348

ARTICLE Refractive outcomes of surgery in patients

Niranjan Manoharan, MD, Jennifer L. Patnaik, PhD, Levi N. Bonnell, MPH, Jeffrey R. SooHoo, MD, Mina B. Pantcheva, MD, Malik Y. Kahook, MD, Brandie D. Wagner, PhD, Anne M. Lynch, MD, Leonard K. Seibold, MD

Purpose: To evaluate refractive outcomes after phacoemulsifica- (P Z .0061) and 11.2% (P Z .0011) in the glaucoma group. Primary tion in patients with glaucoma. open-angle glaucoma (POAG) (n Z 154 ), chronic angle-closure glaucoma (n Z 18 eyes), and pseudoexfoliation glaucoma (n Z 23 Setting: University of Colorado Health Center, Aurora, Colo- eyes) had odds ratios of 1.90 (P Z .1760), 14.54 (P Z .0006), and rado, USA. 7.27 (P Z .0138), respectively, of refractive surprise greater than G1.0 D compared with patients without glaucoma. Refractive Design: Retrospective case series. surprise was noted more often in POAG eyes with axial lengths longer than 25.0 mm (P Z .0298). Glaucoma eyes had worse Methods: The incidence of refractive surprise was evaluated in mean postoperative corrected distance visual acuity than control patients with and without glaucoma after phacoemulsification cata- eyes (glaucoma: 0.1088 logarithm of the minimum angle of ract surgery. Refractive surprise was defined as the difference in resolution [logMAR]; controls: 0.0358 logMAR; P Z .01). spherical equivalent of the refractive target and postoperative refraction in diopters (D). Conclusion: Patients with a diagnosis of glaucoma were more likely to have a refractive surprise and/or worse visual outcome after Results: The study comprised 206 eyes in the glaucoma group phacoemulsification cataract surgery. and 1162 control eyes. The refractive surprise greater than G0.5 D and G1.0 D was 29.9% and 4.9% in the control group and 40.3% J Cataract Refract Surg 2018; 44:348–354 Q 2018 ASCRS and ESCRS

he leading causes of blindness worldwide are glau- Although these anatomic changes might be beneficial to- coma and .1 In the United States, the inci- ward aqueous outflow and IOP, it is unclear how they might T dences of glaucoma and cataracts are estimated to affect refractive outcomes, especially in patients with glau- increase over the coming decades because of the rapidly aging coma. As demand for precise refractive outcomes has population.2 Given that glaucoma and cataracts are both age- increased, there has been continuing research into opti- related diseases, studies have shown up to 20% of patients mizing refractive outcomes with newer formulas for intra- have concomitant disease3 and a significant proportion of ocular (IOL) power calculation.7 It remains unclear phacoemulsification cataract surgery is performed in patients what contribution a previous diagnosis of glaucoma has with glaucoma. The growing numbers of glaucoma patients on ocular biometry and refractive outcomes. In addition, having cataract surgery has led to many recent studies there is limited research on the effect of different types of – regarding a variety of outcome measures.4 6 glaucoma on cataract surgery refractive outcomes. Recent studies have shown phacoemulsification and The purpose of our study was to address these gaps in the manual small-incision cataract surgery can lead to signifi- literature by evaluating the refractive outcomes of patients cant (IOP) reductions after surgery.4 with glaucoma compared with patients without glaucoma Moreover, anterior chamber and angle anatomy have also in a cohort of patients from Colorado who had phacoemul- been shown to change after modern cataract surgery.4 sification cataract surgery.

Submitted: June 7, 2017 | Final revision submitted: December 2, 2017 | Accepted: December 19, 2017 From the Department of (Manoharan, Patnaik, Bonnell, SooHoo, Pantcheva, Kahook, Wagner, Lynch, Seibold), University of Colorado School of Medicine and the Department of Biostatistics and Informatics (Wagner), University of Colorado School of Public Health, Aurora, Colorado, USA. Presented at ASCRS Symposium on Cataract, IOL and Refractive Surgery, New Orleans, Louisiana, USA, May 2016. Corresponding author: Leonard K. Seibold, MD, Department of Ophthalmology, University of Colorado School of Medicine, 1675 Aurora Court, Mailstop F731, Aurora, Colorado 80045, USA. E-mail: [email protected].

Q 2018 ASCRS and ESCRS 0886-3350/$ - see frontmatter Published by Elsevier Inc. https://doi.org/10.1016/j.jcrs.2017.12.024 CATARACT OUTCOMES IN GLAUCOMA PATIENTS 349

PATIENTS AND METHODS no previous diagnosis of glaucoma. The mean age of pa- This retrospective cohort study was approved by the Colorado tients with and without glaucoma was 71.2 years G 12.5 Multiple Institutional Review Board. The University of Colorado (SD) and 68.9 G 10.1 years, respectively (P Z .0329). ’ Department of Ophthalmology s Cataract Outcomes Registry Table 1 and Table 2 show comparisons of the periopera- was used to identify patients for inclusion in this study. The Uni- versity of Colorado Eye Center is a tertiary care academic medical tive characteristics and refractive outcomes, respectively, center. In brief, every patient who has cataract surgery has a between patients with and without glaucoma. There was a comprehensive review of their medical record. The registry in- significantly higher frequency of both small and large cludes data on demographic information, medical history, preop- magnitude refractive surprise (OG0.5 diopter [D] and erative medication history, intraoperative and postoperative G1.0 D) and myopic surprise (OÀ1.0 D) in glaucoma pa- complications, and preoperative and postoperative eye examina- tients compared with patients without glaucoma. The pro- tions. The data are collected for up to 12 months postoperatively. OC Between January 2014 and June 2015, 1732 patients (2559 eyes) portion of patients with a hyperopic surprise ( 1.0 D) had phacoemulsification cataract surgery with IOL implantation. was also higher in the glaucoma group compared with pa- Intraocular power was calculated using partial coherence interfer- tients without glaucoma; however, this association did not ometry (IOLMaster 500, Carl Zeiss Meditec AG) supplemented reach statistical significance. with ultrasound immersion as needed. The formulas used were per surgeon preference and generally as follows: Hoffer Q8 was Table 3A and Table 3B show the odds ratio (OR) of the used for axial lengths (ALs) shorter than 23.0 mm, Holladay 19 subtypes of glaucoma for refractive surprise. In patients for ALs between 23.0 mm and 26.0 mm, and SRK/T10 for having phacoemulsification cataract surgery alone 26.0 mm and longer. A total of 1191 (46.5%) eyes were excluded (Table 3A), patients with a preoperative diagnosis of glau- for the following reasons: preexisting comorbid conditions (ie, coma had greater odds for refractive surprise than patients traumatic cataract, history of retinal detachment, severe age- ! related macular degeneration) (n Z 369); previous refractive sur- without glaucoma (P .01). Patients with POAG, ACG, gery (n Z 110); cataract surgery combined with non-glaucoma and PXG had higher ORs of refractive surprise than pa- related surgeries (ie, ) (n Z 102); complications during tients without glaucoma. These trends were also noted surgery (capsule tear, vitreous loss, trauma, retained lens frag- when eyes that had cataract surgery combined with MIGS Z ments, zonular dialysis, and choroidal hemorrhage) (n 19); and were included in the glaucoma groups (Table 3B). records that did not contain target refractions or postoperative Z Table 4 shows the OR of combination glaucoma surgeries manifest refractions (n 591) (Figure 1). G Among records excluded because of missing target or follow-up for refractive surprises greater than 1.0 D. The relation- refractions, 20.8% had glaucoma versus 15.1% for records ship of any combination of MIGS and phacoemulsification included. Eyes with a diagnosis of glaucoma of any kind were cataract surgery with refractive surprises was not statisti- included in the glaucoma group and patients with no history of cally significant. glaucoma were included in the control group. In addition, patients with glaucoma were categorized in the following subgroups: pri- Table 5 shows a comparison of risk factors between pa- mary open-angle glaucoma (POAG), chronic angle-closure glau- tients who had a refractive surprise versus patients who coma (ACG), pseudoexfoliation glaucoma (PXG), and “other had no refractive surprise in POAG. In POAG, eyes with glaucoma.” Several glaucoma patients had microinvasive glau- a refractive surprise after cataract surgery had a statistically coma surgery (MIGS) performed at the time of their phacoemul- significant higher frequency of AL longer than 25.0 mm sification cataract surgery. These patients were marked specifically for subgroup analysis to determine the effect of these additional than patients who had no refractive surprise. In chronic glaucoma procedures on refractive outcomes. ACG and PXG, none of the risk factors assessed were signif- The primary outcome measure was refractive surprise, defined as a icantly related to refractive surprise. difference in the spherical equivalent (SE) refractive target and post- No statistically significant difference was noted in history of operative SE. Postoperative refractions were taken from the 1-month filtration surgery (n Z 22) or type of surgery (cataract surgery to 1-year postoperative visits. Other variables included in the analysis were age, sex, preoperative IOP, IOP change (difference between pre- alone versus combination with MIGS) between patients with operative IOP and IOP at postoperative visit), AL, lens formula used, and without a refractive surprise in all glaucoma subtypes. mean keratometry (K), previous filtration surgery, preoperative and The mean preoperative IOP for patients with a history of filtra- postoperative corrected distance visual acuity (CDVA). tion surgery was 14.2 G 5.1 mm Hg. There was no significant difference in mean K between the glaucoma group and the Statistical Analysis control group (Table 1). Furthermore, mean K was not a Information was included from patients who had both unilateral risk factor for refractive surprise in the POAG (P Z .87), and bilateral cataract surgery. Two-sample t tests were used to chronic ACG (P Z .96), and PXG (P Z .37) subgroups. compare the mean age between patients with glaucoma and control In control patients, there was no statistically significant patients without glaucoma. All other comparisons were performed difference in the mean AL between patients with and using logistic regressions with general estimating equations to ac- G count for the intrasubject correlation. A P value less than 0.05 without a refractive surprise (mean 24.3 2.1 mm and was considered statistically significant. The statistical analysis was 24.0 G 1.4 mm, respectively, P Z .3229). For controls performed using SAS software (version 9.4, SAS Institute, Inc.). with refractive surprise, 15 ALs (26.8%) were shorter than 23.0 mm, 24 (42.9%) were between 23.0 mm and RESULTS 25.0 mm, and 17 (30.4%) were longer than 25.0 mm. For A total of 941 patients (1368 eyes) had cataract surgery be- controls without refractive surprise, 213 ALs (19.5%) were tween January 2014 and June 2015 and met study inclusion shorter than 23.0 mm, 663 (60.7%) were between criteria. Of these, 156 patients (206 eyes) had a diagnosis of 23.0 mm and 25.0 mm, and 216 (19.8%) were longer than glaucoma before surgery and 785 patients (1162 eyes) had 25.0 mm. Controls with longer ALs were more likely to

Volume 44 Issue 3 March 2018 350 CATARACT OUTCOMES IN GLAUCOMA PATIENTS

Figure 1. Consort diagram of study patients and eyes (ECP Z endocyclophotocoagulation).

have a refractive surprise (P Z .0236), although controls Our results show that patients with a preoperative diagnosis with shorter ALs were not more likely (P Z .0736). of glaucoma had increased odds for refractive surprise after Glaucoma eyes had slightly better preoperative logarithm phacoemulsification cataract surgery than patients without of the minimum angle of resolution (logMAR) corrected glaucoma. Given there are several types of glaucoma with distance visual acuity (CDVA) than control patients different anatomic characteristics, our study further evalu- (glaucoma: 0.3282 G 0.40 logMAR; controls: 0.3633 G 0.52 ated this higher incidence of refractive surprise in glaucoma logMAR; P Z .28), but this difference was not significant. patients by examining each major glaucoma subtype. Pri- However, glaucoma eyes had a statistically significantly worse mary open-angle glaucoma, chronic ACG, and PXG all postoperative logMAR CDVA than control eyes (glaucoma: showed higher odds of a refractive surprise when compared 0.1088 G 0.21 logMAR, controls: 0.0358 G 0.15 logMAR; with controls but only chronic ACG and PXG were statis- P Z .01). Patients with a reduced visual acuity (worse tically significant. postoperative CDVA than preoperative CDVA) were 10 In previous work analyzing refractive outcomes, Kim glaucoma eyes (4.8%) and 10 control eyes (0.9%), and this et al.11 compared refractive outcomes between chronic difference was statistically significant (P Z .0001). ACG and POAG and reported no significant difference be- tween the 2 groups. Their study noted that a large lens vault DISCUSSION (defined as distance from horizontal line connecting 2 To our knowledge, this study is the first comparison of scleral spurs and anterior lens pole) was a significant pre- refractive outcomes in patients with and without glaucoma. dictor of a poor refractive outcome. Seo et al.5 examined

Table 1. Perioperative characteristics of eyes with and without glaucoma. Glaucoma Group Control Group

Parameter Eyes, N Mean ± SD Eyes, N Mean ± SD P Value Preop IOP (mm Hg) 204 15.2 G 4.5 1158 14.5 G 2.9 .0823 Axial length (mm) 203 24.3 G 1.3 1148 24.0 G 1.4 .0611 Keratometry (D) 204 44.0 G 1.6 1155 44.2 G 1.5 .2243 IOP change (mm Hg) 171 1.5 G 3.9 828 1.6 G 3.0 .8024

IOP Z intraocular pressure

Volume 44 Issue 3 March 2018 CATARACT OUTCOMES IN GLAUCOMA PATIENTS 351

Table 2. Refractive outcomes of eyes with and without glaucoma. Number (%)

Parameter Glaucoma Group Control Group P Value Refractive surprise (OG1.0 D) 23 (11.2) 57 (4.9) .0011 Refractive surprise (OG0.5 D) 83 (40.3) 347 (29.9) .0061 Hyperopic surprise (OC1.0 D) 9 (4.4) 23 (2.0) .0633 Myopic surprise (!À1.0 D) 14 (6.8) 34 (2.9) .0082 the biometric factors associated with the accuracy of IOL Gung€ or€ et al.13 reported that phacoemulsification cataract power predictions for cataract surgery in POAG. Their re- surgery induced a more significant anterior chamber depth sults suggest that preoperative relative lens vault was a sig- change in patients with PXF syndrome (0.46 G 0.3 mm) nificant factor in predicting refractive outcomes after compared with control patients (0.12 G 0.1 mm). This cataract surgery in POAG patients. This difference in lens anatomic change after cataract surgery could explain the vault in chronic ACG eyes might explain the inaccuracy higher incidence of refractive surprise in our PXG cohort. of current IOL calculation formulas, as shown by our study. Myopia has long been understood to be a risk factor for A limitation in our study is that we did not use anterior open-angle glaucoma (OAG).14 Therefore, it is typical to segment optical coherence tomography to evaluate the pre- see longer ALs in OAG patients as in our study. The only operative lens vault. Further studies are necessary to under- statistically significant risk factor noted in our study to be stand this difference in lens vault in chronic ACG eyes to related to higher refractive surprises in OAG patients was improve IOL power calculations and refractive outcomes. longer AL. However, in controls with refractive surprise, Pseudoexfoliation glaucoma patients had an OR of 7.3 of there was also a higher incidence of eyes with ALs longer having a refractive surprise after phacoemulsification cata- than 25.0 mm. There was no higher rate of refractive sur- ract surgery compared with controls. To our knowledge, prise between IOL formulas, Holladay 19 or SRK/T,10 in only 1 study has evaluated refractive outcomes in pseudoex- the longer AL group. It is possible these refractive surprises foliation (PXF) patients. Ishikawa et al.12 compared refrac- in POAG as with controls are caused by IOL formula incon- tive outcomes between PXF patients and controls and sistencies with longer ALs. – found no difference in refractive outcomes at 1 month. Several previous studies15 18 found that a history of pre- However, PXF patients were not clearly defined and the vious filtration surgery is a risk factor for unpredictable study did not look specifically at PXG patients. It is possible phacoemulsification cataract surgery refractive outcomes. that our study found a difference because patients with PXG However, in our study, a previous filtration study did not are at a higher risk for refractive surprise than patients correlate with a higher incidence of refractive surprise. without glaucoma but with accumulation of PXF material. Muallem et al.19 similarly found no statistically significant

Table 3A. Odds ratio estimates for the subtypes of glaucoma for refractive surprise in eyes that had phacoemulsification cataract surgery alone. Patients

W/Refractive Surprise Glaucoma Subtype Total, N > ±1.0 D, N (%) OR 95% CI P Value All subtypes Yes 80 12 (15.0) 3.42 1.7, 6.9 .0006 No 1162 57 (4.9) Ref dd POAG Yes 56 5 (8.9) 1.90 0.7, 4.8 .1760 No 1162 57 (4.9) ddd Chronic ACG Yes 7 3 (42.9) 14.54 3.2, 66.8 .0006 No 1162 57 (4.9) ddd PXG Yes 11 3 (27.3) 7.27 1.5, 35.3 .0138 No 1162 57 (4.9) ddd Other glaucoma Yes 7 1 (14.3) 2.70 0.3, 22.9 .3609 No 1162 57 (4.9) ddd

ACG Z angle-closure glaucoma; CI Z confidence interval; No Z control group, patients without glaucoma; OR Z odds ratio; POAG Z primary open-angle glaucoma; PXG Z pseudoexfoliation glaucoma; Ref Z reference group

Volume 44 Issue 3 March 2018 352 CATARACT OUTCOMES IN GLAUCOMA PATIENTS

Table 3B. Odds ratio estimates for the subtypes of glaucoma for refractive surprise in eyes with phacoemulsification cataract surgery combined with MIGS. Patients

W/Refractive Surprise Glaucoma Subtype Total, N > ±1.0 D, N (%) OR 95% CI P Value All subtypes Yes 206 23 (11.2) 2.44 1.4, 4.2 .0011 No 1162 57 (4.9) Ref dd POAG Yes 154 12 (7.8) 1.64 0.9, 3.1 .1378 No 1162 57 (4.9) ddd Chronic ACG Yes 18 4 (22.2) 5.54 1.7, 18.0 .0045 No 1162 57 (4.9) ddd PXG Yes 23 5 (21.7) 5.38 1.7, 16.6 .0034 No 1162 57 (4.9) ddd Other glaucoma Yes 11 2 (18.2) 4.31 0.9, 20.6 .0671 No 1162 57 (4.9) ddd

ACG Z angle-closure glaucoma; CI Z confidence interval; No Z control group, patients without glaucoma; OR Z odds ratio; POAG Z primary open-angle glaucoma; PXG Z pseudoexfoliation glaucoma; Ref Z reference group difference in refractive surprise between eyes with particularly useful in reducing IOP and/or medication previous filtration surgery and controls. Lower pre- dependence in combination with already planned phaco- phacoemulsification IOP was, however, correlated with a emulsification cataract surgery for a visually significant myopic surprise after cataract surgery. Zhang et al.18 also cataract. Several recent publications have noted the effective showed in their study that after cataract surgery refractive IOP-lowering results of iStent, Trabectome, and endocyclo- – outcomes were more myopic in eyes photocoagulation (ECP).20 22 However, there is limited when compared with controls. Yeh et al.15 found that research studying the refractive outcomes of these combined eyes with trabeculectomy and IOP of 9 mm Hg or lower surgeries. Luebke et al.23 looked at refractive and visual out- had significantly more myopic surprises than the control comes after combined cataract and trabectome surgery. group. Eyes with a history of trabeculectomy and preoper- They noted no difference in refractive outcomes between ative IOP higher than 9 mm Hg did not have more myopic the study group with glaucoma patients who had trabectome errors than controls. The mean preoperative IOP of eyes combined with cataract surgery and control patients. They with previous filtration surgery in our study was higher also found no difference in significant biometry prediction than 9 mm Hg (mean 14.2 mm Hg) and therefore could errors or visual outcomes after combined surgery versus explain why these eyes did not have an increased incidence cataract alone. Sheybani et al.24 studied the effect of ECP of refractive surprise. on refractive outcomes when combined with cataract sur- Microinvasive glaucoma surgeries have gained traction in gery. Their group found a statistically significant myopic treating mild to moderate glaucoma. These surgeries are shift and variability in patients who had combined ECP-

Table 4. Univariate relationship of combined and refractive surprise between patients with glaucoma and patients without glaucoma. Patients

W/Refractive Surprise Surgery/Glaucoma Status Total, N > ±1.0 D, N (%) OR 95% CI P Value Phaco alone w/glaucoma 80 12 (15.0) 3.42 1.7, 6.9 .0006 Phaco alone w/o glaucoma 1162 57 (4.9) ddd Phaco plus microbypass stent w/glaucoma 25 3 (12.0) 2.64 0.7, 9.5 .1378 Phaco alone w/o glaucoma 1162 57 (4.9) ddd Phaco plus ECP w/glaucoma 81 6 (7.4) 1.55 0.6, 3.8 .3300 Phaco alone w/o glaucoma 1162 57 (4.9) ddd Phaco plus microbypass stent plus ECP w/glaucoma 20 2 (10.0) 2.15 0.5, 9.2 .2991 Phaco alone w/o glaucoma 1162 57 (4.9) ddd

CI Z 95% confidence interval; ECP Z endocyclophotocoagulation; OR Z odds ratio; phaco Z phacoemulsification cataract surgery

Volume 44 Issue 3 March 2018 CATARACT OUTCOMES IN GLAUCOMA PATIENTS 353

Table 5. Analysis of risk factors for refractive surprise in POAG patients. Parameter Refractive Surprise (N Z 12) No Refractive Surprise (N Z 142) P Value Preoperative IOP (mm Hg) Mean G SD 15.3 G 2.5 15.0 G 4.1 .6471 Eyes, n 12 140 IOP change (mm Hg) Mean G SD 2.3 G 2.7 1.4 G 3.5 .3299 Eyes, n 9 118 Previous filtration surgery, n (%) 2 (16.7%) 17 (12.0%) .6633 Axial length (mm) Mean G SD 25.0 G 1.4 24.4 G 1.3 .1268 !23.0 mm, n (%) 1 (8.3%) 15 (10.7%) .7417 23.0–25.0 mm, n (%) 4 (33.3%) 88 (62.9%) Ref O25.0 mm, n (%) 7 (58.3%) 37 (26.4%) .0298 Keratometry (D) Mean G SD 44.0 G 1.8 43.9 G 1.5 .8671 Eyes, n 12 141

IOP Z intraocular pressure; Ref Z reference group cataract surgery versus cataract surgery alone. This vari- significantly more likely to be excluded than controls ability and myopic error was noted to consistently occur because of these missing data. This trend was typically across Hoffer Q,8 Holladay 1,9 and SRK/T10 formulas. Simi- attributable to different protocols in glaucoma clinics espe- larly, Wang et al.25 showed that phacoemulsification with cially in satellite clinics than to follow-up compliance or is- ECP had decreased predictability of postoperative refraction sues specific to the glaucoma cohort. It is possible this might and a small myopic shift when compared with phacoemul- have skewed the results, but we cannot determine how or in sification alone in ACG patients. The results of our study what way that effect might be. Another limitation of our contrast with these previous findings. Among glaucoma pa- study was the small sample sizes of individual glaucoma tients who had cataract surgery in combination with MIGS subtypes. (microbypass stent [iStent] and/or ECP), there was no sig- In the current era of phacoemulsification, clear corneal nificant increased risk for refractive surprise compared incisions and advanced IOL technology, there is a high with controls. However, the sample size of MIGS combina- expectation from patients to achieve optimal refractive out- tion surgeries was limited in our study. comes. Our study shows that glaucoma patients have higher In addition to a higher risk for refractive surprise, we odds of refractive surprise postoperatively compared to found that glaucoma patients had on average worse normal controls. Primary OAG eyes with long ALs, chronic postoperative CDVA compared with control patients. ACG patients, and PXG patients are at particular risk for This was despite glaucoma patients having similar this undesirable outcome. However, even with appropriate preoperative CDVA compared with controls. Lundstrom€ refractive correction, glaucoma patients can still have worse et al.26 noted that glaucoma patients had worse visual visual outcomes. It is important to further elucidate the ef- outcomes compared with patients without an ocular fect of glaucoma and its various subtypes on cataract sur- comorbidity after routine cataract surgery. They noted gery visual and refractive outcomes as well as the etiologic that 89.7% of glaucoma patients had a CDVA of 20/40 or factors. This knowledge will assist ophthalmologists in better compared with 98.2% of patients without any ocular making any necessary adjustments to their preoperative comorbidity. They also noted that 21.5% of glaucoma IOL calculations and further improve visual and refractive patients had worse postoperative CDVA than preoperative outcomes. CDVA compared with 10.0% of patients with no ocular comorbidities. In our study, more glaucoma patients had reduced visual acuity after cataract surgery (worse WHAT WAS KNOWN postoperative CDVA than preoperative CDVA). We also  There has been no concrete evidence from previous litera- excluded other confounders including eyes with surgical ture suggesting differences in refractive outcomes of pa- complications and/or with preexisting visually threatening tients with and without a previous diagnosis of glaucoma. ocular comorbidities. Therefore, it remains unclear why WHAT THIS PAPER ADDS glaucoma patients are at risk for having worse postoperative  Patients with a diagnosis of glaucoma were more likely than visual outcomes apart from their increased risk for patients without glaucoma to have worse refractive and vi- refractive surprise. sual outcomes after phacoemulsification cataract surgery. Our study was limited by its retrospective nature and  Patients with PXG and chronic ACG had higher odds for large proportion of excluded data because of missing target refractive surprise than patients without glaucoma. or postoperative refraction (23%). Glaucoma eyes were

Volume 44 Issue 3 March 2018 354 CATARACT OUTCOMES IN GLAUCOMA PATIENTS

REFERENCES 16. Ong C, Nongpiur M, Peter L, Perera SA. Combined approach to 1. The Eye Diseases Prevalence Research Group. Causes and prevalence of phacoemulsification and trabeculectomy results in less ideal refractive among adults in the United States. Arch Ophthalmol outcomes compared with the sequential approach. J Glaucoma – 2004; 122:477–485. Available at: http://www.ophed.com/system/files 2016; 25:e873 e878 /2010/01/causes-visual-impairment-adults-us-2207-2207.pdf. Accessed 17. Bae HW, Lee YH, Kim DW, Lee T, Hong S, Seong GJ, Kim CY. Effect February 2, 2018 of trabeculectomy on the accuracy of calculations in 2. Pascolini D, Mariotti SP. Global estimates of visual impairment: 2010. Br J patients with open-angle glaucoma. Clin Exp Ophthalmol 2016; – Ophthalmol 2012; 96:614–618 44:465 471 3. Tseng VL, Yu F, Lum F, Coleman AL. Risk of fractures following cataract 18. Zhang N, Tsai PL, Catoira-Boyle YP, Morgan LS, Hoop JS, Cantor LB, surgery in Medicare beneficiaries. JAMA 2012; 308:493–501. Available WuDunn D. The effect of prior trabeculectomy on refractive outcomes of – at: http://jama.jamanetwork.com/data/Journals/JAMA/24715/joc120058 cataract surgery. Am J Ophthalmol 2013; 155:858 863 _493_501.pdf. Accessed February 2, 2018 19. Muallem MS, Nelson GA, Osmanovic S, Quinones R, Viana M, Edward DP. 4. Sengupta S, Venkatesh R, Krishnamurthy P, Nath M, Mashruwala A, Predicted refraction versus refraction outcome in cataract surgery – Ramulu PY, Robin AL, Lee P. Intraocular pressure reduction after phaco- following trabeculectomy. J Glaucoma 2009; 18:284 287. Available at: emulsification versus manual small-incision cataract surgery; a randomized http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2729937/pdf/nihms-857 controlled trial. Ophthalmology 2016; 123:1695–1703. Available at: http: 84.pdf. Accessed February 2, 2018 //www.aaojournal.org/article/S0161-6420(16)30129-4/pdf. Accessed 20. Arriola-Villalobos P, Martínez-de-la-Casa JM, Díaz-Valle D, Fernandez- February 2, 2018 Perez C, García-Sanchez J, García-Feijoo J. Combined iStent trabecular 5. Seo S, Lee CE, Kim YK, Lee SY, Jeoung JW, Park KH. Factors affecting micro-bypass stent implantation and phacoemulsification for coexistent refractive outcome after cataract surgery in primary angle-closure glau- open-angle glaucoma and cataract: a long-term study. Br J Ophthalmol – coma. Clin Exp Ophthalmol 2016; 44:693–700. Available at: http://onlineli 2012; 96:645 649 brary.wiley.com/doi/10.1111/ceo.12762/epdf. Accessed February 2, 21. Yildirim Y, Kar T, Duzgun E, Sagdic SK, Ayata A, Unal MH. Evaluation 2018 of the long-term results of trabectome surgery. Int Ophthalmol 2016; – 6. Lee HS, Park JW, Park SW. Factors affecting refractive outcome after cata- 36:719 726 ’ ’ ract surgery in patients with a history of acute primary angle closure. Jpn J 22. Ezzouhairi S. [Evaluation de l efficacitedelECP a laser diode combinee a Ophthalmol 2014; 58:33–39 une chirurgie de cataracte dans le glaucome [Evaluation of the efficacy of 7. Kane JX, Van Heerden A, Atik A, Petsoglou C. Intraocular lens power for- diode laser endocyclophotocoagulation combined with cataract surgery – mula accuracy: comparison of 7 formulas. J Cataract Refract Surg 2016; in glaucoma]]. J Fr Ophtalmol 2015; 38:844 854 42:1490–1500 23. Luebke J, Boehringer D, Neuburger M, Anton A, Wecker T, Cakir B, 8. Hoffer KJ. The Hoffer Q formula: a comparison of theoretic and regression Reinhard T, Jordan JF. Refractive and visual outcomes after combined formulas. J Cataract Refract Surg 1993; 19:700–712; errata, 1994; 20:677; cataract and trabectome surgery. Graefes Arch Clin Exp Ophthalmol – 2007; 33:2–3 2015; 253:419 423 9. Holladay JT, Prager TC, Chandler TY, Musgrove KH, Lewis JW, Ruiz RS. A 24. Sheybani A, Saboori M, Kim JM, Gammon H, Lee AY, Bhorade AM. three-part system for refining intraocular lens power calculations. J Cataract Effect of endoscopic cyclophotocoagulation on refractive outcomes Refract Surg 1988; 14:17–24 when combined with cataract surgery. Can J Ophthalmol 2015; – 10. Retzlaff JA, Sanders DR, Kraff MC. Development of the SRK/T intraocular 50:197 201 € lens implant power calculation formula. J Cataract Refract Surg 1990; 25. Wang JC-C, Campos-Moller X, Shah M, Sheybani A, Ahmed II. Effect of en- 16:333–340; erratum, 528 docyclophotocoagulation on refractive outcomes in angle-closure eyes af- 11. Kim YC, Sung MS, Heo H, Park SW. Anterior segment configuration as a ter phacoemulsification and posterior chamber intraocular lens – predictive factor for refractive outcome after cataract surgery in patients implantation. J Cataract Refract Surg 2016; 42:132 137 € with glaucoma. BMC Ophthalmol 2016; 16:179. Available at: https: 26. Lundstrom M, Barry P, Henry Y, Rosen P, Stenevi U. Visual outcome of //www.ncbi.nlm.nih.gov/pmc/articles/PMC5070221/pdf/12886_2016_Artic cataract surgery; study from the European Registry of Quality Outcomes le_359.pdf. Accessed February 2, 2018 for Cataract and Refractive Surgery. J Cataract Refract Surg 2013; – 12. Ishikawa N, Hayashi Y, Miyamoto T, Saika S. Errors in the prediction of post- 39:673 679; erratum, 971 operative refraction following intraocular lens implantation in eyes with pseudoexfoliation syndrome. J Cataract Refract Surg 2013; 39:649–650 Disclosures: None of the authors has a financial or proprietary in- 13. Gung€ or€ SG, Akman A, Asena L, Aksoy M, Sezenoz€ AS. Changes in anterior terest in any material or method mentioned. chamber depth after phacoemulsification in pseudoexfoliative eyes and their effect on accuracy of intraocular lens power calculation. Turk J Oph- thalmol 2016; 46:255–258. Available at: https://www.ncbi.nlm.nih.gov /pmc/articles/PMC5177780/pdf/TJO-46-255.pdf. Accessed February 2, First author: 2018 Niranjan Manoharan, MD 14. Marcus MW, de Vries MM, Montolio FGJ, Jansonius NM. Myopia as a risk factor for open-angle glaucoma: a systematic review and meta-analysis. Department of Ophthalmology, University Ophthalmology 2011; 118:1989–1994.e2 of Colorado School of Medicine, Aurora, 15. Yeh OL, Bojikian KD, Slabaugh MA, Chen PP. Refractive outcome of cata- Colorado, USA ract surgery in eyes with prior trabeculectomy: risk factors for postoperative myopia. J Glaucoma 2017; 26:65–70

Volume 44 Issue 3 March 2018