 (7.6) 2224 (92.4) 2406 (100)
IOL Z intraocular lens
cross-tabulation of these data for all records, then the subset of Acrysof and Tecnis records. Note that about 90% of IOLs (8934 of 10 176) were in the non-ideal orientation, suggesting that reorientation would be helpful to some degree in all these cases. Note, too, that approximately 30% (2971/10 176) of all IOLs were oriented as intended but were not at an ideal orientation. The percentage of IOLs in this category (oriented as intended Z yes; ideal Z no) was significantly lower for the Tecnis IOLs than for the Acrysof IOLs (P ! .01). It was unclear whether the eye was overcorrected or undercorrected in fewer than 0.7% of cases (51/7977), all with a not specified IOL, so they were removed from the summary in Table 2. Table 2 shows the
Table 3. Direction of difference from intended. Direction of Difference, n (%) Ideal Orientation Total Yes No Total Acrysof Yes No Total Tecnis Yes No Total
CCW
None
CW
Total
8 75 (24.3) 1805 (50.2) 919 (25.5) 3599 (100) 3697 (56.2) 2880 (43.8) 6577 (100) 4572 (44.9) 1805 (17.7) 3799 (37.3) 10 176 (100) 555 (24.5) 1119 (49.4) 592 (26.1) 1674 (50.9) 1612 (49.1) 2229 (40.1) 1119 (20.2) 2204 (39.7)
2266 (100) 3286 (100) 5552 (100)
137 (26.6) 256 (49.6) 123 (23.8) 1236 (65.4) 654 (34.6) 1373 (57.1) 256 (10.6) 777 (32.3)
516 (100) 1890 (100) 2406 (100)
CCW Z counterclockwise; CW Z clockwise
IOL
Total (n)
Over (n)
Under (n)
% Over
3499 1319 84 58 2966 7926
1478 670 21 16 1258 3443
2021 649 63 42 1708 4483
42.2 50.8 25.0 27.6 42.4 43.4
Acrysof Tecnis Trulign Staar Not specified Total
IOL Z intraocular lens; Over Z overcorrected; Under Z undercorrected
percentage of eyes undercorrected or overcorrected by IOL type. There was an overall bias to undercorrection; undercorrection was significantly more likely with other IOL types than with the Tecnis IOL (P ! .01), which was the only IOL for which there appeared to be an equal likelihood of overcorrection or undercorrection. Table 3 shows the direction of the difference between the intended orientation and actual orientation for all IOLs and for the Acrysof and Tecnis IOLs. The row percentages are provided, for both the IOLs within G5 degrees of the intended orientation and for IOLs more than 5 degrees from the intended orientation. In the former group, the directional differences were relatively evenly spread. However, for IOLs with an orientation more than 5 degrees different from the intended, there was a statistically significant difference by IOL, with the Acrysof IOLs showing no particular directional bias and the Tecnis IOLs oriented more often in a relative counterclockwise direction from the intended (P ! .01). The expected outcome from rotating the IOL already in the eye is of primary interest to users of the site. The magnitude of residual astigmatism was expected to be 0.50 D or less after IOL rotation in 37% of eyes (4835/12 812). There was a significant difference in this success metric between IOLs, with a higher percentage of Tecnis IOLs calculated to provide residual astigmatism of 0.50 D or less relative to Acrysof IOLs (Tecnis 45.2% [1088/2407] versus Acrysof 37.0% [2053/5552]) (P ! .01). This was independent of the cylinder power (low or high) of the toric IOLs in question. Without regard to changes in the angle of astigmatism, there was an expected 50% mean percentage reduction in the magnitude of the refractive cylinder through reorientation of the IOLs, with a median reduction of 54%. This percentage reduction in cylinder from reorientation was significantly higher with Tecnis IOLs than with Acrysof IOLs (56.5% G 30.3% [SD] versus 49.5% G 31.9%) (P ! .01).
J CATARACT REFRACT SURG - VOL 42, NOVEMBER 2016
1600
ANALYSIS OF RESIDUAL ASTIGMATISM FROM AN ONLINE TORIC IOL BACK-CALCULATOR
DISCUSSION This large data set available from a toric backcalculation website provided a unique opportunity to assist surgeons in reducing postoperative refractive astigmatism after toric IOL implantation and to analyze the factors associated with this residual astigmatism. The mean residual astigmatism magnitude entered was 1.89 D, with a 90th percentile range of 0.50 to 3.00 D, which is significantly higher than that observed in normal clinical populations.2,5 Knowing the intended orientation of the IOL before surgery and calculating the ideal orientation after surgery (based on the residual refractive astigmatism and the known IOL orientation) allowed us to make relative comparisons. We believe this is the first large-scale effort to analyze postoperative refractive astigmatism after toric IOL implantation. In 30% of cases overall, the IOLs were oriented as intended; however, the intended orientation was not ideal. This suggests there are deficiencies in the preoperative calculation or changes to the cornea at the time of surgery that might affect the calculated orientation. Preoperative effects would include things such as variability in SIA, effects of posterior corneal astigmatism that are not accounted for, and variability in keratometric measurement.5 The relative importance of these factors has not been well quantified. A more detailed analysis of these factors will be the subject of future analyses. To better characterize the factors might require that additional data be collected on the website. The difference between intended and ideal orientation in this data set varied by IOL model, as did the likelihood of overcorrection or undercorrection; this is likely a function of the toric calculator associated with each IOL. We have presumed in the given time period that the toric IOL calculator associated with a given IOL was used to determine cylinder power and orientation so that when implanting Acrysof toric or Tecnis toric IOLs, the surgeon would use the AcrysofC and TecnisD toric calculators, respectively. There are known differences between different toric calculators that could explain some of the differences observed in the current study. For instance, the Acrysof toric calculatorC yields a result that never allows residual refractive astigmatism in the orthogonal meridian (ie, the axis of astigmatism cannot be flipped). This results in a bias toward undercorrection, as seen here and consistent with results in previous studies15 and expectations from a simulation of the effects of the calculator.16 More specifically, a comparison between the Tecnis and Acrysof toric calculators indicates that the Acrysof toric calculatorC suggested a lower cylinder power than the Tecnis toric calculator,D which might explain why the bias for
undercorrection is not present when using the Tecnis toric IOL.15 With new generic calculators now available,13,15 this might not be as true in the future. Recording the toric calculator used for the original IOL power calculation would appear to be an important addition to consider for the website. There was also a significant difference in the nature of misalignment by IOL model, with some IOLs showing a directional bias. It has been suggested that the direction of rotation might be the result of surgical factors specifically related to the capsulorhexis or to IOL factors such as the haptics.17 A subsequent manuscript is planned to explore this issue in more detail. On the basis of the results presented here, using a toric back-calculator for postoperative analysis of residual astigmatism after toric IOL implantation can be recommended. Data show that reorienting a toric IOL would be likely to have some benefit in 90% of cases. In the current data set, the median expected reduction in cylinder magnitude was 54%; 37% of eyes were expected to have less than 0.50 D of cylinder postoperatively. There were, however, several cases in which reorientation appeared insufficient to provide a significant reduction in postoperative refractive cylinder. In recognition of this, a future version of the website is planned that would include suggested IOL exchange parameters. There are limitations to the data set analyzed here. Although input data were screened, duplicate or erroneous records might have been retained. It is expected that the large sample mitigates the effect of these filtering errors. Results are limited to a small percentage (!1.0%) of eyes with a toric IOL. The website is used only when there is significant residual refractive cylinder; as such, the collected data are not representative of the typical results with a toric IOL. There are also other options for toric IOL back-calculation, so not all eyes with residual astigmatism will be captured. However, the data that are captured provide the opportunity to better understand the differences between IOLs and eyes that are associated with residual astigmatism after toric IOL implantation. This preliminary analysis also highlighted additional variables that might be helpful to collect with future versions of the website to better delineate the causative factors of postoperative refractive astigmatism. In conclusion, the online toric back-calculator can be helpful to surgeons when they encounter residual refractive astigmatism after toric IOL implantation and provides a useful analytical data set to help better understand the causes of residual refractive astigmatism. The latter may provide objective support for
J CATARACT REFRACT SURG - VOL 42, NOVEMBER 2016
ANALYSIS OF RESIDUAL ASTIGMATISM FROM AN ONLINE TORIC IOL BACK-CALCULATOR
changes to planning and performing cataract surgery with toric IOLs. WHAT WAS KNOWN Postoperative refractive astigmatism is present in some patients after toric IOL implantation. Rotation of toric IOLs might be used to reduce or eliminate this residual refractive astigmatism.
10.
11.
12.
WHAT THIS PAPER ADDS Use of a toric back-calculator to analyze refractive astigmatism after toric IOL implantation is recommended to predict the effect of IOL reorientation. Analysis of toric back-calculator inputs showed overall, and IOL-specific, factors that are associated with postoperative refractive astigmatism.
13.
14.
15.
REFERENCES €tz WW. Analysis of biometry and prevalence 1. Hoffmann PC, Hu data for corneal astigmatism in 23,239 eyes. J Cataract Refract Surg 2010; 36:1479–1485 2. Kessel L, Andresen J, Tendal B, Erngaard D, Flesner P, Hjortdal J. Toric intraocular lenses in the correction of astigmatism during cataract surgery; a systematic review and metaanalysis. Ophthalmology 2016; 123:275–286. Available at: http://www.aaojournal.org/article/S0161-6420(15)01148-3/pdf. Accessed September 10, 2016 3. Visser N, Bauer NJC, Nuijts RMMA. Toric intraocular lenses: historical overview, patient selection, IOL calculation, surgical techniques, clinical outcomes, and complications. J Cataract Refract Surg 2013; 39:624–637 4. Waltz KL, Featherstone K, Tsai L, Trentacost D. Clinical outcomes of TECNIS toric intraocular lens implantation after cataract removal in patients with corneal astigmatism. Ophthalmology 2015; 122:39–47 5. Hirnschall N, Hoffmann PC, Draschl P, Maedel S, Findl O. Evaluation of factors influencing the remaining astigmatism after toric intraocular lens implantation. J Refract Surg 2014; 30:394–400 6. Potvin R, Gundersen KG, Masket S, Osher RH, Snyder ME, Vann RR, Solomon KD, Hill WE. Prospective multicenter study of toric IOL outcomes when dual zone automated keratometry is used for astigmatism planning. J Refract Surg 2013; 29:804–809 7. Browne AW, Osher RH. Optimizing precision in toric lens selection by combining keratometry techniques. J Refract Surg 2014; 30:67–72 8. Epitropoulos AT, Matossian C, Berdy GJ, Malhotra RP, Potvin R. Effect of tear osmolarity on repeatability of keratometry for cataract surgery planning. J Cataract Refract Surg 2015; 41:1672–1677 9. Zheng T, Chen Z, Lu Y. Influence factors of estimation errors for total corneal astigmatism using keratometric astigmatism in
16.
17.
1601
patients before cataract surgery. J Cataract Refract Surg 2016; 42:84–94 Hill W. Expected effects of surgically induced astigmatism on AcrySof toric intraocular lens results. J Cataract Refract Surg 2008; 34:364–367 Savini G, Næser K. An analysis of the factors influencing the residual refractive astigmatism after cataract surgery with toric intraocular lenses. Invest Ophthalmol Vis Sci 2015; 56:827– 835. Available at: http://iovs.arvojournals.org/article.aspx?articleidZ2212837. Accessed September 10, 2016 Goggin M, Moore S, Esterman A. Outcome of toric intraocular lens implantation after adjusting for anterior chamber depth and intraocular lens sphere equivalent power effects. Arch Ophthalmol 2011; 129:998–1003. correction, 1494. Available at: http://archopht.jamanetwork.com/data/Journals/OPHTH/22540/ ecs05115_998_1003.pdf. Correction Available at: http://archopht.jamanetwork.com/data/Journals/OPHTH/22547/ecx15011_ 1494_1494.pdf. Both Accessed September 10, 2016 Abulafia A, Hill WE, Franchina M, Barrett GD. Comparison of methods to predict residual astigmatism after intraocular lens implantation. J Refract Surg 2015; 31:699–707 Cha D, Kang SY, Kim S-H, Song J-S, Kim H-M. New axismarking method for a toric intraocular lens: mapping method. J Refract Surg 2011; 27:375–379 Park HJ, Lee H, Woo YJ, Kim EK, Seo KY, Kim HY, Kim TI. Comparison of the astigmatic power of toric intraocular lenses using three toric calculators. Yonsei Med J 2015; 56:1097–1105. Available at: https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC4479840/pdf/ymj-56-1097.pdf. Accessed September 10, 2016 Hill W, Potvin R. Monte Carlo simulation of expected outcomes with the AcrySofÒ toric intraocular lens. BMC Ophthalmol 2008; 8:22. Available at: http://www.ncbi.nlm.nih.gov/pmc/articles/ PMC2586009/pdf/1471-2415-8-22.pdf. Accessed September 10, 2016 Shah GD, Praveen MR, Vasavada AR, Vasavada VA, Rampal G, Shastry LR. Rotational stability of a toric intraocular lens: influence of axial length and alignment in the capsular bag. J Cataract Refract Surg 2012; 38:54–59
OTHER CITED MATERIAL A. Berdahl J, Hardten D. Toric Results Analyzer. Available at: www.astigmatismfix.com. Accessed September 10, 2016 B. Toric IOL use reported in the quarterly survey of U.S. cataract surgeons. St. Louis, MO, Market Scope, LLC C. Alcon Surgical, Inc. AcrySofÒ Toric IOL Web Based Calculators. Available at: http://www.acrysoftoriccalculator.com. Accessed September 10, 2016 D. Abbott Medical Optics, Inc. Calculate the Proper Cylinder Power for the TechnisÒ Toric IOL. Available at: http://www.tecnisiol. com/us/physician/calculation. Accessed September 10, 2016
FINANCIAL DISCLOSURES Drs. Berdahl and Hardten are owners of Ocular Surgical Data LLC, makers of astigmatismfix.com. Dr. Berdahl is a consultant to Abbott Medical Optics, Inc., Alcon Laboratories, Inc., and Bausch & Lomb, Inc. Dr. Hardten is a consultant to Abbott Medical Optics, Inc., ESI, and TLC Laser Eye Center, Dr. Potvin is a consultant to Alcon Lab€ te GmbH. Mr. oratories, Inc., Haag-Streit AG, and Oculus Optikgera Kramer has no financial or proprietary interest in any material or method mentioned.
J CATARACT REFRACT SURG - VOL 42, NOVEMBER 2016