The traditional ophthalmic microscope, a cornerstone of surgical precision, has undergone a remarkable evolution.1 In the early years of development, several modifications were adopted because of concerns raised regarding phototoxic maculopathy.2–17 Presently, operative microscopes confer a fundamental advantage to operators, whereby small shortcomings have the potential to be overshadowed. Thus, even in the current iteration, any deleterious effects of prolonged intense light exposure on ocular structures may not have been thoroughly explored.13–18 The transition to digital microscopy brings about a paradigm shift, offering ophthalmic surgeons a host of new benefits, not the least of which being that images derived from the analog microscope and displayed on the 3D flat panel screen have been digitally enhanced, thus providing an alternative method to coaxial illumination for improved visualization, as was exemplified in our previous pilot study.19–24

Over the past 5 years, high-definition 3D digital surgical visualization systems have been used successfully and safely in anterior segment surgeries.19,21,24 Previous studies on this platform have demonstrated equivalent surgical times and complication rates, improved patient comfort at the time of surgery, beneficial ergonomics, and decreased light intensity (in lux) from coaxial illumination.19–29 Several papers have alluded to an enhanced visual recovery when surgeons operated at overall lower light intensity percentages, stopping short of identifying the mechanism or rationale behind this phenomenon.19,20,25

The purpose of this study was not only to investigate whether the effect of decreased illumination can improve visual recovery after cataract surgery, but also if implementing digital visualization systems could yield fewer cases of macular damage such as cystoid macular edema (CME), especially in those patients at increased risk (eg, diabetics) of such changes.

METHODS

This study represents a prospective, randomized study of 214 consecutive eyes of 132 patients who underwent cataract surgery between January 2021 and January 2022 at the Long Island Ambulatory Surgery Center in Brentwood, NY. Patients were randomized using a random number generator (www.calculatorsoup.com) on the day of surgery and grouped into 2 surgical visualization approaches: (1) use of only the OPMI Lumera T operating microscope (Carl Zeiss Meditech AG) (“traditional group”) or (2) use of the NGENUITY 3D Visualization System (Alcon Laboratories, Inc.) affixed to the OPMI Lumera T microscope (“digital group”) (Figure 1). Surgeries were performed by a single surgeon, using the same phacoemulsification systems, operating microscope, 3D digital visualization system, and femtosecond laser (if used) for all patients. For those cases performed with adjunctive use of a femtosecond laser (Catalys Laser Precision System, Johnson & Johnson Vision), identical femtosecond parameters were implemented for capsulorrhexis formation, nucleus segmentation, and wound construction. This study received Institutional Review Board (IRB) approval from the Western IRB (Protocol #20204061) and conformed to tenets of the Declaration of Helsinki.

Figure 1.:

Photograph illustrating the NGENUITY 3D Visualization System affixed to the OPMI LUMERA T operating microscope.

Using conservative estimates of a visual recovery rate from postoperative day 1 to final at 80% ± 20% while implementing 3D digital visualization compared with a 55% recovery rate in the medical control treatment alone group, while using a 2 independent sample study with a dichotomous endpoint, 1:1 enrollment ratio, alpha/beta of 0.05, and power of 95%, it was calculated that the minimum number of eyes needed to enroll as 179 to provide statistically significant data. Accounting for a 15% drop out in the study, the minimum number needed to enroll was estimated in 210 eyes. Inclusionary criteria consisted of a diagnosis of mild to moderate cataract with surgery performed either in association with or without the femtosecond laser within the study period cited above. All consecutive cases of cataract surgery were included with the exception of cases associated with any of the exclusionary criteria: (1) planned complex or combined surgery (eg, surgery requiring additional intraocular manipulation as a result of the use of pupil expansion devices, minimally invasive glaucoma surgery, etc.); (2) any preoperative evidence of corneal dystrophy; (3) other corneal pathologies (corneal scarring, etc.) limiting accuracy and penetrance of biometry measurements; (4) previous intraocular surgery requiring microscopic or endoscopic illumination, excluding intravitreal injections and laser procedures; (5) adjunctive use of intraoperative aberrometry because the use of intraoperative aberrometry requires extensive manipulation of microscope light intensity, as well as additional operative time (thereby excluding patients receiving toric and presbyopia-correcting intraocular lenses from this study, resulting in the exclusive use of monofocal aspheric single-piece intraocular lenses); and (6) preoperatively planned myopic refractive target. These additional elements would have introduced confounding variables with potential for impacting light intensity, light exposure time, and postoperative visual acuity analysis.

Data Collection

Assessments of surgical coaxial light exposure time, light intensity (%), cumulative dissipated energy (CDE), and femtosecond laser utilization were recorded at the time of surgery. The light intensity, recorded as a percentage of maximum intensity illumination on the operating microscope, was set at the beginning of the surgery, recorded, and not changed for the duration of the case. Development of any intraoperative surgical complications (defined as development of vitreous loss and/or inability to place an intraocular lens in the capsular bag) was recorded.

Information regarding perioperative course was obtained from a medical chart review. For each patient, note was made of preoperative corrected distance visual acuity (CDVA), spectral domain optical coherence tomography (SD-OCT) findings, prior retinal diagnoses, and medically diagnosed diabetes with the most recent hemoglobin A1c (HgA1c). In addition, the uncorrected distance visual acuity (UDVA) was checked on postoperative day 1 (POD1), postoperative week 1 (POW1), and postoperative month 1 (POM1), with SD-OCT's performed at POW1 and POM1 time points. All SD-OCT images were read by a third-party retina specialist (EM) and analyzed for the presence of CME and retinal central subfield mean thickening.

A subanalysis was performed on a group of 71 eyes (of the total 214 eyes) in whom a clinical diagnosis of diabetes was confirmed by their medical doctor with a confirmatory HgA1c ≥6.5% (48 mmol/mol). These eyes were compared with the 143 nondiabetic eyes in the study and evaluated for the incidence of CME and retinal central subfield mean thickening.

Statistical Considerations

Differences between patients in the 2 arms of the study were analyzed by appropriate statistical methods. The chi-square (χ2) test was used for analyses involving exclusively nominal variables (eg, association between visualization method and age). Analyses involving continuous variables (eg, association between visualization method and light intensity) used either the t test or the Mann-Whitney U test, depending on the outcome of the Kolmogorov-Smirnov test for normality of distribution. A 2-way analysis of variance (ANOVA) was performed in the analysis of postoperative visual acuity improvement on the basis of both mean light intensity and visualization method. Post hoc Bonferroni analyses were used to help to interpret multiplicity with multiple secondary or post hoc endpoints. P values less than or equal to 0.05 were considered statistically significant.

RESULTS

Overall, 214 eyes (132 patients) met the inclusionary/exclusionary criteria and were analyzed. There were 126 eyes of women and 88 eyes of men (t test, P = .009). Of all eyes enrolled, 54% had the right eye treated (n = 115), 46% had the left eye treated (n = 99), and 62% of the patients enrolled had sequential bilateral treatment. The mean age of all patients was 69.8 ± 9.0 years (range 40 to 93 years). Postoperative CME was noted on the SD-OCT of 13 patients, at a rate of 6.1% for all patients enrolled in the study.

Cataract surgery on 119 eyes was performed with use of the Zeiss OPMI Lumera T microscope alone, comprising the traditional group, and 95 eyes with use of the Alcon NGENUITY 3D Visualization System affixed to the OPMI Lumera T microscope in the digital group. In keeping with the previous pilot study, the light intensity selected for each case was such that the surgery could be performed as safely as possible with an adequate red reflex, etc. No significant difference was found between the 2 different surgical visualization groups regarding age (t test, P = .99), diabetic status (χ2 test, P = .16), preoperative CDVA (t test, P = .87), macular degeneration status (wet: n = 2, χ2 test, P = .873; dry: n = 29, χ2 test, P = .65), laterality of the surgical eye (χ2 test, P = .91), femtosecond laser use (χ2 test, P = .78), or CDE (t test, P = .1) (Table 1). There were 2 intraoperative complications of anterior capsular tears that all occurred in the traditional group with no postoperative complications in either group.

Table 1. - Distribution of preoperative and perioperative characteristics for the traditional and digital visualization groups Characteristic Traditional group (n = 118) Digital group (n = 96) P valuea Age (y) 69.8 69.8 .99 Diabetic status, % (n) 37 (44) 28.1 (27) .16 Macular degeneration status, % (n)  Dry 11.9 (14) 15.6 (15) .42  Wet 0.8 (1) 1 (1) .88 Eye, % (n)  Right 53.4 (63) 54.2 (52) .91  Left 46.6 (55) 45.8 (44) Femtosecond laser use, % (n) 37.3 (44) 35.4 (34) .78 CDE 5.8 5.2 .10 Light exposure time (min) 12.9 12.8 .78 Light intensity, % 49.2 19.5 .00 Postop CME, % (n) 9.3 (11) 2.1 (2) .03

CDE = cumulative dissipated energy; CME = cystoid macular edema

aChi-square test for categorical variables and t test for continuous variables

The light exposure time was 12.9 ± 0.3 minutes in the traditional group compared with 12.8 ± 0.2 minutes in the digital group, showing no statistical difference in times between the 2 groups (t test, P = .78). However, there was a highly significant lower light intensity used throughout the duration of each surgery in the digital group (19.5% ± 0.5%; range: 5% to 26%) compared with the traditional group (49.2% ± 0.6%; range: 40% to 80%; t test, P < .01) (Table 1). Analysis of decimal postoperative uncorrected visual acuities in the 2 surgical visualization groups revealed a superior POD1 UDVA of 0.60 ± 0.02 in the digital group compared with 0.51 ± 0.03 in the traditional group (t test, P = .03); meanwhile, POW1 and POM1 UDVA showed no difference between the 2 groups (0.68 ± 0.03 and 0.68 ± 0.03, t test P = .92; and 0.72 ± 0.03 and 0.68 ± 0.03, t test, P = .27, respectively) (Figure 2). CME was noted on the SD-OCT of 9.2% of patients in the traditional group and only 2.1% in the digital group (χ2 test, P = .03) (Table 1).

Figure 2.:

Visual acuity (noted in decimal representation). (a) Vertical bar graph depicting mean visual acuity of all participants in the study across each of the 4 time points (preoperative, POD, POW1, and POM1). (b) Horizontal bar graph comparing the visual acuity data between the 2 different visualization groups (traditional vs digital) at each time point. POD1 = postoperative day 1; POM1 = postoperative month 1; POW1 = postoperative week 1

A subanalysis of 71 eyes in whom a clinical diagnosis of diabetes was confirmed by their medical doctor, where a confirmatory HgA1c ≥6.5% (48 mmol/mol) was performed and compared with the 143 nondiabetic eyes. No significant differences were found between the 2 groups regarding light intensity (t test, P = .17), CDE (t test, P = .94), or femtosecond laser use (χ2 test, P = .24). Decimal POD1 UDVA was 0.50 ± 0.03 in the diabetic group and 0.58 ± 0.02 in the nondiabetic group (t test, P = .05), while a comparable statistical difference was noted on the decimal POM1 UDVA (0.63 ± 0.03, 0.73 ± 0.02, t test P = .01). The rate of SD-OCT diagnosed CME was 6-fold higher in the diabetic group compared with the nondiabetic group (12.7% vs 2.1%, χ2 test, P = .002). Given that both the visualization method and diabetic status were not independent of CME findings, the population was divided into 4 subgroups. Within the nondiabetic group, the rate of CME was noted at 2.7% in the traditional subgroup and 1.4% in the digital subgroup; meanwhile, within the diabetic group, the rate of CME was noted at 20.5% in the traditional subgroup and 3.7% in the digital subgroup (2-way ANOVA, P = .01) (Figure 3). The occurrence of retinal central subfield mean thickening of greater than 10 μm mirrored the results of CME (9.6%, 1.5%, 38.1%, and 11.5%, respectively).

Figure 3.:

Incidence of CME. Vertical bar graph depicting the incidence of CME by the type of surgical visualization (traditional vs digital) and diabetic status of the patient (diabetic vs nondiabetic). CME = cystoid macular edema

DISCUSSION

Microscope manufacturers and ophthalmic surgeons have made concerted efforts to decrease the risk for microscope light-induced toxicity since its derivation.30–33 Today's fiberoptic illumination systems offer a more homogeneous illumination at the level of the retina in contrast to older incandescent lamp systems that produced nonuniform illumination with local “hot spots.”30 Longer wavelengths, operating microscope tilt, and opaque “eclipse” filters represent mitigation strategies developed to protect the macula over the last several decades.31–33 However, when referencing the microscope user manuals, the consistent recommendation has been for the use of the lowest light intensity possible for the shortest period of time while still maintaining optimal visualization.7,17,34 The manual for the OPMI Lumera T microscope, based on published data, advises use of the xenon light at a 50% light intensity percentage for a total of only 4.2 minutes (red reflex illumination) and 1.3 minutes (with surrounding field illumination); meanwhile, at a 25% light intensity, the manual recommends a maximum dose of 8.2 minutes and 2.5 minutes, respectively.30 Of note, there are no published recommendations for light intensity percentage below 25% conceivably because there has been no precedent operating safely beneath that threshold. Nevertheless, it remains significant that these recommended safe exposure times are substantially shorter than the length of time required for a typical cataract surgery.

Through the advent and implementation of 3D high-definition digital surgical visualization systems, surgeons have begun leveraging image-enhancing software that permits the use of significantly lower illumination levels without compromise in visualization. In our previous 2021 pilot study, we found that cataract surgery could be performed equally safely and timely in both visualization groups, with the digital group requiring up to 4-fold less light intensity than in the traditional group.19 Furthermore, there was an association noted between faster postoperative visual acuity recovery and lower light intensity used at the time of surgery. However, the study was a retrospective one and consisted of a smaller cohort of patients (51 eyes) than this study.

In strong agreement with our previous findings and other published literature, this prospective randomized trial found comparable results regarding use of intraoperative digital microscopy and decreased need for high coaxial illumination light intensity levels; however, in this study, we used a different operative microscope, operative setting, mean duration of case, and real-world heterogeneity regarding femtosecond laser use.19–25 To our knowledge, this represents the first prospective, randomized trial evaluating possible confounding preoperative and intraoperative variables, such as surgical time, CDE, and femtosecond laser use regarding light intensity used. In addition, this study was uniquely powered to assess visual recovery after cataract surgery under the differing lighting conditions afforded by the digital group vs the traditional group. We found that eyes in the digital group yielded better POD1 uncorrected visual acuities when compared with those in the traditional group. One explanation may be that phototoxicity-associated macular changes in the immediate postoperative period are more likely to occur in the traditional group than in the digital group. By POW1, any variances between the 2 groups were negligible, and therefore, we cannot postulate on any long-term effects to suprathreshold light exposure limits (as determined by the OPMI Lumera T microscope manual).

CME is one of the most common reasons for unexpected visual loss after cataract surgery.35–37 Several studies have attempted to identify the risk factors associated with postoperative CME; however, the exact cause of these phenomena still remains unknown.35–40 Regardless, it has been generally accepted that uveitis, posterior capsular rupture, CDE, and iris manipulation increase the relative risk for the development of CME.39,40 Over the past several decades, diabetic status has also been identified as a risk factor, with a prevailing theory that the retina has a lower threshold for withstanding injury.38 Given that several studies have shown a strong correlation between diabetics and CME, we wished to evaluate the role light intensity plays in the development of the pathology within this patient population.38,41,42

Optical coherence tomography is routinely used to quantitate retinal changes after surgery and aide in the diagnoses of CME.36,38 Interestingly, CME rates after cataract surgery have been found to be as high as 19% using fluorescein angiography criteria and as low as 0.82% when assessed clinically.35–37,43 In our current study, we noted the overall rate of SD-OCT diagnosed CME to be 6.1% in all patients, which is consistent with previous OCT-based studies (7.5%).43 When stratified, we found a 4-fold higher rate of CME in the traditional group vs the digital group and a 6-fold higher rate in the diabetics vs the nondiabetics. When further apportioned into the 4 subgroups, we noted that lower light intensity decreased the CME incidence in both the diabetic and nondiabetic groups. Therefore, coupled with the visual acuity data, we postulate that decreased light intensity at the time of cataract surgery induced less retinal phototoxicity, with the largest net-effect appreciated in “at-risk” retinas (ie, diabetics).

Limitations to this study include the absence of a per case discrete light intensity measurement in lux; however, previous studies confirm a tight correlation between total light intensity percentage of the operating microscope and macular light exposure in lux.25 In addition, there are several other preexisting conditions which could yield a higher susceptibility to macular phototoxicity outside of diabetes, including unknown conditions, that have not been openly studied here. Finally, it is difficult to draw definitive conclusions regarding pathological and subclinical changes occurring at the level of the retina because histological examination of postoperative retinas would be required. In future studies, special attention should be paid to patients describing scotomas or negative dysphotopsias, as well as to those patients at possibly increased risk for retinal phototoxicity. In addition, newer and more sophisticated retinal diagnostic modalities such as fundus photography, fundus autofluorescence, and fluorescein angiography would be helpful in elucidating macular changes.

Interestingly, light intensity modulation with modern-day cataract surgery has not been postulated as a predisposing factor for CME formation since microscope modifications in the early 2000s, plausibly because there had been no alternative available for conducting a comparative study. 3D digital visualization systems have repeatedly demonstrated that it is possible for surgeons to perform cataract surgery safely and effectively using a fraction of the microscope light intensity traditionally used. Given that most surgeons are still routinely operating well above the maximum light exposure limit recommended by surgical microscope manufactures, any technological advances with a proven efficacy of positively impacting surgical outcomes warrant further examination.30 Therefore, it is our firm belief that surgeons should once again reconsider light to be analogous to a drug, and abide by the recommended dosing guidelines.WHAT WAS KNOWN Digital visualization systems provided a safe alternative to traditional analog microscope coaxial illumination systems while also allowing for comparable visualization at a fraction of the light intensity. There was an association between improved postoperative day 1 visual outcomes and lower surgical light intensity levels.

WHAT THIS PAPER ADDS High-definition 3D digital visualization systems deliver an improved postoperative day 1 UDVA, and thus, a faster rate of visual recovery when compared with traditional analog microscopes. Digital visualization systems reduce the rate of SD-OCT diagnosed cystoid macular edema and retinal central subfield mean thickening in both the diabetic and nondiabetic populations. Acknowledgments

Edward Marcus, MD, provided assistance with retinal imaging analysis. Mary Iaconis, COA assisted with all elements of research coordination.

REFERENCES 1. Barraquer JI. The history of the microscope in ocular surgery. J Microsurg 1980;1:288–299 2. Boldrey EE, Ho BT, Griffith RD. Retinal burns occurring at cataract extraction. Ophthalmology 1984;91:1297–1302 3. Brod RD, Barron BA, Suelflow JA, Franklin RM, Packer AJ. Phototoxic retinal damage during refractive surgery. Am J Ophthalmol 1986;102:121–123 4. Byrnes GA, Antoszyk AN, Mazur DO, Kao TC, Miller SA. Photic maculopathy after extracapsular cataract surgery. A prospective study. Ophthalmology 1992;99:731–737; discussion 737–738 5. Dogramaci M, Williams K, Lee E, Williamson TH. Foveal light exposure is increased at the time of removal of silicone oil with the potential for phototoxicity. Graefes Arch Clin Exp Ophthalmol 2013;251:35–39 6. Kay CN, Pavan PR, Burrows A, Martino AZ. Neurosensory retinal detachment of the macula in retinal phototoxicity documented by optical coherence tomography. Retin Cases Brief Rep 2010;4:143–145 7. Khwarg SG, Linstone FA, Daniels SA, Isenberg SJ, Hanscom TA, Geoghegan M, Straatsma BR. Incidence, risk factors, and morphology in operating microscope light retinopathy. Am J Ophthalmol 1987;103:255–263 8. Kramer T, Brown R, Lynch M, Sternberg P Jr, Buchek G, L'Hernault N, Grossniklaus HE. Molteno implants and operating microscope-induced retinal phototoxicity. A clinicopathologic report. Arch Ophthalmol 1991;109:379–383 9. Kweon EY, Ahn M, Lee DW, You IC, Kim MJ, Cho NC. Operating microscope light-induced phototoxic maculopathy after transscleral sutured posterior chamber intraocular lens implantation. Retina 2009;29:1491–1495 10. Lindquist TD, Grutzmacher RD, Gofman JD. Light-induced maculopathy. Potential for recovery. Arch Ophthalmol 1986;104:1641–1647 11. Long VW, Woodruff GH. Bilateral retinal phototoxic injury during cataract surgery in a child. J AAPOS 2004;8:278–279 12. McDonald HR, Harris MJ. Operating microscope-induced retinal phototoxicity during pars plana vitrectomy. Arch Ophthalmol 1988;106:521–523 13. McDonald HR, Irvine AR. Light-induced maculopathy from the operating microscope in extracapsular cataract extraction and intraocular lens implantation. Ophthalmology 1983;90:945–951 14. Robertson DM, Feldman RB. Photic retinopathy from the operating room microscope. Am J Ophthalmol 1986;101:561–569 15. Ross WH. Light-induced maculopathy. Am J Ophthalmol 1984;98:488–493 16. Kleinmann G, Hoffman P, Schechtman E, Pollack A. Microscope-induced retinal phototoxicity in cataract surgery of short duration. Ophthalmology 2002;109:334–338 17. Hochheimer BF, D'Anna SA, Calkins JL. Retinal damage from light. Am J Ophthalmol 1979;88:1039–1044 18. Arafat A, Dutton G, Wykes W. Subclinical operating microscope retinopathy: the use of static perimetry in its detection. Eye (Lond) 1994;8:467–472 19. Rosenberg ED, Nuzbrokh Y, Sippel KC. Efficacy of 3D digital visualization in minimizing coaxial illumination and phototoxic potential in cataract surgery: pilot study. J Cataract Refract Surg 2021;47:291–296 20. Qian Z, Wang H, Fan H, Lin D, Li W. Three-dimensional digital visualization of phacoemulsification and intraocular lens implantation. Indian J Ophthalmol 2019;67:341–343 21. Weinstock RJ, Diakonis VF, Schwartz AJ, Weinstock AJ. Heads-up cataract surgery: complication rates, surgical duration, and comparison with traditional microscopes. J Refract Surg 2019;35:318–322 22. Ohno H. Utility of three-dimensional heads-up surgery in cataract and minimally invasive glaucoma surgeries. Clin Ophthalmol 2019;13:2071–2073 23. Matsumoto CS, Shibuya M, Makita J, Shoji T, Ohno H, Shinoda K, Matsumoto H. Heads-up 3D surgery under low light intensity conditions: new high-sensitivity HD camera for ophthalmological microscopes. J Ophthalmol 2019;2019:5013463 24. Srinivasan S, Tripathi AB, Suryakumar R. Evolution of operating microscopes and development of 3D visualization systems for intraocular surgery. J Cataract Refract Surg 2023;49:988–995 25. Gualino V, Pierne K, Manassero A, Bruneau S, Couturier A, Tadayoni R. Comparing microscope light-associated glare and comfort between heads-up 3D digital and conventional microscopes in cataract surgery: a randomised, multicentre, single-blind, controlled trial. BMJ Open Ophthalmol 2023;8:e001272 26. Tan NE, Wortz BT, Rosenberg ED, Radcliffe NM, Gupta PK. Impact of heads-up display use on ophthalmologist productivity, wellness, and musculoskeletal symptoms: a survey study. J Curr Ophthalmol 2022;34:305–311 27. Tan NE, Wortz BT, Rosenberg ED, Radcliffe NM, Gupta PK. Digital survey assessment of factors associated with musculoskeletal complaints among US ophthalmologists. Clin Ophthalmol 2021;15:4865–4874 28. Muecke TP, Casson RJ. Three-dimensional heads-up display in cataract surgery: a review. Asia Pac J Ophthalmol (Phila) 2022;11:549–553 29. Wang K, Song F, Zhang L, Xu J, Zhong Y, Lu B, Yao K. Three-dimensional heads-up cataract surgery using femtosecond laser: efficiency, efficacy, safety, and medical education—a randomized clinical trial. Trans Vis Sci Tech 2021;10:4 30. Michael R, Wegener A. Estimation of safe exposure time from an ophthalmic operating microscope with regard to ultraviolet radiation and blue-light hazards to the eye. J Opt Soc Am A Opt Image Sci Vis 2004;21:1388–1392 31. Ham WT Jr, Mueller HA, Sliney DH. Retinal sensitivity to damage from short wavelength light. Nature 1976;260:153–155 32. Landry RJ, Miller SA, Byrnes GA. Study of filtered light on potential retinal photic hazards with operation microscopes used for ocular surgery. Appl Opt 2002;41:802–804 33. Brod RD, Olsen KR, Ball SF, Packer AJ. The site of operating microscope light-induced injury on the human retina. Am J Ophthalmol 1989;107:390–397 34. Irvine AR, Wood I, Morris BW. Retinal damage from the illumination of the operating microscope. An experimental study in pseudophakic monkeys. Arch Ophthalmol 1984;102:1358–1365 35. Chu CJ, Johnston RL, Buscombe C, Sallam AB, Mohamed Q, Yang YC; United Kingdom Pseudophakic Macular Edema Study Group. Risk factors and incidence of macular edema after cataract surgery: a database study of 81984 eyes. Ophthalmology 2016;123:316–323 36. –43. See additional references; supplemental digital content (https://links.lww.com/JRS/B63)