sign in sign up
<<
Home , Retinitis, Retinitis pigmentosa

Evaluating Structural Progression of Pigmentosa After Surgery

JOAQUIN O. DE ROJAS, KASPAR SCHUERCH, PRIYA M. MATHEWS, THIAGO CABRAL, ALBERT HAZAN, JANET SPARROW, STEPHEN H. TSANG, AND LEEJEE H. SUH

PURPOSE: To determine whether cataract surgery ac- ity in RP patients and that it does not seem to be celerates disease progression in (RP). associated with faster disease progression as measured DESIGN: Retrospective cohort study. using SDOCT. (Am J Ophthalmol 2017;180: METHODS: Seventy eyes of 40 patients with RP were 117–123. Ó 2017 Elsevier Inc. All rights reserved.) categorized as having had phacoemulsification with intra- ocular implantation vs no cataract surgery at a single tertiary-level institution. Spectral-domain optical coher- ETINITIS PIGMENTOSA (RP) IS A GENETICALLY HET- ence tomography (SDOCT) was used to measure the erogeneous inherited retinal dystrophy character- ellipsoid zone (EZ) width, which has been demonstrated R ized by a slow loss of rod followed by cone photoreceptors that leads to a progressive, functional loss to be a reliable marker of RP severity, at baseline and 1 throughout follow-up (median 768 days). RP progression of and visual field. Common complications in the course of RP are formation of , cystoid mac- was calculated as the loss of EZ width over time for all pa- 2 3 tients. Additional postoperative data were collected for ular edema (CME), and (ERM). the cataract surgery group, including preoperative and Cataracts are more commonly seen in patients with RP postoperative best-corrected visual acuity, incidence of compared to the general population, occur at an earlier , posterior capsular opacification, epireti- age, are more commonly of the posterior subcapsular cata- ract (PSC) type, tend to be more visually significant, and nal membrane, and intraocular lens subluxation. 4–6 RESULTS: Multivariable analysis including age, base- often present with significant glare. line EZ width, mode of inheritance, and cataract surgery Cataract surgery is frequently performed in RP patients, status showed that there was no significant difference in but that does not make it benign, as it has been postulated that cataract extraction and intraocular lens placement RP progression between the cataract surgery and control 7 groups (P [ .23). Mode of inheritance was associated may increase outer retinal atrophy through light damage and/or anterior chamber inflammation,8 as well as increase with RP progression, with autosomal recessive RP 6 m the likelihood or severity of CME formation and posterior progressing at 148 m/year and autosomal dominant 9 RP progressing at 91 mm/year (P [ .003). Visual acuity capsular opacification (PCO). RP patients may also have a higher prevalence of zonular instability, increasing the risk improved in almost all eyes that underwent surgery (17/ 10 19, 89%) and remained stable in remaining eyes (2/19, of intraocular lens dislocation during or after surgery. 11%). There was a high incidence of postsurgical poste- Despite its potential risks and downfalls in RP patients, cataract surgery has been shown to improve visual acuity rior capsular opacification (18/19, 95%). There were 5,6,9 no serious complications, such as lens subluxation or in these patients, although to varying degrees. . In order to measure the structural severity of RP accu- CONCLUSIONS: Our findings suggest that cataract sur- rately and precisely, special techniques using spectral- gery is a safe and effective means of improving visual acu- domain optical coherence tomography (SDOCT) are being used to measure and monitor ellipsoid zone (EZ) measurements (ie, photoreceptor inner segment/outer Supplemental Material available at AJO.com. segment junction layer measurements). The EZ line, visu- Accepted for publication May 30, 2017. From the Edward S Harkness Eye Institute (J.O.D., K.S., P.M.M., T.C., alized using this technique, is a known correlate of where 11 A.H., J.S., S.H.T., L.H.S.), Barbara & Donald Jonas & retinal function is partly or fully intact. The purpose of Regenerative Medicine Laboratory, and Bernard & Shirlee Brown our study is to determine whether RP progression is Laboratory, Department of (T.C., S.H.T.), and Department of Pathology & Cell Biology, Institute of Human altered by cataract surgery by comparing EZ line loss (ie, Nutrition, College of Physicians and Surgeons (S.H.T.), Columbia progression) in RP patients who underwent uncompli- University Medical Center, New York, New York; Department of cated cataract surgery with RP patients who did not. Ophthalmology, Federal University of Espı´rito Santo, Vitoria, Brazil (T.C.); and Department of Ophthalmology, Federal University of Sao Our study is the first to quantitatively measure the EZ Paulo, Sao Paulo, Brazil (T.C.). for this purpose. We will also look at potential complica- Inquiries to Leejee H. Suh, Associate Professor of Ophthalmology, tions and other postoperative outcomes to further investi- Edward S. Harkness Eye Institute, Columbia University Medical Center, 635 West 165th St, New York, NY 10032; e-mail: lhs2118@cumc. gate if cataract surgery is safe and effective in this special columbia.edu population of patients.

0002-9394/$36.00 © 2017 ELSEVIER INC.ALL RIGHTS RESERVED. 117 http://dx.doi.org/10.1016/j.ajo.2017.05.026 METHODS izontal single-line SDOCT (9 mm, ART, average of a min- imum of 50 scans) was acquired in high-resolution mode. SUBJECTS: This retrospective cohort study adhered to All images were acquired using a 30-degree lens. The hor- the tenets of the Declaration of Helsinki and was approved izontal scan through the fovea was used to evaluate the hor- by the Institutional Review Boards of the Harkness Eye izontal width of the residual EZ line using the measuring Institute and Columbia University. The study protocol tool (Heidelberg Eye Explorer, software version 1.9.10.0; adhered to the Health Insurance Portability and Account- Heidelberg Engineering, Heidelberg, Germany). This is ability Act. further described by Sujirakul and associates.1 The termi- All study subjects were diagnosed (or confirmed) with nation of the EZ was determined using a described proto- RP by the same clinician (S.H.T.), using the same criteria col.12 EZ width was measured in the horizontal line incorporating clinical history, examination, SDOCT from the nasal to temporal termination. This is SDOCT, and full-field electroretinogram results. This the most sensitive marker of RP progression when the EZ cohort of typical RP patients was followed in the electro- line falls under 3000 mm.1 The Figure depicts an example diagnostics clinic at Columbia University’s Harkness Eye of baseline and follow-up SDOCTs in 2 patients, with Institute for at least 2 visits. Owing to our international calculation of EZ lines displayed. referral system for RP, many patients were only seen for 1 visit and care was transferred back to the primary provider. STATISTICAL ANALYSES: The SDOCT scans at baseline Therefore, these patients did not return for a second visit and all follow-up visits were independently analyzed by 2 and were excluded from our study. No cases of unilateral authors (K.S. and T.C.). The authors were masked to the RP, X-linked RP in female patients, or paravenous RP patients’ cataract surgery status when interpreting the im- were included in the study. Patients who developed RP ages. Intraobserver and interobserver reliability was before 30 years of age were also excluded, as such cases controlled for by measuring the horizontal width of the are often part of a that does not exemplify typical EZ line (ie, the EZ width) at 2 different times several weeks RP progression. SDOCT was used to evaluate structural apart. The average of 4 EZ line measurements at each time disease progression and disease status at every time point. point (2 from each author) was used for this analysis. Test- Patients with advanced RP who exhibited no visible EZ retest reliability using this method has been previously line in the 30-degree field were excluded. Exclusionary calculated and verified.1 The baseline time point was criteria also included patients with SDOCT scans of poor deemed ‘‘day 0’’ and corresponded to the first recorded visit quality from which the EZ line could not be assessed; how- in the non–cataract surgery or control group, and it corre- ever, none of the patients who underwent cataract surgery sponded to the preoperative visit closest to the day of oper- were excluded for this reason. ation in the cataract surgery group. The baseline visits for The patients were divided into 2 groups: those who had the latter group all occurred fewer than 90 days prior to cataract surgery and those who did not. Patients who were the respective surgery date for each patient. All subsequent included in the cataract surgery group met the following follow-up visits were recorded in days from baseline exam- criteria: (1) all cataract surgeries were performed at our ination. Baseline characteristics for the RP patients who institution; (2) all cataracts were removed using similar, underwent cataract surgery vs those who did not were typical phacoemulsification technique; and (3) documen- compared with Student t test for continuous variables tation of the preoperative and follow-up visits were avail- and x2 test for categorical variables. The rate of RP progres- able for review. Patients were excluded if they had a sion was computed by first creating a scatterplot for each second procedure along with cataract surgery, such as glau- eye displaying the change in EZ line width over time. coma tube implantation or vitrectomy. All patients under- Univariable linear regression analysis was used for each going cataract surgery had baseline SDOCTs taken within eye to calculate the slope of the best-fit line (unit ¼ micro- 3 months before the date of surgery. Sixteen eyes were meters per day). The mean rates of progression in the RP implanted with SN60WF intraocular lenses (AcrySof IQ patients undergoing cataract surgery vs those with no surgi- Natural SN60WF; Alcon Laboratories, Fort Worth, Texas, cal intervention were compared using Student t test. The USA) and 3 were implanted with SA60AT lenses (Acrysof rate of RP progression was also compared with mode of in- Natural Single-Piece SA60AT; Alcon Laboratories, Fort heritance, sex, and disease severity using Student t test. Worth, Texas, USA), based on surgeon preference. Shapiro-Wilk test was used to assess normality of data dis- tribution for continuous variables. The association be- IMAGE ACQUISITION: SDOCT and fundus infrared tween age or baseline EZ line width and the rate of reflectance images were acquired with an 870-nm light progression was determined using a univariable linear source at a resolution of 1536 3 1536 pixels and field of regression model. Multivariable analyses were used for view of 30 degrees, using an automatic real-time registra- the covariables found to be significant. All statistical ana- tion program from the Spectralis HRA þ OCT device lyses were performed with STATA version 14 (STATA (Heidelberg Engineering, Heidelberg, Germany). The hor- Corp, College Station, Texas, USA).

118 AMERICAN JOURNAL OF OPHTHALMOLOGY AUGUST 2017 FIGURE. Retinitis pigmentosa (RP) progression as measured by serial ellipsoid zone (EZ) width measurements in 2 patients. (Left column) Control patient. Three sequential spectral-domain optical coherence tomography (SDOCT) horizontal scans were taken 1 year apart for a patient with RP (shown in order from top to bottom). (Right column) An RP patient who underwent cataract surgery. Images represent the baseline scan taken 1 month before cataract surgery (top) and SDOCT scans taken at 1 year postoperatively (middle) and 2 years postoperatively (bottom). Dashed lines indicate the widths of the EZ line for each patient at respective time points. Progressive reductions in the horizontal EZ line width across time are depicted as a shortening of the solid white lines for the control patient (Left column) and the patient who underwent cataract surgery (Right column).

RESULTS to 8 visits for all patients. The baseline best-corrected visual acuity (BCVA, logMAR) was higher in the cataract surgery A TOTAL OF 84 PATIENTS WITH TYPICAL RP SEEN IN THE group (0.15 vs 0.33, P < .01). There was a higher prevalence electrodiagnostics clinic at Columbia University’s Hark- of PSC in the patients who proceeded to have cataract sur- ness Eye Institute were initially included in the study. After gery (84% vs 27%, P < .01). There was no difference in the exclusion of patients based on the criteria detailed in the percentage of patients with ERM or CME at baseline (P > methods section, 40 patients and 70 eyes were included .05). The mean EZ width was higher in the control group for data analysis. (2857 mm, SD 1991 mm) compared with the cataract sur- Table 1 summarizes baseline demographics and clinical gery group (1869 mm, SD 1375 mm), although this did data for our patients. Of the 70 total eyes, 19 eyes had un- not reach statistical significance (P ¼ .051). dergone cataract surgery and 51 eyes did not have any sur- Table 2 summarizes the postoperative outcomes in the gical intervention. The cataract surgery group was older 19 eyes that underwent cataract surgery. There were 6 than the patients who did not have surgery (51 vs 44 years, eyes (32%) that showed CME on SDOCT; however, only P ¼ .04) and sex did not differ in the groups (P > .05). The 2 of the 6 cases were new after surgery. Eighteen eyes most common form of inheritance was autosomal recessive (95%) developed PCO during the follow-up period. The RP (ARRP) (n ¼ 50, 71%), followed by autosomal domi- average final BCVA (logMAR) after surgery was 0.19 nant (ADRP) (n ¼ 18, 26%), and lastly X-linked RP (standard deviation [SD] 0.17), which represented an (XLRP) (n ¼ 2, 3%). The distribution by inheritance average improvement of 0.14 (SD 0.09) from baseline. pattern (autosomal dominant, autosomal recessive, X- No eyes had worsened BCVA after surgery. There were linked recessive) was similar between cataract surgery and no cases of vitreous prolapse requiring vitrectomy, dropped control groups. Three patients (5 eyes) with a diagnosis of nucleus, IOL dislocation, or endophthalmitis. were included in the autosomal recessive As mentioned in the first paragraph of this section, the group for purposes of subsequent data analysis. The baseline EZ width was not different between groups (P ¼ follow-up time was shorter in the cataract surgery group .051). A histogram was created for each group, which (median 259 days, range 105–916 days) compared with showed that baseline EZ width was highly positively skewed the non-surgery group (median 1007 days, range 98– (nonnormality of distribution confirmed by Shapiro-Wilk 2210 days), and the frequency of follow-up ranged from 2 test, P < .001). We identified 8 outliers, all in the

VOL. 180 RETINITIS PIGMENTOSA PROGRESSION AFTER CATARACT SURGERY 119 TABLE 1. Baseline Characteristics of Retinitis Pigmentosa Patient Eyes in a Retrospective Cohort Studya

All (N ¼ 70)b Cataract Surgery (N ¼ 19) No Cataract Surgery (N ¼ 51)b P Valuec

Age in years, mean 6 SD 46 6 12 51 6 13 44 6 11 .04 Men 40 (57%) 13 (68%) 27 (53%) - Inheritance pattern - Autosomal dominant 18 (26%) 4 (21%) 14 (27%) Autosomal recessive 50 (71%) 15 (79%) 35 (69%) X-linked recessive 2 (3%) 0 (0%) 2 (4%) Duration of follow-up (days), median (range) 768 (98–2210) 259 (105–916) 1007 (98–2210) - BCVA (logMAR), mean 6 SD 0.20 6 0.19 0.33 6 0.20 0.15 6 0.15 <.01 Presence of PSC 29 (43%) 16 (84%) 13 (27%) <.01 Presence of ERM 17 (24%) 4 (21%) 13 (25%) - Presence of CME 15 (22%) 4 (21%) 11 (23%) - EZ width, mean 6 SD (mm) 2588 (1887) 1869 (1375) 2857 (1991) -

BCVA ¼ best-corrected visual acuity; CME ¼ cystoid macular edema; ERM ¼ epiretinal membrane; EZ ¼ ellipsoid zone; PSC ¼ posterior subcapsular cataract. Results are n (%) unless otherwise indicated. aBaseline prevalence values listed (BCVA, PSC, ERM, CME, EZ width) include those obtained at the first visit in the ‘‘No cataract surgery’’ group or within 3 months prior to surgery in the ‘‘Cataract surgery’’ group. bThree patients in the ‘‘No cataract surgery’’ group were excluded from BCVA, PSC, and CME analyses as these data were missing for these patients. cP values are nonsignificant (>.05) unless otherwise indicated.

non-surgery group, with baseline EZ widths greater than (0.42 vs 0.33 mm/day or 155 vs 120 mm/year, P ¼ 6000 mm (all other eyes had baseline widths below .074). The rates of progression were found to vary by 5000 mm). Subsequent statistical analyses regarding EZ mode of inheritance when ADRP was compared with width measurements were performed excluding these out- ARRP (Table 3). For purposes of statistical analysis, the liers, as it has been shown that eyes with large baseline EZ X-linked recessive eyes were excluded (n ¼ 2). Baseline widths have more variable rates of progression, which could EZ width was not statistically different between ARRP potentially confound comparisons between groups.1 With and ADRP groups (2325 mm vs 1958 mm, respectively, the outliers excluded, the mean baseline EZ of the non- P ¼ .28). The ARRP group had a significantly higher surgery group (n ¼ 43) was 2128 mm (SD 1095 mm), which rate of RP progression than the ADRP group (0.41 was, again, not statistically different from the cataract sur- vs 0.25 mm/day, or 148 vs 91 mm/year, P ¼ .005). Multi- gery group (n ¼ 19) (1869 mm; SD 1375 mm; P ¼ .43). variable analysis including age, baseline EZ width, mode of Owing to a nonnormal distribution of the baseline EZ inheritance, and cataract surgery status showed that mode width despite exclusion of outliers (Shapiro-Wilk P value of inheritance was independently associated with rate of <.0001), the logarithm of the measurement was used for progression (P ¼ .003). Cataract surgery was still not asso- the linear regression analysis (Shapiro-Wilk P value ciated with RP progression in this analysis (P ¼ .23). >.05). The log of the baseline EZ value was not associated Associations between rate of progression and other cova- with rate of RP progression (micrometers per day) riates were also analyzed. Age was not associated with RP (b ¼0.045, 95% confidence interval [CI], 0.12 to progression (b ¼ 0.7 mm per 1 year increase in age at base- 0.035, P ¼ .26). Additional subgroup analyses were done line, 95% CI ¼0.005 to 0.002, P ¼ .43). Sex, presence of on eyes with ‘‘very severe RP,’’ defined as having a baseline CME at baseline, presence of ERM at baseline, and pres- EZ width less than 1000 mm, against eyes with ‘‘severe RP,’’ ence of PSC were also not associated with rate of RP pro- defined as having baseline EZ widths >_1000 mm. The rate of gression (P > .05). progression in the very severe RP group was not signifi- cantly different from that in the severe RP group (P ¼ .36). The improvement in vision after surgery in the very severe RP group was similar to that in the severe DISCUSSION RP group (P ¼ .37). The rate of RP progression in the cataract surgery group OUR STUDY PROVIDES EVIDENCE THAT THE NATURAL HIS- was not statistically different from the non-surgery group tory of structural RP progression, as measured by change

120 AMERICAN JOURNAL OF OPHTHALMOLOGY AUGUST 2017 TABLE 2. Outcomes and Complications After Cataract TABLE 3. A Comparison of Retinitis Pigmentosa Progression Surgery in Eyes of Patients With Retinitis Pigmentosa (N ¼ 19 Between Groups Using Rate of Ellipsoid Zone Loss Over Eyes) Time

Postoperative Outcome Total Eyes Rate of EZ Loss (mm/day), Mean (SD) P Valuec Postoperative cystoid macular edema,a n (%) 6 (32%) a Posterior capsular opacification, n (%) 18 (95%) All eyes (n ¼ 62) 0.36 (0.20) - Final BCVA (logMAR), mean 6 SD 0.19 6 0.17 Cataract surgery status - Change in final BCVAb (logMAR), mean 6 SD 0.14 6 0.09 Cataract surgery 0.42 (0.16) IOL dislocation, n (%) 0 (0%) No cataract surgery 0.33 (0.16) Endophthalmitis, n (%) 0 (0%) Mode of inheritance .005 Autosomal dominant 0.25 (0.16) BCVA ¼ best-corrected visual acuity; IOL ¼ intraocular lens. Autosomal recessive 0.41 (0.19) a Four out of 6 eyes were noted to have had preexisting cystoid Disease severity classified - macular edema prior to surgery. by baseline EZ widthb b Measured for each patient by subtracting baseline BCVA Very severe RP (<1000 mm) 0.37 (0.20) from BCVA at last follow-up visit. Negative value indicates Severe RP (>_1000 mm) 0.30 (0.20) improvement in BCVA after surgery. No patients had worsened BCVA after surgery. EZ ¼ ellipsoid zone. aResults displayed above are for patients with baseline EZ widths <_6000 mm. Eight patients from the original cohort were excluded based on this criterion. of EZ width over time, is unaffected by cataract surgery. bTen patients with baseline EZ width <1000 mm were classi- Cataract surgery improved visual acuity in all eyes in this fied as having ‘‘very severe RP.’’ The remaining patients (n ¼ group, except for 2 eyes in which visual acuity remained 52) were classified as having ‘‘severe RP.’’ c stable after surgery. Serious postoperative complications P values are nonsignificant (>.05) unless otherwise indicated. such as intraocular lens dislocation or endophthalmitis were not reported after cataract surgery in any patient. In accordance with previous findings,9,10 there was a high incidence of PCO postoperatively, but all cases were progressed at a significantly faster rate than the ADRP effectively treated with neodymium–yttrium-aluminum- group and at almost 60% faster progression (148 mm/year garnet laser posterior capsulotomy. Similarly, the 2 new vs 91 mm/year loss in EZ line, respectively). This relation- cases of postoperative CME were effectively treated with ship held true even after controlling for potential con- oral acetazolamide and vision improved reliably as a result. founders such as age, baseline EZ width, and cataract To our knowledge, our study is the first to characterize surgery status. Our previous study did not look at this rela- the rate of structural RP progression after cataract surgery tionship, as it only compared X-linked RP to ARRP pro- using SDOCT. We chose to concentrate on the outcome gression rates and found no significant difference between measurements of EZ line width and its change over time these groups (Supplemental Table 3 in Sujirakul and asso- because of its objective precision,1 good interrater and ciates1). The discrepancy in structural RP progression using intrarater reliability, and strong correlation with retinal SDOCT between ARRP and ADRP is a novel and inter- function in RP patients.12–15 We were concerned that esting finding that has not yet been reported but substanti- cataract surgery could potentially accelerate the rate of ates prior research that reports an association with outer retinal atrophy in RP patients by allowing more autosomal recessive inheritance and worse central visual light to reach the posterior pole and exert phototoxic acuity, as compared with autosomal dominant RP, which damage to the outer . This process has been is associated with a better visual prognosis.19 hypothesized to worsen outer retinal atrophy in an RHO Our previous study suggests that there is a relationship animal model.7 We were also concerned that between baseline EZ width and rate of RP progression. In anterior segment inflammation could increase photore- that study, eyes with baseline EZ widths >3000 mm ceptor damage in RP patients,8 but our results suggest progressed at a significantly faster rate than eyes with EZ that neither process had a significant effect on RP progres- widths <3000 mm.1 We could not replicate this finding sion. The average rate of RP progression among all patients in our current study because the majority of our patients was 0.36 mm/day, or 131 mm/year, which is almost iden- had more advanced RP with narrower EZ widths at base- tical to the rate of 130 mm/year published in our past line, the majority of which were less than 3000 mm. To study1 and comparable to rates quoted in a few previous investigate the relationship between baseline EZ width studies.16–18 and RP progression further, we recategorized cases into Mode of inheritance was one factor that did appear to in- ‘‘very severe RP,’’ defined as baseline EZ width fluence the rate of RP progression. The ARRP group <1000 mm, and ‘‘severe RP,’’ defined as baseline EZ width

VOL. 180 RETINITIS PIGMENTOSA PROGRESSION AFTER CATARACT SURGERY 121 >_1000 mm, but, despite this, we found no significant differ- postoperative period. However, when we reexamined the ence between these groups in terms of RP progression. SDOCTs of cases with greater than 2 years of postoperative Cataract surgery improved vision in 17 of 19 (89%) pa- follow-up, we did not see evidence of accelerated progres- tients and did not alter BCVA in the remaining 2 (11%) sion in these patients. Finally, there was a discrepancy in patients. No patient in our study had worsened visual acu- the total number of cases within groups, with more eyes ity after cataract surgery. A previous study by Yoshida and in the non-surgery group compared with the cataract sur- associates reported a >40% degree of improvement in vi- gery group. Retinitis pigmentosa is a rare disease, and sual acuity after cataract surgery,6 and another group re- even as a tertiary referral center we were not able to find ported a 77% degree of improvement.9 Postoperative and include more patients in the cataract surgery group BCVA may be more unpredictable in RP patients with that met our inclusion and exclusion criteria. Although poor preoperative vision,10 but all eyes in our study had pre- future, higher-powered studies are necessary to further operative BCVAs better than or equal to 20/200. test our hypothesis, our study is important in that it is the Unlike findings reported in 2 previous studies,5,6 EZ first to look at structural RP progression after cataract sur- integrity did not predict final visual acuity after surgery. gery and begins to fill a large gap in our knowledge. In Nakamura and associates,5 visual acuity did not reliably New tools in ophthalmology such as SDOCT are allow- improve in patients with preoperative EZs that were ing clinicians and scientists to better characterize diseases completely atrophied on OCT. We did not find such like RP with objective, structural measures that correlate severely progressed patients at baseline in our study group, well with visual function. These quantitative methods are and, thus, cannot comment on whether those few patients less time-consuming or less dependent on patient perfor- with no identifiable EZ line at baseline would benefit from mance than techniques such as full-field electroretinog- cataract surgery. raphy or visual field testing. Our study is the first to use Other potential limitations of our study include a shorter SDOCT to quantify EZ integrity across time in RP patients average follow-up time for the cataract surgery group after cataract surgery, and to determine that the normal compared with the non–cataract surgery group. This is a rate of structural progression in RP is not affected by cata- potential limitation because visual acuities may worsen in ract surgery but may be affected by mode of inheritance of RP patients several years after cataract surgery, or the RP. Our findings suggest that cataract surgery is safe and rate of RP progression may accelerate years later in the effective in RP patients, regardless of disease severity.

FUNDING/SUPPORT: THE JONAS CHILDREN’S VISION CARE AND BERNARD & SHIRLEE BROWN GLAUCOMA LABORATORY ARE supported by the National Institute of Health, New York, NY (5P30EY019007, R01EY018213, R01EY024698, 1R01EY026682, R21AG050437), National Cancer Institute Core (5P30CA013696), the Research to Prevent Blindness (RPB) Physician-Scientist Award, and unrestricted funds from RPB, New York, New York, USA. T.C. is supported by the ICO-Retina Research Foundation Helmerich Fellowships, sponsored by the Retina Research Foundation in cooperation with the ICOFoundation (International Council of Ophthalmology). S.H.T. is a member of the RD-CURE Consortium and is supported by the Tistou and Charlotte Kerstan Foundation, the Schneeweiss Stem Cell Fund, New York State (C029572), the Foundation Fighting Blindness New York Regional Research Center Grant (C-NY05-0705-0312), and the Gebroe Family Foundation. Financial Disclosures: The following authors have no finan- cial disclosures: Joaquin O. De Rojas, Kaspar Schuerch, Priya M. Mathews, Thiago Cabral, Albert Hazan, Janet Sparrow, Stephen H. Tsang, and Leejee H. Suh. All authors attest that they meet the current ICMJE criteria for authorship.

REFERENCES cataract surgery in patients with retinitis pigmentosa. Br J Ophthalmol 2015;99(4):508–511. 1. Sujirakul T, Lin MK, Duong J, Wei Y, Lopez-Pintado S, 6. Yoshida N, Ikeda Y, Murakami Y, et al. Factors affecting vi- Tsang SH. Multimodal imaging of central retinal disease pro- sual acuity after cataract surgery in patients with retinitis gression in a 2-year mean follow-up of retinitis pigmentosa. pigmentosa. Ophthalmology 2015;122(5):903–908. Am J Ophthalmol 2015;160(4):786–798.e784. 7. Cideciyan AV, Jacobson SG, Aleman TS, et al. In vivo dy- 2. Huckfeldt RM, Comander J. Management of cystoid macular namics of retinal injury and repair in the mutant edema in retinitis pigmentosa. Semin Ophthalmol 2017;32(1): dog model of human retinitis pigmentosa. Proc Natl Acad 43–51. Sci U S A 2005;102(14):5233–5238. 3. Chebil A, Touati S, Maamouri R, Kort F, El Matri L. 8. Nagasaka Y, Ito Y, Ueno S, Terasaki H. Increased aqueous Spectral domain optical coherence tomography findings flare is associated with thickening of inner retinal layers in in patients with retinitis pigmentosa. Tunis Med 2016; eyes with retinitis pigmentosa. Sci Rep 2016;6:33921. 94(4):265–271. 9. Jackson H, Garway-Heath D, Rosen P, Bird AC, Tuft SJ. 4. Berson EL, Rosner B, Sandberg MA, et al. Further evaluation Outcome of cataract surgery in patients with retinitis pigmen- of in patients with retinitis pigmentosa tosa. Br J Ophthalmol 2001;85(8):936–938. receiving treatment: subgroup analyses. Arch 10. Dikopf MS, Chow CC, Mieler WF, Tu EY. Cataract extrac- Ophthalmol 2004;122(9):1306–1314. tion outcomes and the prevalence of zonular insufficiency 5. Nakamura Y, Mitamura Y, Hagiwara A, et al. Relationship in retinitis pigmentosa. Am J Ophthalmol 2013;156(1): between retinal microstructures and visual acuity after 82–88.e82.

122 AMERICAN JOURNAL OF OPHTHALMOLOGY AUGUST 2017 11. Greenstein VC, Duncker T, Holopigian K, et al. Structural Usher syndrome. Invest Ophthalmol Vis Sci 2016;57(6): and functional changes associated with normal and abnormal 2428–2442. fundus autofluorescence in patients with retinitis pigmentosa. 16. Birch DG, Locke KG, Wen Y, Locke KI, Hoffman DR, Retina 2012;32(2):349–357. Hood DC. Spectral-domain optical coherence tomography 12. Ramachandran R, Cai XC, Lee D, et al. Reliability of a manual measures of outer segment layer progression in patients with procedure for marking the EZ endpoint location in patients X-linked retinitis pigmentosa. JAMA Ophthalmol 2013; with retinitis pigmentosa. Transl Vis Sci Technol 2016;5(3):6. 131(9):1143–1150. 13. Hariri AH, Zhang HY, Ho A, et al. Quantification of ellipsoid 17. Grover S, Fishman GA, Birch DG, Locke KG, Rosner B. zone changes in retinitis pigmentosa using en face spectral Variability of full-field electroretinogram responses in sub- domain-optical coherence tomography. JAMA Ophthalmol jects without diffuse disease. Ophthal- 2016;134(6):628–635. mology 2003;110(6):1159–1163. 14. Smith TB, Parker M, Steinkamp PN, et al. Structure-function 18. Seiple W, Clemens CJ, Greenstein VC, Carr RE, modeling of optical coherence tomography and standard auto- Holopigian K. Test-retest reliability of the multifocal electro- mated perimetry in the retina of patients with autosomal domi- retinogram and Humphrey visual fields in patients with reti- nant retinitis pigmentosa. PLoS One 2016;11(2):e0148022. nitis pigmentosa. Doc Ophthalmol 2004;109(3):255–272. 15. Sun LW, Johnson RD, Langlo CS, et al. Assessing 19. Fishman GA. Retinitis pigmentosa. Visual loss. Arch Ophthal- photoreceptor structure in retinitis pigmentosa and mol 1978;96(7):1185–1188.

VOL. 180 RETINITIS PIGMENTOSA PROGRESSION AFTER CATARACT SURGERY 123