Outcomes in resident-performed cataract surgeries with iris challenges: Results from the Perioperative Care for Intraocular Lens study

1

ARTICLE

Outcomes in resident-performed cataract surgeries with iris challenges: Results from the Perioperative Care for Intraocular Lens study Giannis A. Moustafa, MD, Durga S. Borkar, MD, K. Matthew McKay, MD, Emily A. Eton, MD, Nicole Koulisis, MD, Alice C. Lorch, MD, Carolyn E. Kloek, MD, the PCIOL Study Group

Purpose: To assess the outcomes of resident-performed cataract surgeries with iris challenges and to compare these outcomes with similar surgeries performed by attending surgeons. Setting: Massachusetts Eye and Ear Infirmary, Harvard Medical School, Boston, Massachusetts, USA. Design: Retrospective chart review. Methods: All cases of cataract extraction by phacoemulsification with intraocular lens implantation, performed by comprehensive ophthalmologists between January 1 and December 31, 2014, were reviewed. Cases with preoperative or intraoperative miosis, iris prolapse, and intraoperative floppy iris syndrome, were included for analysis. Visual outcomes and the rate of perioperative adverse events were compared between resident and attending surgeon cases. Factors predicting adverse events were also assessed.

Results: In total, 1931 eye cases of 1434 patients were reviewed, and 65 resident cases and 168 attending surgeon cases were included. The mean logarithm of the minimum angle of resolution

B

alancing patient safety and trainee participation in operative procedures is a critical topic for surgical residency programs. At present, cataract surgeries account for more than one third of all procedures performed by ophthalmology residents in the United States.A,1 The Accreditation Council for Graduate Medical Education (ACGME) ophthalmology case logsA show a continued rise in the number of resident-performed

corrected distance visual acuity was better in the resident group 1 month after surgery (0.051 G 0.10 [SD] versus 0.132 G 0.30, P Z .03); however, the difference was eliminated when controlling for macular disease. The mean operative time was 43.8 G 26.5 minutes and 30.9 G 12.6 minutes for cases performed by resident surgeons and attending surgeons, respectively (P .0001). Residents utilized supplemental pharmacologic dilation or retraction more frequently than attending surgeons (98% versus 87% of cases, P Z .008). The overall rate of adverse events was no different between residents and attending surgeons (P Z 0.16). Dense nuclear sclerosis predicted adverse events in cataract cases with iris challenges (adjusted odds ratio, 1.86; 95% confidence interval, 1.17–2.94; P Z .001).

Conclusion: Although requiring longer operative times and more surgical manipulation, residents who performed cataract surgeries with iris challenges achieved outcomes comparable to those performed by attending surgeons, and residents should be given the opportunity to operate on these eyes. J Cataract Refract Surg 2018; -:-–- Q 2018 ASCRS and ESCRS

phacoemulsification procedures from a mean of 143.8 cases per resident in 2010 to 186.4 in 2017. Many studies2–29 have described surgical curricula and outcomes of resident-performed and attending surgeonperformed cataract surgeries in different settings. Reported risk factors for surgical complications in residentperformed cataract surgeries include dense nuclear sclerosis and poor red reflex, zonular weakness, older patient age,

Submitted: May 11, 2018 | Final revision submitted: July 16, 2018 | Accepted: August 5, 2018 From the Department of Ophthalmology (Moustafa, Borkar, McKay, Lorch, Kloek), Massachusetts Eye and Ear Infirmary, Harvard Medical School, Boston, Massachusetts, Retina Service (Borkar), Wills Eye Hospital, Philadelphia, Pennsylvania, Harvard Medical School (Eton), Boston, Massachusetts, and University of Southern California Roski Eye Institute (Koulisis), Keck School of Medicine, Los Angeles, California, USA. Presented at the Association for Research in Vision and Ophthalmology annual meeting, Honolulu, Hawaii, USA, May 2018. Supported by a research grant from the American Society of Cataract and Refractive Surgery, Fairfax, Virginia, USA (Drs. Borkar and Kloek). Miriam J. Haviland, MSPH, reviewed the methodology and statistical analysis, and Demetrios G. Vavvas, MD, PhD, read and commented on the manuscript. Corresponding author: Carolyn E. Kloek, MD, Assistant Professor of Ophthalmology, Massachusetts Eye and Ear Infirmary, 243 Charles Street, Boston, Massachusetts 02114, USA. Email: [email protected]. Q 2018 ASCRS and ESCRS Published by Elsevier Inc.

0886-3350/$ - see frontmatter https://doi.org/10.1016/j.jcrs.2018.08.019

2

RESIDENT-PERFORMED CATARACT SURGERY WITH IRIS CHALLENGES

longer phacoemulsification time, and a history of vitrectomies.11,14,30,31 Small pupil size has been associated with a high rate of surgery-related complications for both residents and attending surgeons, such as vitreous loss, retained nuclear material, posterior capsule tear, and posterior capsule opacification.21,32–35 Encountering a small pupil is fairly common, with one study11 reporting an 11% incidence among cataract surgeries performed by residents. Intraoperative floppy iris syndrome (IFIS) is a well-described condition characterized by a combination of iris flaccidity and billowing, tendency of the iris to prolapse through the wounds, and progressive pupil constriction despite standard preventive interventions.36 It manifests intraoperatively in 0.9% to 2.5% of the general cataract surgery population and in 29.6% to 93% of patients using tamsulosin.20,36,37 Recent use of tamsulosin increases the risk for serious postoperative adverse events by more than 2-fold.38 A significantly higher rate of intraoperative complications is also observed in patients with previous tamsulosin use compared with patients who have never received a-antagonists, and this further increases in the presence of IFIS characteristics.39–41 Complications that have been described in patients with previous tamsulosin exposure and/or IFIS include iris trauma, corneal damage and endothelial cell loss, posterior capsule rupture, lost lens or lens fragment, and retinal detachment.38–40,42 Similar complications have been attributed to iris prolapse with or without IFIS.43,44 Pharmacologic agents such as atropine or bisulfate-free epinephrine–lidocaine (epi-Shugarcaine) and intracameral epinephrine, the use of high-viscosity retentive ophthalmic viscosurgical devices, mechanical iris retractors, and modification of surgical techniques are used intraoperatively to address these iris challenges. The potential for a technically challenging surgery and increased complication rate in patients with iris challenges might cause attending surgeons to hesitate involving residents in these cases. Given the surgical planning and technical skills required to manage iris challenges during cataract surgery coupled with the increasing number of patients who are treated with a-blockers, it is critical that ophthalmology residents have the opportunity to perform supervised surgery on this group of patients during training. At the same time, patient safety must be carefully considered. This study aimed to identify whether cataract surgeries with iris challenges performed by resident surgeons can achieve outcomes comparable to those performed by attending surgeons. Moreover, we sought to identify whether preoperative and intraoperative variables predict adverse events in these cases. MATERIALS AND METHODS Study Population and Case Selection The electronic medical records (EMRs) of patients who had cataract extraction with intraocular lens (IOL) implantation by phacoemulsification at the Comprehensive Ophthalmology Service, Massachusetts Eye and Ear Infirmary, Harvard Medical School, Boston, Massachusetts, USA, between January 1, 2014 to December 31, 2014, were reviewed. Institutional Review Board Volume - Issue - - 2018

approval for the retrospective review of the data was obtained through the Massachusetts Eye and Ear Human Studies Committee. A waiver of patient consent was granted given the retrospective nature of the study. The study was conducted in accordance with the tenets of the Declaration of Helsinki. Current Procedural Terminology codes 66982 (extracapsular cataract extraction with insertion of IOL prosthesis, complex) and 66984 (extracapsular cataract extraction with insertion of IOL prosthesis) were used to identify cases for review. A subsequent detailed chart review of the EMR was performed for each case. The resident ACGME surgical case logs were reviewed to determine which cases were performed primarily by resident surgeons. Cases that were logged as primary surgeries by the resident were considered resident-performed surgeries. Cases not logged as primary surgeries by the residents were considered attending surgeon-performed surgeries. According to the ACGME, a case should be logged as primary surgeon if “a resident is present for all of the critical portions, and must perform the majority of the critical portions of the procedure under appropriate faculty supervision.” In our institution, residents and faculty included in this study are educated about these guidelines and are asked to follow them when logging. In case of uncertainty about whether a case should be logged as resident or as primary surgeon, the resident and faculty member are asked to discuss the case together to reach a decision. Cases with inadequate pupil dilation or intraoperative miosis, iris prolapse, and IFIS, were selected for inclusion based on a detailed review of the operative report. Cases with synechiae were included in the study if a miotic pupil was described in the operative note. One case was excluded because of incomplete data. No exclusion criteria based on case baseline characteristics, intraoperative features and events, or outcomes were applied. Data Collection Chart reviews were conducted by 4 well-trained study personnel. Before initiating the data collection, each study personnel underwent training with a study investigator to review the standardized chart review methodology and the definitions of the variables intended to be reviewed. For each surgical case, the preoperative consultation closest to the date of surgery, operative report, postoperative day 1 visit, postoperative week 1 visit, and postoperative month 1 visit were reviewed and recorded. A visit qualified as a postoperative week 1 or a postoperative month 1 if it was performed between postoperative days 5 to 14 or postoperative weeks 3 to 8, respectively. Preoperative Data The preoperative data elements used for analysis were as follows: age, sex, race/ethnicity, operative eye, nuclear sclerosis, preoperative corrected distance visual acuity (CDVA), preoperative intraocular pressure (IOP), and the presence of macular disease. Nuclear sclerosis values for each case were derived by calculating the average of nuclear opalescence and nuclear color grading as described within the Lens Opacities Classification System III.45 Resident surgical logs were also reviewed to determine the previous number of cataract procedures each resident had performed up to the date of the procedure being examined. Operative Data Data abstracted from the operative report included whether surgery was performed by a resident or an attending surgeon, operative time as measured from incision to wound closure, posterior capsule tear, anterior capsule tear, zonular dehiscence, IOL placed in the sulcus or anterior chamber, and iris trauma. Only severe iris trauma described in the operative note, such as large iris or angle tears, multiple transillumination defects, and sphincter tears, were considered iris complications for analysis; temporary distortions or single transillumination defects were not included as iris complications. Postoperative Data The postoperative variables included were the IOP (at 1 day postoperatively), the CDVA (at 1 month

RESIDENT-PERFORMED CATARACT SURGERY WITH IRIS CHALLENGES

postoperatively), the presence of a retained lens fragment, corneal edema, a corneal epithelial defect, Descemet membrane tear, inflammation, eye pain (at 1 month postoperatively), a new retinal tear or detachment, a new epiretinal membrane, macular edema, infectious endophthalmitis, and need for referral to an ophthalmology specialist. For corneal edema, only cases characterized in the note as severe were included. Inflammation was deemed present for analysis if anterior chamber flare was graded 3C or more on either of the postoperative visits. Muhtaseb Score The Muhtaseb scoring system is a widely studied and established preoperative risk stratification formula for cataract surgery performed by phacoemulsification. This formula uses baseline preoperative characteristics of the patient to predict the risk for operative complications.46 Variables including refractive error, anterior chamber depth, the presence of corneal scarring, phacodonesis, pseudoexfoliation, posterior capsule plaque, posterior polar cataract, or a history of vitrectomies were collected to determine the Muhtaseb score for each case. Poor positioning was not considered for scoring.46 Outcome Variables Outcomes investigated included postopera-

tive uncorrected (UDVA) and corrected (CDVA) distance visual acuities at the postoperative month 1 examination, and adverse events occurring intraoperatively or up to 8 weeks after the procedure. Surgical Approach A standard combination of phenylephrine 2.5% and tropicamide 1% was administered to all patients preoperatively on the day of the surgery. Anesthesia was delivered by monitored anesthesia care and peribulbar block in all cases, except for 1 residentperformed case and 3 cases performed by attending surgeons, which were performed under topical anesthesia. All resident cataract surgeries were performed by postgraduate year 4 residents under direct attending surgeon supervision. The assignment of cases as a resident case or an attending surgeon case was based on the attending surgeon’s judgement of case difficulty and an estimation of the resident’s skills and previous experience. Both resident and attending surgeons used a clear corneal approach, except for one attending surgeon who primarily used a scleral tunnel incision; attending surgeons performed either a divide-and-conquer, stop-and-chop, or a quick-chop nuclear disassembly technique, depending on judgement and personal preference, whereas residents predominately used the former. Intraoperative interventions to mitigate small pupil size and IFIS included the use of iris hooks, placement of a Malyugin ring,47 and instillation of intracameral bisulfate-free epinephrine 0.025% and lidocaine 0.75% in a fortified balanced salt solution.48 Ophthalmic viscosurgical devices were used according to the surgeon’s discretion. No surgeries were combined with other ophthalmologic procedures, such as vitrectomy or glaucoma surgery. Statistical Analysis Data were analyzed using SPSS software (version 23.0, IBM Corp.). Continuous variables were expressed as means G SD. Categorical variables were summarized using frequencies and percentages. Snellen CDVA values were converted to the logarithm of the minimum angle of resolution (logMAR) notation for statistical analysis (counting fingers Z C2, hand motion Z C2.2, and light perception Z C2.6).49 Subsequently, logMAR values were converted back to the Snellen fraction equivalent for interpretation of the results. The variable describing patients experiencing at least one event versus no events was named “event rate” and was treated as a dichotomous variable. The Muhtaseb score was assigned values from 0 to 17 and was treated as a continuous variable. Nuclear sclerosis with values ranging from 0 to 6 was treated as a continuous variable. Descriptive statistics were calculated using simple crosstabulation. Differences in continuous variables

3

were evaluated using a Mann-Whitney U test. For categorical variables, Pearson chi-square or Fisher exact tests were performed. Linear-by-linear association was performed to test the association between ordinal variables and nonordered categorical variables. Binomial logistic regression analysis was performed to control for confounding factors when the outcome variable was dichotomous. For visual acuity, stratification was implemented to control for the presence of maculopathy. Potential risk factors for intraoperative and postoperative events were initially subject to univariate regression. Factors with a P value less than 0.10, as well as other factors of specific interest (age, sex, resident versus attending surgeon as the primary surgeon, previous number of cases performed by residents, presence or absence of IFIS), were incorporated into the multivariate models. A P value less than 0.05 was considered statistically significant, and all P values were 2-tailed. Missing data were few in our study (less than 6.8% for each variable tested). Cases with missing data were excluded from the analysis.

RESULTS The EMRs of 1931 cases (1434 patients) were reviewed. Of those, 234 cases (12.1%) manifested as a challenging iris intraoperatively. One case was excluded because whether the primary surgeon was a resident or an attending surgeon could not be confirmed. Sixty-five eyes of 61 patients operated on by 8 residents and 168 eyes of 138 patients operated on by 10 attending surgeons were included for analysis. All resident surgeons were in their third year of residency training with a mean 97.86 G 63.39 (range 18 to 233) of previous cataract surgeries performed by phacoemulsification. For attending surgeon-performed cases, the mean attending surgeon experience was 17.79 G 11.86 years (range 2 to 29 years) in practice at the time of surgery. Table 1 shows the demographics and preoperative characteristics between the cataract surgeries with challenging irises performed by resident and attending surgeons. The mean IOP at baseline was higher in resident-operated eyes than in eyes operated by attending surgeons (P Z .03). The Muhtaseb score was used to stratify cases according to difficulty. Case difficulty was higher in the group of surgeries performed by attending surgeons than by resident surgeons (P Z .008). Analyzing each risk indicator separately, only patient age older than 88 years (P Z .037) and the presence of pseudoexfoliation (P Z .009) demonstrated statistical significance between resident-performed and attending surgeon-performed cases. For the cataract surgeries with challenging irises, the difference in operative time between resident-performed procedures and those performed by attending surgeons was statistically significant (P .0001). Residents used additional pharmacologic agents, including bisulfate-free epinephrine–lidocaine, or mechanical pupillary dilation, including iris hooks and Malyugin rings, more frequently than attending surgeons (64 [98%] of 65 cases versus 146 [87%] of 168 cases, P Z .008). Table 2 shows a comparison of intraoperative characteristics between resident and attending surgeon cases. All 65 cases (100%) in the resident-treated group and 151 (89.6%) of the 168 cases in the group treated by attending surgeons achieved a 1-month postoperative CDVA of Volume - Issue - - 2018

4

RESIDENT-PERFORMED CATARACT SURGERY WITH IRIS CHALLENGES

Table 1. Comparison of patient demographic and preoperative characteristics of eyes undergoing phacoemulsification with challenging irises by resident versus attending surgeons. Variable Mean age (y) G SD Sex, n (%) Male Female Race/ethnicity, n (%) White Black/African American Asian Hispanic Other/undeclared Operative eye, n (%) Right Left Mean baseline CDVA (logMAR) G SD Mean baseline IOP (mm Hg) G SD Macular disease, n (%) Muhtaseb risk indicators, n (%) Age O88 years Myopia or hyperopia O6 D Brunescent/white/dense/total cataract/no fundus view Corneal scarring Phacodonesis Pseudoexfoliation Posterior capsule plaque Posterior polar cataract Previous PPV Anterior chamber depth !2.5 mm Small pupil Mean Muhtaseb score G SD

Residents (n Z 65 Cases)

Attending Surgeons (n Z 168 Cases)

73.14 G 7.62

72.91 G 10.97

33 (50.77) 32 (49.23)

102 (60.71) 66 (39.29)

43 (66.15) 11 (16.92) 5 (7.69) 0 (0.00) 6 (9.23)

137 (81.55) 15 (8.93) 13 (7.74) 1 (0.60) 2 (1.19)

33 (50.77) 32 (49.23) 0.48 G 0.54 15.87 G 3.46 5 (7.69)

87 (51.79) 81 (48.21) 0.44 G 0.46 14.69 G 2.91 27 (16.07)

.85 .03* .09

0 (0.00) 5 (7.69) 15 (23.08) 3 (4.62) 0 (0.00) 2 (3.08) 3 (4.62) 0 (0.00) 0 (0.00) 0 (0.00) 65 (100) 1.54 G 1.47

11 (6.55) 23 (13.69) 39 (23.21) 5 (2.98) 1 (0.60) 26 (15.48) 7 (4.17) 0 (0.00) 10 (5.95) 7 (4.17) 161 (95.8) 2.16 G 1.75

.037* .21 .95 .69 1.00 .009* 1.00 d .07 .20 .20 .008*

P Value .98 .17

.009*

.89

CDVA Z corrected distance visual acuity; IOP Z intraocular pressure; logMAR: logarithm of the minimum angle of resolution; PPV Z pars plana vitrectomy *Statistically significant

20/40 or better (P Z .007). Cases that reached a CDVA of 20/20 or better included 47 (72.3%) of the 65 residentperformed cases and 89 (53.0%) of 168 attending surgeon-performed cases (P Z .01). Seven cases (4.2%) in the attending surgeon group achieved a final postoperative CDVA of 20/100 or worse. These patients had significant underlying macular disease, such as macular hole with multiple past vitrectomies and wet age-related macular

degeneration. As a result, the CDVA at 1 month postoperatively was 20/23 Snellen and 20/27 Snellen in the resident and attending surgeon groups, respectively (P Z .03). Maculopathy-specific subgroup stratification showed no statistical difference in postoperative CDVA outcomes. The UDVA at 1 month postoperatively was identical for both residents and attending surgeons (20/43 Snellen, P Z .84). Table 3 shows a summary of visual outcomes.

Table 2. Comparison of intraoperative characteristics of eyes undergoing phacoemulsification with iris challenges by resident versus attending surgeons. Variable Iris challenge, n (%) Miosis Prolapse IFIS Mean operative time (min) G SD Supplemental dilation,†n (%)

Residents (n Z 65 Cases)

Attending Surgeons (n Z 168 Cases)

55 (84.62) 0 (0.00) 10 (15.38) 43.8 G 26.5 64 (98)

123 (73.21) 7 (4.17) 37 (22.02) 30.9 G 12.6 146 (87)

.11

IFIS Z intraoperative floppy iris syndrome *Statistically significant † Pharmacologic dilation with bisulfate-free epinephrine–lidocaine or mechanical iris retraction by iris hooks or Malyugin ring

Volume - Issue - - 2018

P Value

!.0001* .008*

5

RESIDENT-PERFORMED CATARACT SURGERY WITH IRIS CHALLENGES

Table 4 shows the frequency of intraoperative and postoperative adverse events in the cohort. The adverse event rate was significantly associated with the Muhtaseb score, as shown by linear-by-linear association (P ! .05). In the resident-performed surgeries, the event rate was not associated with the residents’ previous cataract surgery volume (Mann-Whitney U test, P Z .42). Forty-one (63.1%) of 65 resident-performed cases and 89 (53.0%) of 168 attending surgeon-performed cases experienced an event (P Z .16). The rate of adverse events was not statistically different between the 2 groups, even when adjusted for case difficulty (P Z .074). Among the 40 resident-performed and 89 attending surgeon-performed surgeries associated with events, 13 (32.5%) and 36 (40.4%) of cases experienced 2 or more adverse events, respectively (P Z .10). In the resident-performed and attending surgeon–performed groups, the mean Muhtaseb scores for patients experiencing 1 adverse event were 1.68 G 1.68 and 2.34 G 1.98, and for patients with 2 or more adverse events, 1.54 G 1.27 and 2.67 G 1.81, respectively. A significantly greater incidence of elevated postoperative IOP on postoperative day 1 was observed in eyes operated on by residents as compared with attending surgeons (P Z .02). However, the group of resident-operated eyes had a higher mean baseline IOP (Table 1); as a result, the incidence of elevated IOP on the first postoperative day did not differ between the 2 groups when controlling for this difference (P Z .15). Overall, the longer operative time was correlated with a higher IOP on postoperative day 1 (Spearman coefficient Z 0.19, P Z .004). Table 5 shows the factors that predict adverse events associated with cataract surgery in cases with iris challenges performed by residents. No statistically significant association with preoperative or intraoperative characteristics was identified. Table 6 shows the factors that predict adverse events associated with cataract surgery in cases with iris challenges. Nuclear sclerosis, case difficulty, and macular disease were significantly associated with events in the univariate analyses (P Z .001, P Z .047, and P Z .028, respectively). In the multivariate analysis, only nuclear sclerosis maintained a significant association (P Z .001). Case difficulty, patient age, sex, resident versus attending as the primary surgeon, macular disease, and IFIS were not statistically significant contributors to the rate of events.

DISCUSSION Residents in ophthalmology must learn to perform cataract surgery on the growing subset of patients with iris challenges from miosis, IFIS, and other contributors. This analysis of surgical outcomes in resident-performed versus attending surgeon-performed cataract surgeries with iris challenges showed that in the real surgical setting under attending surgeon supervision, resident-performed surgeries achieve outcomes comparable to those of attending ophthalmologists. After stratifying for macular disease, there was no statistical difference in the postoperative visual outcomes between resident and attending surgeons. Other studies reporting visual outcomes in resident-performed cataract surgeries2,21,27,28,31,50 show that vision universally fails to improve or deteriorates in the presence of perioperative complications and baseline comorbidities, including macular disease. Taking all these studies together, an improvement of 20/40 or better is achieved in 73% to 93% of resident-operated eyes, with the rate increasing to 92% to 100% when excluding eyes with comorbidities.51 In our cohort, 100% of resident cases achieved vision better than 20/40. These favorable results were likely in part attributable to the underrepresentation of factors that have a major impact on postoperative vision, such as macular disease, in our resident-performed cataract surgeries. No statistically significant differences were observed in the incidence of the majority of intraoperative and postoperative adverse events between the cohort of surgeries performed by residents and attending surgeons. The rate of posterior capsule tear was the same in resident and attending surgeon-performed surgeries, and lower than that reported in most other studies, in which the rate often exceeds 4%.51 On the other hand, anterior capsule tears occurred more frequently in the resident group than the attending surgeon group (3% versus 1.8%). Both the resident and attending surgeon cohorts of patients with iris challenges reported in this study showed higher rates of anterior capsule tear as compared with previously published anterior capsule tear rates in resident-performed and attending surgeon-performed cataract surgeries not exclusively examining patients with iris challenges.8,52 These results emphasize the importance of a well-dilated and fixated pupil in steps of cataract surgery, such as capsulorhexis.52

Table 3. Comparison of visual outcomes after phacoemulsification by resident versus attending surgeons. Full Cohort

Postop Mo 1 VA (logMAR) Mean UDVA G SD Mean CDVA G SD

Residents (n Z 65 Cases)

Attending Surgeons (n Z 168 Cases)

0.328 G 0.34 0.328 G 0.34 0.051 G 0.10 0.132 G 0.30

Cases w/o Macular Disease

P Value .84 .03*

Residents (n Z 60 Cases)

Attending Surgeons (n Z 141 Cases)

0.330 G 0.35 0.269 G 0.26 0.042 G 0.09 0.065 G 0.13

Cases w/ Macular Disease

Residents P Value (n Z 5 Cases) .56 .46

Attending Surgeons (n Z 27 Cases)

0.304 G 0.20 0.622 G 0.52 0.175 G 0.15 0.462 G 0.56

P Value .19 .71

CDVA Z corrected distance visual acuity; logMAR Z logarithm of the minimum angle of resolution; UDVA Z uncorrected distance visual acuity; VA Z visual acuity *Statistically significant

Volume - Issue - - 2018

6

RESIDENT-PERFORMED CATARACT SURGERY WITH IRIS CHALLENGES

Table 4. Comparison of intraoperative and postoperative event rates of eyes with iris challenges that had phacoemulsification by resident versus attending surgeons. Number (%)

Event Intraoperative Anterior capsule tear Posterior capsule tear Zonular dehiscence Out-of-the-bag IOL Iris trauma Postoperative Inflammation IOP O21 mm Hg Corneal edema Epithelial defect Retained lens fragment Descemet membrane tear Eye pain Retinal tear/detachment Epiretinal membrane Macular edema Infectious endophthalmitis Referral to specialist R1 event(s)

Residents (n Z 65 Cases) 2 2 1 4 3

Attending Surgeons (n Z 168 Cases)

(3.1) (3.1) (1.5) (6.2) (4.2)

3 5 0 6 12

4 (6.2) 28 (43.1) 1 (1.5) 1 (1.5) 2 (3.1) 1 (1.5) 2 (3.1) 0 (0.0) 3 (4.2) 1 (1.5) 0 (0.0) 2 (3.1) 41 (63.1)

(1.8) (3.0) (0.0) (3.6) (7.1)

P Value .62 1.00 .28 .47 .57

15 (8.9) 46 (27.4) 4 (2.4) 3 (1.8) 2 (1.2) 0 (0.0) 8 (4.8) 1 (0.6) 2 (1.2) 1 (0.6) 0 (0.0) 5 (3.0) 89 (53.0)

.49 .02* 1.00 1.00 .31 .28 .73 1.00 .34 .49 d 1.00 .16

IOL Z intraocular lens; IOP Z intraocular pressure *Statistically significant

We also found an increased incidence (3%) of retained lens fragment in eyes with challenging irises as compared with phacoemulsification cohorts not exclusively examining patients with iris challenges. Tzamalis et al.9 observed a 1% rate of retained lens material among 502 residentperformed phacoemulsification cases, whereas Puri et al.8 reported a 0.6% rate of retained lens fragment under experienced attending surgeon supervision. Likewise, in a large study of 3389 resident-operated eyes,53 0.8% of cases were reoperated for retained lens fragments. The attending surgeon rate of postoperative retained lens fragment in our study (1.2%) was similar to the 1.1% incidence of retained lens fragments reported in the United Kingdom National Cataract Surgery Survey.54 Iris trauma was similarly more often seen in our study as compared with other phacoemulsification cohorts22,54,55; however, the fact that trauma is variously defined among

several studies complicates comparison of outcomes. In our study, iris trauma was more frequently observed in attending surgeon-performed (7%) versus residentperformed (4%) surgeries. This unexpected result could be explained by the lower average Muhtaseb scores of these resident cases compared with the attending surgeon cases. An elevated IOP on the first postoperative day was more frequently observed in resident cases, and this difference was statistically significant. In a group of residentoperated eyes, Elfersy et al.56 similarly reported a 2 to 5 times higher incidence of elevated IOP on the first postoperative day as compared with attending surgeons for all levels of intraocular hypertension. In our study, adjusting our findings for baseline IOP eliminated the significance of this finding. The operative time was longer in the cataract surgeries with iris challenges performed by residents than attending

Table 5. Odds ratios of events during or after resident-performed phacoemulsification in eyes with challenging irises.* Predictor Age Male sex Previous cases performed IFIS Macular disease Muhtaseb score

Unadjusted OR (95% CI)

Unadjusted P Value

Adjusted OR (95% CI)

Adjusted P Value

0.97 (0.90, 1.04) 0.95 (0.35, 2.61) 0.99 (0.99, 1.01) 1.44 (0.33, 6.19) 0.13 (0.13, 1.19) 1.14 (0.79, 1.64)

.39 .92 .45 .62 .07 .49

0.99 (0.92, 1.06) 0.76 (0.24, 2.43) 0.99 (0.99, 1.01) 0.90 (0.16, 4.99) 0.14 (0.01, 1.61) 1.10 (0.75, 1.60)

.73 .64 .54 .90 .11 .65

CI Z confidence interval; IFIS Z intraoperative floppy iris syndrome; OR Z odds ratio *Analysis of 65 cases

Volume - Issue - - 2018

7

RESIDENT-PERFORMED CATARACT SURGERY WITH IRIS CHALLENGES

Table 6. Odds ratios of events during or after phacoemulsification by residents and attending surgeons in eyes with challenging irises.* Predictor Age Male sex Resident IFIS Nuclear sclerosis Macular disease Muhtaseb score

Unadjusted OR (95% CI)

Unadjusted P Value

Adjusted OR (95% CI)

Adjusted P Value

1.00 (0.98, 1.03) 0.63 (0.37, 1.08) 1.52 (0.84, 2.73) 1.23 (0.88, 1.71) 1.88 (1.30, 2.71) 0.42 (0.20, 0.91) 1.18 (1.01, 1.38)

.77 .09 .17 .23 .001† .028† .047†

0.99 (0.96, 1.02) 0.61 (0.34, 1.09) 1.48 (0.78, 2.82) 1.12 (0.77, 1.62) 1.86 (1.17, 2.94) 0.50 (0.22, 1.14) 1.02 (0.83, 1.24)

.44 .10 .23 .55 .001† .10 .86

CI Z confidence interval; IFIS Z intraoperative floppy iris syndrome; OR Z odds ratio *Analysis of 228 cases † Statistically significant

surgeons, and longer for both groups as compared with previously reported operative times in broader cohorts of patients not exclusively manifesting iris challenges. Two studies by Taravella et al.10,19 identified a mean operative time (including patch time) of 38.86 G 12.33 minutes for residents who had previously completed similar number of cases as residents in our group. Wiggins and Warner57 estimated a mean operative time of 30.0 G 6.91 and 27.2 G 6.91 minutes for subsequent resident cases 84 to 88 and 119 to 123, respectively (w98 cases mean value in our group). Moreover, Taravella et al.19 characterized residents as competent if they performed phacoemulsification within 38 minutes (1 SD above the attending surgeons’ mean time), completed surgery without complications, and required no assistance. They reported that 42% of residents with a previous experience of 75 to 100 surgeries met these standards. According to these criteria, only 15% of residents achieved this level of competence in our study, underlining the challenges of managing a problematic iris in cataract surgery. Likewise, attending surgeon time of 30.9 G 12.6 minutes was higher than surgery time between 25.75 and 26.80 minutes previously published for attending surgeons operating on standard cataract eyes.10,19,58 Our study highlighted advanced nuclear sclerosis to be a risk factor for complications in eyes undergoing phacoemulsification with iris challenges. In the study of Briszi et al.,14 a white cataract increased the risk by 3.9 times for posterior capsule tears and by 4.3 times for vitreous loss. The risk for both posterior capsule tears and vitreous loss was 4.7 times higher in the presence of dense nuclear sclerosis. Multivariate analysis showed only dense nuclear sclerosis to predict intraoperative complications in resident-performed cataract surgery, increasing the risk by 3.2 times.14 In a study exploring risk factors for major complications in resident-performed cataract surgery, Rutar et al.31 found dense nuclear sclerosis to demonstrate a strong association with complications, increasing the risk by almost 19 times. Najjar and Awwad59 assigned the highest point in their scale for a dense cataract, and a study validating their scoring system60 showed an increased risk for complications among residents operating on eyes with dense nuclear sclerotic cataract or poor red reflex. We were not able to identify risk factors predicting adverse intraoperative or postoperative events for surgeries with iris

challenges performed by residents. In a study by Lomi et al.,6 previous resident experience was reported to be the only factor predicting complications in cataract surgery performed by residents with the rate of complications decreasing with increasing experience. In our study, no association was shown between resident experience and the rate of adverse events, indicating that a problematic iris can be a challenge even for experienced residents. Challenges in reporting and comparing cataract outcomes include the varying definitions used to describe complications throughout the literature, the different numbers of complications analyzed to estimate the overall complication rate, as well as differing ways in which a case is defined as difficult. To keep our methodology simple, we grouped intraoperative and postoperative events together and created a broad definition of adverse event. Although other studies reporting cataract surgery outcomes characterize a case as difficult in a simplified way,15,18,19,31 we determined case difficulty using an established risk scoring system.46 Moreover, by treating the Muhtaseb score as a continuous variable, we were able to extract more information about intermediate levels of case difficulty. Similarly, resident experience was expressed as the number of previous phacoemulsification procedures performed, which is a more accurate measure as compared with residency year used in other studies.4,18,31 Although we reviewed a large number of patients, a limitation of our study is the low statistical power with regard to specific outcomes with effect size similar between the groups and infrequently occurring events, which might result in failure to reach statistical significance in some cases. Another limitation is its retrospective nature. As a result, inclusion of cases was based on data from the operative note and medical record. Medical records included in this study had relatively few omissions, and data collection was done manually by well-trained study personnel. This study was performed at one site and thus potentially limits broader generalization of the results. Finally, surgeries performed by residents were chosen by attending surgeons. Controlling for the bias introduced by attending surgeon case selection, as well as the variable degree of attending surgeon supervision and direct involvement in the surgical procedure, was beyond the scope of this study. Characterizing the variable degree of attending surgeon supervision Volume - Issue - - 2018

8

RESIDENT-PERFORMED CATARACT SURGERY WITH IRIS CHALLENGES

in each case was not feasible to assess retrospectively with the data available. The Muhtaseb scores indicate that the cohort of cases performed by residents were of lesser difficulty than those performed by attending surgeons. We deliberately did not control for this variable because we aimed to present data that reflect the “real world” teaching environment and can be applied to other teaching surgical settings. This analysis could be an interesting direction for future studies. To our knowledge, this is the first study comparing resident outcomes with those of attending surgeons in complex cataract cases. Our results suggest that cataract surgeries with iris challenges performed by ophthalmology residents might require a longer time in the operating room and more surgical manipulation but can achieve outcomes comparable to surgeries performed by attending surgeons. Our findings suggest that residents should be given the opportunity to operate on these cases. However, an increased level of alertness and close resident supervision is recommended in cataract surgeries with anticipated iris challenges and dense nuclear sclerosis.

8.

9.

10. 11.

12.

13.

14.

15.

16.

WHAT WAS KNOWN  Phacoemulsification with iris challenges, including preoperative or intraoperative miosis, prolapse, or IFIS, is associated with an increased risk for complications for both residents and attending surgeons. However, there are no data on how in comparison to attending surgeons, residents perform in cataract cases with iris challenges.

17.

WHAT THIS PAPER ADDS

19.

 Although residents required longer operating times for surgeries performed on cataract cases with iris challenges, they achieved visual outcomes and adverse event rates comparable to those achieved by attending surgeons.  Dense nuclear sclerosis was associated with a higher risk for adverse events in cataract cases with iris challenges.

18.

20.

21.

22.

23.

REFERENCES 1. Klein BE, Howard KP, Lee KE, Klein R. Changing incidence of lens extraction over 20 years: the Beaver Dam eye study. Ophthalmology 2014; 121:5–9 2. Payal AR, Gonzalez-Gonzalez LA, Chen X, Cakiner-Egilmez T, Chomsky A, Baze E, Vollman D, Lawrence MG, Daly MK. Outcomes of cataract surgery with residents as primary surgeons in the Veterans Affairs Healthcare System. J Cataract Refract Surg 2016; 42:370–384 3. Finn AP, Borboli-Gerogiannis S, Brauner S, Peggy Chang HY, Chen S, Gardiner M, Greenstein SH, Kloek C, Miller JW, Chen TC. Assessing resident cataract surgery outcomes using medicare physician quality reporting system measures. J Surg Educ 2016; 73:774–779 4. Menda SA, Driver TH, Neiman AE, Blumberg S, Naseri A, Stewart JM. Risk factors for return to the operating room after resident-performed cataract surgery. Semin Ophthalmol 2018; 33:210–214 5. Mangan MS, Atalay E, Arici C, Tuncer I, Bilgec MD. Comparison of different types of complications in the phacoemulsification surgery learning curve according to number of operations performed. Turk J Ophthalmol 2016; 46:7–10 6. Lomi N, Sharma R, Khokhar S, Dada T, Vanathi M, Agarwal T. Risk factors for intra-operative complications during phacoemulsification performed by residents. Int Ophthalmol 2016; 36:401–406 7. Schmidt CM, Sundararajan M, Biggerstaff KS, Orengo-Nania S, Coffee RE, Khandelwal SS. Indications and outcomes of resident-performed cataract

Volume - Issue - - 2018

24.

25.

26.

27.

28.

29.

30.

surgery requiring return to the operating room. J Cataract Refract Surg 2016; 42:385–391 Puri S, Kiely AE, Wang J, Woodfield AS, Ramanathan S, Sikder S. Comparing resident cataract surgery outcomes under novice versus experienced attending supervision. Clin Ophthalmol 2015; 9:1675–1681 Tzamalis A, Lamprogiannis L, Chalvatzis N, Symeonidis C, Dimitrakos S, Tsinopoulos I. Training of resident ophthalmologists in cataract surgery: a comparative study of two approaches. J Ophthalmol 2015; 2015:932043 Taravella MJ, Davidson R, Erlanger M, Guiton G, Gregory D. Time and cost of teaching cataract surgery. J Cataract Refract Surg 2014; 40:212–216 Hashemi H, Mohammadpour M, Jabbarvand M, Nezamdoost Z, Ghadimi H. Incidence of and risk factors for vitreous loss in residentperformed phacoemulsification surgery. J Cataract Refract Surg 2013; 39:1377–1382 Meeks LA, Blomquist PH, Sullivan BR. Outcomes of manual extracapsular versus phacoemulsification cataract extraction by beginner resident surgeons. J Cataract Refract Surg 2013; 39:1698–1701 Tsinopoulos IT, Lamprogiannis LP, Tsaousis KT, Mataftsi A, Symeonidis C, Chalvatzis NT, Dimitrakos SA. Surgical outcomes in phacoemulsification after application of a risk stratification system. Clin Ophthalmol 2013; 7:895–899 Briszi A, Prahs P, Hillenkamp J, Helbig H, Herrmann W. Complication rate and risk factors for intraoperative complications in resident-performed phacoemulsification surgery. Graefes Arch Clin Exp Ophthalmol 2012; 250:1315–1320 Fong CS, Mitchell P, de Loryn T, Rochtchina E, Hong T, Cugati S, Wang JJ. Long-term outcomes of phacoemulsification cataract surgery performed by trainees and consultants in an Australian cohort. Clin Exp Ophthalmol 2012; 40:597–603 Roensch MA, Charton JW, Blomquist PH, Aggarwal NK, McCulley JP. Resident experience with toric and multifocal intraocular lenses in a public county hospital system. J Cataract Refract Surg 2012; 38:793–798 Khanna RC, Kaza S, Palamaner Subash Shantha G, Sangwan VS. Comparative outcomes of manual small incision cataract surgery and phacoemulsification performed by ophthalmology trainees in a tertiary eye care hospital in India: a retrospective cohort design. BMJ Open 2012; 2: pii e001035 Woodfield AS, Gower EW, Cassard SD, Ramanthan S. Intraoperative phacoemulsification complication rates of second- and third-year ophthalmology residents a 5-year comparison. Ophthalmology 2011; 118:954–958 Taravella MJ, Davidson R, Erlanger M, Guiton G, Gregory D. Characterizing the learning curve in phacoemulsification. J Cataract Refract Surg 2011; 37:1069–1075 Ku TK, Rutar T, Han Y, Porco TC, Naseri A. Resident-performed phacoemulsification surgery in tamsulosin-treated patients. Arch Ophthalmol 2010; 128:967–972 Kim JY, Ali R, Cremers SL, Yun SC, Henderson BA. Incidence of intraoperative complications in cataract surgery performed by left-handed residents. J Cataract Refract Surg 2009; 35:1019–1025 Lee JS, Hou CH, Yang ML, Kuo JZ, Lin KK. A different approach to assess resident phacoemulsification learning curve: analysis of both completion and complication rates. Eye (Lond) 2009; 23:683–687 Khatibi A, Naseri A, Stewart JM. Rate of rhegmatogenous retinal detachment after resident-performed cataract surgery is similar to that of experienced surgeons. Br J Ophthalmol 2008; 92:438 Randleman JB, Wolfe JD, Woodward M, Lynn MJ, Cherwek DH, Srivastava SK. The resident surgeon phacoemulsification learning curve. Arch Ophthalmol 2007; 125:1215–1219 Bhagat N, Nissirios N, Potdevin L, Chung J, Lama P, Zarbin MA, Fechtner R, Guo S, Chu D, Langer P. Complications in residentperformed phacoemulsification cataract surgery at New Jersey Medical School. Br J Ophthalmol 2007; 91:1315–1317 lu K, Akar Y, Altin M. PhacoUnal M, Yucel I, Sarici A, Artunay O, Devranog emulsification with topical anesthesia: resident experience. J Cataract Refract Surg 2006; 32:1361–1365 Randleman JB, Srivastava SK, Aaron MM. Phacoemulsification with topical anesthesia performed by resident surgeons. J Cataract Refract Surg 2004; 30:149–154 Quillen DA, Phipps SJ. Visual outcomes and incidence of vitreous loss for residents performing phacoemulsification without prior planned extracapsular cataract extraction experience. Am J Ophthalmol 2003; 135:732–733 Blomquist PH, Rugwani RM. Visual outcomes after vitreous loss during cataract surgery performed by residents. J Cataract Refract Surg 2002; 28:847–852 Blomquist PH, Morales ME, Tong L, Ahn C. Risk factors for vitreous complications in resident-performed phacoemulsification surgery. J Cataract Refract Surg 2012; 38:208–214

9

RESIDENT-PERFORMED CATARACT SURGERY WITH IRIS CHALLENGES

31. Rutar T, Porco TC, Naseri A. Risk factors for intraoperative complications in resident-performed phacoemulsification surgery. Ophthalmology 2009; 116:431–436 32. Guzek JP, Holm M, Cotter JB, Cameron JA, Rademaker WJ, Wissinger DH, Tonjum AM, Sleeper LA. Risk factors for intraoperative complications in 1000 extracapsular cataract cases. Ophthalmology 1987; 94:461–466 33. Lumme P, Laatikainen LT. Risk factors for intraoperative and early postoperative complications in extracapsular cataract surgery. Eur J Ophthalmol 1994; 4:151–158 34. Goldman JM, Karp CL. Adjunct devices for managing challenging cases in cataract surgery: capsular staining and ophthalmic viscosurgical devices. Curr Opin Ophthalmol 2007; 18:52–57 35. Narendran N, Jaycock P, Johnston RL, Taylor H, Adams M, Tole DM, Asaria RH, Galloway P, Sparrow JM. The Cataract National Dataset electronic multicentre audit of 55,567 operations: risk stratification for posterior capsule rupture and vitreous loss. Eye (Lond) 2009; 23:31–37 36. Chang DF, Campbell JR. Intraoperative floppy iris syndrome associated with tamsulosin. J Cataract Refract Surg 2005; 31:664–673 37. Chen AA, Kelly JP, Bhandari A, Wu MC. Pharmacologic prophylaxis and risk factors for intraoperative floppy-iris syndrome in phacoemulsification performed by resident physicians. J Cataract Refract Surg 2010; 36:898–905 38. Bell CM, Hatch WV, Fischer HD, Cernat G, Paterson JM, Gruneir A, Gill SS, Bronskill SE, Anderson GM, Rochon PA. Association between tamsulosin and serious ophthalmic adverse events in older men following cataract surgery. JAMA 2009; 301:1991–1996 39. Haridas A, Syrimi M, Al-Ahmar B, Hingorani M. Intraoperative floppy iris syndrome (IFIS) in patients receiving tamsulosin or doxazosin-a UK-based comparison of incidence and complication rates. Graefes Arch Clin Exp Ophthalmol 2013; 251:1541–1545 40. Vollman DE, Gonzalez-Gonzalez LA, Chomsky A, Daly MK, Baze E, Lawrence M. Intraoperative floppy iris and prevalence of intraoperative complications: results from ophthalmic surgery outcomes database. Am J Ophthalmol 2014; 157:1130–1135.e1131 41. Ku TK, Naseri A, Han Y, Porco TC, Rutar T. Effect of timing and duration of tamsulosin exposure on complications in resident-performed phacoemulsification. Ophthalmic Surg Lasers Imaging 2011; 42:416–422 42. Storr-Paulsen A, Jorgensen JS, Norregaard JC, Thulesen J. Corneal endothelial cell changes after cataract surgery in patients on systemic sympathetic alpha-1a antagonist medication (tamsulosin). Acta Ophthalmol 2014; 92:359–363 43. Tint NL, Dhillon AS, Alexander P. Management of intraoperative iris prolapse: stepwise practical approach. J Cataract Refract Surg 2012; 38:1845–1852 44. Tint NL, Yeung AM, Alexander P. Management of intraoperative floppy-iris syndrome-associated iris prolapse using a single iris retractor. J Cataract Refract Surg 2009; 35:1849–1852 45. Chylack LT Jr, Wolfe JK, Singer DM, Leske MC, Bullimore MA, Bailey IL, Friend J, McCarthy D, Wu SY. The Lens Opacities Classification System III. The Longitudinal Study of Cataract Study Group. Arch Ophthalmol 1993; 111:831–836 46. Muhtaseb M, Kalhoro A, Ionides A. A system for preoperative stratification of cataract patients according to risk of intraoperative complications: a prospective analysis of 1441 cases. Br J Ophthalmol 2004; 88:1242–1246 47. Malyugin B. Small pupil phaco surgery: a new technique. Ann Ophthalmol (Skokie) 2007; 39:185–193 48. Shugar JK. Use of epinephrine for IFIS prophylaxis. J Cataract Refract Surg 2006; 32:1074–1075 49. Lange C, Feltgen N, Junker B, Schulze-Bonsel K, Bach M. Resolving the clinical acuity categories “hand motion” and “counting fingers” using the

50.

51.

52. 53. 54.

55.

56.

57. 58.

59. 60.

Freiburg Visual Acuity Test (FrACT). Graefes Arch Clin Exp Ophthalmol 2009; 247:137–142 Gross FJ, Garcia-Zalisnak DE, Bovee CE, Strawn JD. A comparison of pop and chop to divide and conquer in resident cataract surgery. Clin Ophthalmol 2016; 10:1847–1851 Kaplowitz K, Yazdanie M, Abazari A. A review of teaching methods and outcomes of resident phacoemulsification. Surv Ophthalmol 2018; 63:257–267 Marques FF, Marques DM, Osher RH, Osher JM. Fate of anterior capsule tears during cataract surgery. J Cataract Refract Surg 2006; 32:1638–1642 Schwartz SG, Holz ER, Mieler WF, Kuhl DP. Retained lens fragments in resident-performed cataract extractions. CLAO J 2002; 28:44–47 Desai P, Minassian DC, Reidy A. National cataract surgery survey 1997-8: a report of the results of the clinical outcomes. Br J Ophthalmol 1999; 83:1336–1340 Misra A, Burton RL. Incidence of intraoperative complications during phacoemulsification in vitrectomized and nonvitrectomized eyes: prospective study. J Cataract Refract Surg 2005; 31:1011–1014 Elfersy AJ, Prinzi RA, Peracha ZH, Kim DD, Crandall DA, DarnleyFisch DA, Imami NR. IOP elevation after cataract surgery: results for residents and senior staff at Henry Ford Health System. J Glaucoma 2016; 25:802–806 Wiggins MN, Warner DB. Resident physician operative times during cataract surgery. Ophthalmic Surg Lasers Imaging 2010; 41:518–522 Wong T, Hingorani M, Lee V. Phacoemulsification time and power requirements in phaco chop and divide and conquer nucleofractis techniques. J Cataract Refract Surg 2000; 26:1374–1378 Najjar DM, Awwad ST. Cataract surgery risk score for residents and beginning surgeons. J Cataract Refract Surg 2003; 29:2036–2037 Blomquist PH, Sargent JW, Winslow HH. Validation of Najjar-Awwad cataract surgery risk score for resident phacoemulsification surgery. J Cataract Refract Surg 2010; 36:1753–1757

OTHER CITED MATERIAL A. Accreditation Council for Graduate Medical Education (ACGME). Ophthalmology Case Logs: National Data Report. 2010-2017. Available at: www.acgme.org /Data-Collection-Systems/Case-Logs-Statistical-Reports. Accessed August 27, 2018

Disclosures: None of the authors has a financial or proprietary interest in any material or method mentioned. Group Information: The Perioperative Care for Intraocular Lens Study Group members are Sheila Borboli-Gerogiannis, MD, Stacey Brauner, MD, H. Peggy Chang, MD, Kenneth Chang, MD, Sherleen H. Chen, MD, Matthew Gardiner, MD, Scott H. Greenstein, MD, Carolyn E. Kloek, MD, Ann-Marie Lobo, MD, Z. Katie Luo, MD, PhD

First author: Giannis A. Moustafa, MD Department of Ophthalmology, Massachusetts Eye and Ear Infirmary, Harvard Medical School, Boston, Massachusetts, USA

Volume - Issue - - 2018