Advertisement

Vision Outcomes Following Anti–Vascular Endothelial Growth Factor Treatment of Diabetic Macular Edema in Clinical Practice

Open AccessPublished:April 20, 2018DOI:https://doi.org/10.1016/j.ajo.2018.04.010

      Purpose

      To determine monitoring and treatment patterns and vision outcomes in real-world patients initiating anti–vascular endothelial growth factor (anti-VEGF) therapy for diabetic macular edema (DME).

      Design

      Retrospective interventional cohort study.

      Methods

      Setting: Electronic medical record analysis of Geisinger Health System data. Study Population: A total of 110 patients (121 study eyes) initiating intravitreal ranibizumab or bevacizumab for DME during January 2007‒May 2012, with baseline corrected visual acuity of 20/40 to 20/320, and ≥1 ophthalmologist visit during follow-up. Main Outcome Measures: Intravitreal injections per study eye during the first 12 months; corrected visual acuity, change in corrected visual acuity from baseline, proportions of eyes with ≥10 or ≥15 approximate Early Treatment Diabetic Retinopathy Study letter gain/loss at 12 months; number of ophthalmologist visits.

      Results

      Over 12 months, mean number of ophthalmologist visits was 9.2; mean number of intravitreal injections was 3.1 (range, 1–12), with most eyes (68.6%) receiving ≤3 injections. At 12 months, mean corrected visual acuity change was +4.7 letters (mean 56.9 letters at baseline); proportions of eyes gaining ≥10 or ≥15 letters were 31.4% and 24.0%, respectively; proportions of eyes losing ≥10 or ≥15 letters were 10.8% and 8.3%, respectively. Eyes receiving adjunctive laser during the first 6 months (n = 33) showed similar change in corrected visual acuity to non–laser-treated eyes (n = 88) (+3.1 vs +5.3 letters at 12 months).

      Conclusions

      DME patients receiving anti-VEGF therapy in clinical practice undergo less frequent monitoring and intravitreal injections, and achieve inferior vision outcomes to patients in landmark clinical trials.
      Diabetes mellitus has reached epidemic proportions worldwide. In the United States alone, an estimated 29.1 million people (9.3% of the population) have diabetes.
      • Centers for Disease Control and Prevention
      National Diabetes Statistics Report: Estimates of Diabetes and Its Burden in the United States, 2014.
      Diabetic retinopathy/maculopathy, one of several microvascular complications of diabetes, is a leading cause of acquired blindness among the working-age population in developed countries.
      • Liew G.
      • Michaelides M.
      • Bunce C.
      A comparison of the causes of blindness certifications in England and Wales in working age adults (16-64 years), 1999-2000 with 2009-2010.
      • Buch H.
      • Vinding T.
      • La Cour M.
      • Appleyard M.
      • Jensen G.B.
      • Nielsen N.V.
      Prevalence and causes of visual impairment and blindness among 9980 Scandinavian adults: the Copenhagen City Eye Study.
      Central vision impairment occurs when edema affects the central retina or macula.
      • Antonetti D.A.
      • Klein R.
      • Gardner T.W.
      Diabetic retinopathy.
      A recent pooled analysis of 35 population-based studies in the United States, Australia, Europe, and Asia indicates that the global, age-standardized prevalence of diabetic retinopathy and diabetic macular edema (DME) in diabetic patients younger than 80 years of age is ∼35% and 7.5%, respectively.
      • Yau J.W.
      • Rogers S.L.
      • Kawasaki R.
      • et al.
      Global prevalence and major risk factors of diabetic retinopathy.
      In the United States, there are an estimated 7.7 million cases of diabetic retinopathy and 0.8 million cases of DME in diabetic individuals older than 40 years (2010 statistics).
      • National Eye Institute
      Diabetic retinopathy.
      • Varma R.
      • Bressler N.M.
      • Doan Q.V.
      • et al.
      Prevalence of and risk factors for diabetic macular edema in the United States.
      The pathogenesis of DME is multifactorial, involving angiogenic, inflammatory, hypoxic, and hemodynamic processes that lead to breakdown of the blood-retinal barrier and intraretinal fluid leakage.
      • Antcliff R.J.
      • Marshall J.
      The pathogenesis of edema in diabetic maculopathy.
      • Zhang X.
      • Zeng H.
      • Bao S.
      • Wang N.
      • Gillies M.C.
      Diabetic macular edema: new concepts in patho-physiology and treatment.
      • Ehrlich R.
      • Harris A.
      • Ciulla T.A.
      • Kheradiya N.
      • Winston D.M.
      • Wirostko B.
      Diabetic macular oedema: physical, physiological and molecular factors contribute to this pathological process.
      Vascular endothelial growth factor (VEGF), a potent mediator of retinal vessel growth and permeability,
      • Ferrara N.
      • Gerber H.P.
      • LeCouter J.
      The biology of VEGF and its receptors.
      is closely implicated in blood-retinal barrier disruption in diabetes,
      • Zhang X.
      • Zeng H.
      • Bao S.
      • Wang N.
      • Gillies M.C.
      Diabetic macular edema: new concepts in patho-physiology and treatment.
      • Qaum T.
      • Xu Q.
      • Joussen A.M.
      • et al.
      VEGF-initiated blood-retinal barrier breakdown in early diabetes.
      making it an attractive target for pharmacologic intervention. Laser photocoagulation, the mainstay of treatment for patients with DME for many years, has proved to be effective in stabilizing vision in DME but of limited effect in improving impaired vision.
      • Ford J.A.
      • Lois N.
      • Royle P.
      • Clar C.
      • Shyangdan D.
      • Waugh N.
      Current treatments in diabetic macular oedema: systematic review and meta-analysis.
      Recent years have seen the emergence of intravitreal anti-VEGF and intravitreal corticosteroid therapies that offer the promise of reversing vision loss in DME.
      • Bandello F.
      • Cunha-Vaz J.
      • Chong N.V.
      • et al.
      New approaches for the treatment of diabetic macular oedema: recommendations by an expert panel.
      Currently, intravitreal anti-VEGF agents—used either alone or as an adjunct to laser photocoagulation—are the preferred first-line treatment for DME, particularly for DME with foveal involvement and associated vision loss.
      • Bandello F.
      • Cunha-Vaz J.
      • Chong N.V.
      • et al.
      New approaches for the treatment of diabetic macular oedema: recommendations by an expert panel.
      Randomized clinical trials in DME have demonstrated that the anti-VEGF agents ranibizumab, bevacizumab, and aflibercept, used either alone or in conjunction with laser photocoagulation, produce greater improvement in visual acuity than laser therapy alone.
      • Massin P.
      • Bandello F.
      • Garweg J.G.
      • et al.
      Safety and efficacy of ranibizumab in diabetic macular edema (RESOLVE Study): a 12-month, randomized, controlled, double-masked, multicenter phase II study.
      • Mitchell P.
      • Bandello F.
      • Schmidt-Erfurth U.
      • et al.
      The RESTORE study: ranibizumab monotherapy or combined with laser versus laser monotherapy for diabetic macular edema.
      • Nguyen Q.D.
      • Brown D.M.
      • Marcus D.M.
      • et al.
      Ranibizumab for diabetic macular edema: results from 2 phase III randomized trials: RISE and RIDE.
      • Elman M.J.
      • Aiello L.P.
      • Beck R.W.
      • et al.
      Randomized trial evaluating ranibizumab plus prompt or deferred laser or triamcinolone plus prompt laser for diabetic macular edema.
      • Elman M.J.
      • Bressler N.M.
      • Qin H.
      • et al.
      Expanded 2-year follow-up of ranibizumab plus prompt or deferred laser or triamcinolone plus prompt laser for diabetic macular edema.
      • Michaelides M.
      • Kaines A.
      • Hamilton R.D.
      • et al.
      A prospective randomized trial of intravitreal bevacizumab or laser therapy in the management of diabetic macular edema (BOLT study) 12-month data: report 2.
      • Korobelnik J.F.
      • Do D.V.
      • Schmidt-Erfurth U.
      • et al.
      Intravitreal aflibercept for diabetic macular edema.
      • Ishibashi T.
      • Li X.
      • Koh A.
      • et al.
      The REVEAL Study: ranibizumab monotherapy or combined with laser versus laser monotherapy in Asian patients with diabetic macular edema.
      The landmark clinical trials that have shown the most pronounced improvements in visual acuity with anti-VEGF therapy, such as RESOLVE, RISE/RIDE, PROTOCOL I, BOLT, VISTA, and VIVID,
      • Massin P.
      • Bandello F.
      • Garweg J.G.
      • et al.
      Safety and efficacy of ranibizumab in diabetic macular edema (RESOLVE Study): a 12-month, randomized, controlled, double-masked, multicenter phase II study.
      • Nguyen Q.D.
      • Brown D.M.
      • Marcus D.M.
      • et al.
      Ranibizumab for diabetic macular edema: results from 2 phase III randomized trials: RISE and RIDE.
      • Elman M.J.
      • Aiello L.P.
      • Beck R.W.
      • et al.
      Randomized trial evaluating ranibizumab plus prompt or deferred laser or triamcinolone plus prompt laser for diabetic macular edema.
      • Michaelides M.
      • Kaines A.
      • Hamilton R.D.
      • et al.
      A prospective randomized trial of intravitreal bevacizumab or laser therapy in the management of diabetic macular edema (BOLT study) 12-month data: report 2.
      • Korobelnik J.F.
      • Do D.V.
      • Schmidt-Erfurth U.
      • et al.
      Intravitreal aflibercept for diabetic macular edema.
      have featured frequent (typically monthly) monitoring and injections (9–12) during the first year of treatment. In contrast, anti-VEGF utilization and patient follow-up patterns in the treatment of DME in clinical practice are considerably less intensive.
      • Kiss S.
      • Liu Y.
      • Brown J.
      • et al.
      Clinical utilization of anti-vascular endothelial growth-factor agents and patient monitoring in retinal vein occlusion and diabetic macular edema.
      Recent US claims analyses from multiple databases indicate that the mean number of intravitreal bevacizumab injections received by DME patients over the first year of treatment is ∼2 to 4 (Kiss S, et al. JMCP 2014;20[10a Suppl]:S21–22, Abstract E1).
      • Kiss S.
      • Liu Y.
      • Brown J.
      • et al.
      Clinical utilization of anti-vascular endothelial growth-factor agents and patient monitoring in retinal vein occlusion and diabetic macular edema.
      • VanderBeek B.L.
      • Shah N.
      • Parikh P.C.
      • Ma L.
      Trends in the care of diabetic macular edema: analysis of a national cohort.
      Infrequent anti-VEGF dosing might potentially compromise the extent of vision improvement. This study was undertaken to assess vision outcomes in patients receiving anti-VEGF treatment for DME in the clinical practice setting.

      Methods

       Study Design and Data Source

      This retrospective, interventional cohort study was based on electronic medical records (EMRs) from January 1, 2004 to May 31, 2013 sourced from the Geisinger Health System database, supplemented by administrative claims data and manual chart review. Geisinger Health System provides healthcare services to ∼2 million patients (including Medicare-eligible patients) in central and northeastern Pennsylvania, USA, and (in conjunction with other selected health systems) to ∼225 000 members of the Geisinger Health Plan. Geisinger's MedMining database includes ∼3.8 million individuals (1996 to present) and provides EMRs for services provided through the Geisinger Health System, as well as administrative claims data for Geisinger Health Plan enrollees. EMR data include longitudinal patient-level information on demographics; medical and medication histories; ambulatory care visits and inpatient admissions; associated diagnoses; orders for laboratory tests, medications, and procedures; and clinical laboratory and examination results. Healthcare claims data include outpatient and inpatient diagnoses and procedures, outpatient drug utilization, and dates of service. Medical, prescription, and patient encounter data are automatically linked to an underlying International Classification of Diseases, Version 9, Clinical Modification (ICD-9-CM) diagnosis code. All patient data used in the present study were de-identified in compliance with the patient confidentiality requirements of the Health Insurance Portability and Accountability Act of 1996.

       Patient Identification

      The EMR database was screened to identify patients who (1) had a diagnosis of DME (ICD-9-CM code 362.07 alone; or either 362.83 [retinal edema] or 362.53 [cystoid macular degeneration] in association with a prior [within the past 12 months] diagnosis of diabetes mellitus [ICD-9-CM code 250.xx]) from January 2004 onward; (2) received an initial (“index”) intravitreal injection of ranibizumab or bevacizumab between January 2007 and May 2012 (after the qualifying DME diagnosis); and (3) had at least 1 follow-up encounter in the Geisinger Health System within 12 months after the index anti-VEGF injection. Aflibercept was not included as a medication of interest, since its US Food and Drug Administration (FDA) approval (2015) postdated the study period. For study inclusion, patients were required to have a corrected visual acuity (CVA), as measured with patients wearing their habitual glasses or contact lenses, of 20/40 to 20/320 (treated eye) during the baseline period (from 30 days preceding to 7 days following the index anti-VEGF injection) and at least 1 valid CVA measurement in the same eye during the month 3 to month 12 window (day 60 to day 419) after the index anti-VEGF injection. Patients with a co-diagnosis of another retinal disease for which anti-VEGF therapy may have been prescribed, namely retinal vein occlusion (ICD-9-CM code 362.35 or 362.36) or age-related macular degeneration (ICD-9-CM code 362.42, 362.43, 362.52, or 362.50) with an intravitreal injection on the same day, were excluded from the study, as were patients for whom laterality of ocular disease or index treatment could not be determined.

       Study Outcomes

      Healthcare resource utilization during the 12-month post-index follow-up period was assessed in terms of the number of ophthalmologist (all ophthalmology subspecialty) visits made and the number of intravitreal anti-VEGF injections administered per study eye. Anti-VEGF injections were identified from Healthcare Common Procedure Coding System codes for bevacizumab (J3490, J3590, J9035, Q2024, C9257) and ranibizumab (J2778, C9233) that were accompanied by the Current Procedural Terminology code for intravitreal injection (67028). For the purpose of statistical analysis, visual acuity values, which in clinical practice were obtained using Snellen charts, were converted to approximate Early Treatment Diabetic Retinopathy Study (approxETDRS) letter scores using a published algorithm.
      • Gregori N.Z.
      • Feuer W.
      • Rosenfeld P.J.
      Novel method for analyzing Snellen visual acuity measurements.
      Vision outcomes during the 12-month post-index period were expressed in terms of the change in approxETDRS letter score from baseline and the proportions of eyes with ≥10 or ≥15 approxETDRS letters gain/loss in CVA.
      Primary analysis of visual acuity outcomes was performed using the full study population, comprising all study eyes with at least 1 valid postbaseline CVA assessment. Secondary analysis was performed using the subset of study eyes with regular ophthalmologist visits within each quarterly window of the follow-up period: month 3 (60–149 days post-index), month 6 (150–239 days post-index), month 9 (240–329 days post-index), and month 12 (330–419 days post-index). Visual acuity outcomes were determined firstly using an “intention-to-treat” approach, whereby missing data resulting from missed visits or visits with no CVA assessment are imputed using the last-observation-carried-forward (LOCF) method, and secondly using actual observed (ie, nonimputed) data. Sensitivity analyses were conducted to explore the effect of anti-VEGF injection frequency and adjunctive laser therapy on visual acuity outcomes.

       Statistics

      Data were summarized with descriptive statistics. Multivariate linear and logistic regression models were used to explore the effects of covariates (age, baseline CVA, number of anti-VEGF injections, and number of ophthalmologist visits) on visual acuity response (CVA, change from baseline in CVA, and proportion of patients with ≥10 approxETDRS letters gain/loss) at 12 months. Allowance for data correlation owing to inclusion of paired study eyes was provided through the use of clustered standard errors and random-effect models. Statistical significance was set at P < .05.

      Results

       Study Population

      Of 336 patients identified in the EMR database with a diagnosis of DME only (ie, no co-diagnosis of retinal vein occlusion or age-related macular degeneration) and initiating intravitreal anti-VEGF treatment between January 2007 and May 2012, 110 patients (121 eyes) met the full eligibility criteria and were included in the study (Figure 1). Of these, 99 patients provided a single study eye and 11 patients provided paired (left and right) study eyes.
      Figure thumbnail gr1
      Figure 1Patient selection algorithm. AMD = age-related macular degeneration; CVA = corrected visual acuity; DME = diabetic macular edema; IVI = intravitreal injection; RVO = retinal vein occlusion; VEGF = vascular endothelial growth factor.
      The patient population (N = 110) had a mean (standard deviation [SD]) age of 63.3 (11.6) years and was evenly divided between the sexes (Table 1). At the time of the index intravitreal treatment (baseline), the mean (SD) CVA of the study eye population was 56.9 (11.6) approxETDRS letters (approximately 20/80 Snellen equivalent). Intravitreal anti-VEGF therapy was initiated with bevacizumab in 116 eyes (95.9%) and ranibizumab in 5 eyes (4.1%), and was continued unchanged in all but 14 eyes (11.6%), which were switched from bevacizumab to ranibizumab therapy.
      Table 1Baseline Characteristics of Study Eyes
      CharacteristicAll Patients (N = 110)All Eyes (N = 121)Eyes Categorized by Number of Anti-VEGF InjectionsP Value
      ≤3 (N = 83)4–6 (N = 27)≥7 (N = 11)
      Age, mean (SD), y63.3 (11.6)63.7 (11.4)63.2 (11.4)65.6 (11.7)63.0 (10.8).658
      Linear regression models, using clustered standard errors.
      Sex, n (%).953
      Logistic regression models, using clustered standard errors.
       Male61 (55.4)66 (54.6)46 (55.4)14 (51.9)6 (54.5)
       Female49 (44.6)55 (45.4)37 (44.6)13 (48.1)5 (45.5)
      Bilateral DME, n (%)57 (51.2)68 (56.2)46 (55.4)17 (63.0)5 (45.5).611
      Logistic regression models, using clustered standard errors.
      Baseline CVA, mean (SD), approxETDRS letters56.9 (11.6)56.2 (11.9)58.4 (11.1)58.7 (11.2).544
      Linear regression models, using clustered standard errors.
      approxETDRS = approximate Early Treatment Diabetic Retinopathy Study; CVA = corrected visual acuity; DME = diabetic macular edema; VEGF = vascular endothelial growth factor.
      a Linear regression models, using clustered standard errors.
      b Logistic regression models, using clustered standard errors.

       Ophthalmology Visits and Visual Acuity Assessments

      Of 121 study eyes that received ≥12 months' follow-up after the index anti-VEGF injection, 109 (90.1%), 93 (76.9%), 90 (74.4%), and 93 (76.9%) eyes had at least 1 ophthalmologist office visit during the month 3 (60–149 days post-index), month 6 (150–239 days post-index), month 9 (240–329 days post-index), and month 12 (330–419 days post-index) windows, respectively, and 71 eyes (58.7%) received regular (at least quarterly) ophthalmologist visits. The mean and median number of ophthalmologist office visits per study eye over the 12-month post-index period was 9.2 and 8, respectively. Of the full study cohort, 85 eyes (70%) had a valid CVA assessment during the month 12 window (ie, between day 330 and day 419). Among the subset of 71 study eyes with regular (at least quarterly) ophthalmologist visits, 64 eyes (90%) had a valid CVA assessment during the month 12 window.

       Anti–Vascular Endothelial Growth Factor Treatment Frequency

      For the overall study population (N = 121 eyes), a mean (SD) of 3.1 (2.4) (range, 1–12) intravitreal anti-VEGF injections per study eye were administered over the 12-month follow-up period. Most (n = 83; 68.6%) study eyes received ≤3 anti-VEGF injections over this period, 27 (22.3%) study eyes received 4–6 injections, and 11 (9.1%) study eyes received ≥7 injections; only 3% of study eyes received anti-VEGF injections at a frequency (≥10 over 12 months) approximating monthly dosing (Figure 2). No significant differences in demographic and baseline clinical characteristics (eg, age, sex, bilateral DME involvement, baseline CVA) were noted between study eyes categorized according to injection frequency (≤3, 4–6, or ≥7 per 12 months) (Table 1).
      Figure thumbnail gr2
      Figure 2Frequency distribution of intravitreal anti–vascular endothelial growth factor injections over 12 months.

       Visual Acuity Outcomes

      For the overall study population (N = 121 eyes), mean (SD) CVA increased from 56.9 (11.6) approxETDRS letters at baseline to 61.6 (15.4) approxETDRS letters at 12 months, representing a mean (SD) change in CVA of +4.7 (12.3) approxETDRS letters (LOCF analysis), which was achieved with a mean of 3.1 anti-VEGF injections (Figure 3, Top). For the subset of eyes with regular follow-up (n = 71 eyes), mean (SD) CVA increased from 58.2 (10.9) approxETDRS letters at baseline to 61.8 (15.2) approxETDRS letters at 12 months, representing a mean (SD) change in CVA of +3.5 (13.4) approxETDRS letters (LOCF analysis), which was achieved with a mean of 3.2 anti-VEGF injections (Figure 3, Bottom). The distribution of changes in CVA (from baseline to 12 months) in individual study eyes was skewed toward improvement, although a substantial proportion of eyes showed either stability (21%) or deterioration (22%) in CVA (Figure 4). For the overall study population (N = 121 eyes), the proportions of study eyes gaining ≥10 and ≥15 approxETDRS letters at 12 months were 31.4% and 24.0%, respectively, whereas the proportions of eyes losing ≥10 and ≥15 approxETDRS letters were 10.8% and 8.3%, respectively (LOCF analysis) (Table 2). Response rates calculated from observed data (n = 85 eyes) for the overall study population were similar, with 29.4% and 21.2% of study eyes gaining ≥10 and ≥15 approxETDRS letters, respectively, and 14.1% and 10.6% of study eyes losing ≥10 and ≥15 approxETDRS letters, respectively, at 12 months.
      Figure thumbnail gr3
      Figure 3Mean corrected visual acuity and mean change in corrected visual acuity from baseline for all study eyes (Top) and study eyes with regular follow-up (Bottom). ETDRS = Early Treatment Diabetic Retinopathy Study; LOCF = last-observation-carried-forward.
      Figure thumbnail gr4
      Figure 4Individual changes in corrected visual acuity from baseline to 12 months for all study eyes (last-observation-carried-forward analysis).
      Table 2Visual Acuity Outcomes at 12 Months (Last-Observation-Carried-Forward Analysis)
      All Eyes (N = 121)Eyes Categorized by Number of Anti-VEGF InjectionsEyes Categorized by Adjunctive Therapy
      ≤3 (N = 83)4–6 (N = 27)≥7 (N = 11)Laser Therapy (N = 33)No Laser Therapy (N = 88)
      CVA (approxETDRS letters), mean (SD)
       Baseline56.9 (11.6)56.2 (11.9)58.4 (11.1)58.7 (11.2)53.1 (12.3)58.3 (11.1)
       12 months61.6 (15.4)61.2 (15.5)60.8 (15.7)66.5 (14.6)56.2 (14.4)63.6 (14.8)
       Change from baseline+4.7 (12.3)+5.0 (12.2)+2.4 (12.1)+7.7 (14.0)+3.1 (14.3)+5.3 (11.5)
      Visual acuity response rate, n (% eyes)
       Eyes gaining ≥10 letters38 (31.4)25 (30.1)7 (25.9)6 (54.5)10 (30.3)28 (31.8)
       Eyes gaining ≥15 letters29 (24.0)18 (21.7)7 (25.9)4 (36.4)7 (21.2)22 (25.0)
       Eyes losing ≥10 letters13 (10.8)9 (10.8)3 (11.1)1 (9.1)6 (18.2)7 (8.0)
       Eyes losing ≥15 letters10 (8.3)6 (7.2)2 (7.4)1 (9.1)3 (9.1)7 (8.0)
      approxETDRS = approximate Early Treatment Diabetic Retinopathy Study; CVA = corrected visual acuity; VEGF = vascular endothelial growth factor.
      Unadjusted analysis of visual acuity outcome provided no clear evidence of a relationship with anti-VEGF injection frequency. Eyes receiving ≤3 injections, 4–6 injections, and ≥7 injections over 12 months showed mean improvements in CVA at 12 months of 5.0, 2.4, and 7.7 approxETDRS letters, respectively (Table 2). Although the likelihood of achieving an improvement in CVA of ≥10 and ≥15 approxETDRS letters at 12 months appeared to be higher in eyes receiving ≥7 anti-VEGF injections than in those receiving fewer injections, the proportion of study eyes that lost ≥10 and ≥15 approxETDRS letters at 12 months (overall ∼10% and 8%, respectively) did not vary appreciably with anti-VEGF injection frequency (Table 2).
      Multivariate linear and logistic regression analyses adjusting for data clustering provided no evidence of a statistically significant association between intravitreal anti-VEGF injection frequency (included as either a continuous or categorical exposure variable) and visual response outcome at 12 months (CVA, change in CVA from baseline, and proportion of study with ≥10 approxETDRS letters improvement). After adjusting for covariates, baseline CVA was positively associated (P < .05) with CVA at 12 months and negatively associated (P < .05) with ≥10 approxETDRS letters improvement at 12 months. The number of ophthalmologist visits over the first 12 months consistently showed a positive association (P < .05) with CVA, change in CVA from baseline, and ≥10 ETDRS letters improvement at 12 months.
      Adjunctive laser photocoagulation therapy was administered to 33 (27.3%) study eyes during the first 6 months following the index intravitreal anti-VEGF injection. This subset of study eyes had a lower baseline CVA than eyes that received anti-VEGF monotherapy during this period (n = 88; 72.7%) (mean 53.1 vs 58.3 approxETDRS letters). Nevertheless, for the adjunctive laser cohort the improvement in CVA at 12 months (mean +3.1 approxETDRS letters, achieved with a mean of 2.5 anti-VEGF injections) was similar to that recorded in the subset of eyes that received anti-VEGF monotherapy (mean +5.3 approxETDRS letters, achieved with a mean of 3.3 anti-VEGF injections) (Table 2). Visual response rates were also similar in the 2 groups, with 7 (21.2%) adjunctive laser-treated eyes gaining ≥15 approxETDRS letters compared with 22 (25.0%) non–laser-treated eyes, and 3 (9.1%) adjunctive laser-treated eyes losing ≥15 approxETDRS letters compared with 7 (8.0%) non–laser-treated eyes (Table 2).

      Discussion

      In this retrospective analysis of EMRs from a large integrated healthcare system database, we identified 336 patients in the clinical practice setting who initiated intravitreal anti-VEGF therapy (predominantly bevacizumab) for DME from 2007 onward. Although not specifically formulated for ophthalmic use, intravitreal bevacizumab has been widely used off-label since 2006, and in a 2014 survey by the American Society of Retina Specialists was reported as the treatment of choice for DME in US clinical practice.
      • Rezaei K.A.
      • Stone T.W.
      2014 Global trends in retina survey.
      The initial DME diagnosis could occur at any time (from 2004 onward) prior to the index intravitreal injection, and there was no requirement for patients to be newly diagnosed on initiating anti-VEGF therapy. Healthcare resource utilization and visual acuity outcome data were obtained for 110 eligible patients (121 eyes). Our findings indicated that, despite the high overall frequency of ophthalmology subspecialty visits (mean 9.2 over 12 months), only 59% of study eyes had regular (at least quarterly) visits, and fewer than 2% had monthly visits. Most study eyes (69%) received 3 or fewer anti-VEGF injections over the 12-month study period, and only 3% of study eyes received injections at a frequency approaching monthly dosing. This finding is consistent with national trends revealed in contemporaneous administrative claims analyses, which indicate that patients initiating anti-VEGF therapy for newly diagnosed DME receive ∼2 to 4 intravitreal injections during the first year (Kiss S, et al. JMCP 2014;20[10a Suppl]: S21–22, Abstract E1).
      • Kiss S.
      • Liu Y.
      • Brown J.
      • et al.
      Clinical utilization of anti-vascular endothelial growth-factor agents and patient monitoring in retinal vein occlusion and diabetic macular edema.
      • VanderBeek B.L.
      • Shah N.
      • Parikh P.C.
      • Ma L.
      Trends in the care of diabetic macular edema: analysis of a national cohort.
      In contrast, in the majority of phase 3 clinical trials of anti-VEGF therapy in DME, patients were monitored monthly and the mean/median number of intravitreal injections administered in the first year of treatment was more than 2-fold higher.
      • Massin P.
      • Bandello F.
      • Garweg J.G.
      • et al.
      Safety and efficacy of ranibizumab in diabetic macular edema (RESOLVE Study): a 12-month, randomized, controlled, double-masked, multicenter phase II study.
      • Mitchell P.
      • Bandello F.
      • Schmidt-Erfurth U.
      • et al.
      The RESTORE study: ranibizumab monotherapy or combined with laser versus laser monotherapy for diabetic macular edema.
      • Nguyen Q.D.
      • Brown D.M.
      • Marcus D.M.
      • et al.
      Ranibizumab for diabetic macular edema: results from 2 phase III randomized trials: RISE and RIDE.
      • Elman M.J.
      • Aiello L.P.
      • Beck R.W.
      • et al.
      Randomized trial evaluating ranibizumab plus prompt or deferred laser or triamcinolone plus prompt laser for diabetic macular edema.
      • Michaelides M.
      • Kaines A.
      • Hamilton R.D.
      • et al.
      A prospective randomized trial of intravitreal bevacizumab or laser therapy in the management of diabetic macular edema (BOLT study) 12-month data: report 2.
      • Korobelnik J.F.
      • Do D.V.
      • Schmidt-Erfurth U.
      • et al.
      Intravitreal aflibercept for diabetic macular edema.
      • Ishibashi T.
      • Li X.
      • Koh A.
      • et al.
      The REVEAL Study: ranibizumab monotherapy or combined with laser versus laser monotherapy in Asian patients with diabetic macular edema.
      • The Diabetic Retinopathy Clinical Research Network
      Aflibercept, bevacizumab, or ranibizumab for diabetic macular edema.
      Anti-VEGF therapy appeared to be initiated at a similar stage of disease progression in the clinical practice setting (mean baseline CVA 56.9 approxETDRS letters) as in the clinical trial setting (mean baseline best-corrected visual acuity [BCVA] ∼55–65 ETDRS letters),
      • Massin P.
      • Bandello F.
      • Garweg J.G.
      • et al.
      Safety and efficacy of ranibizumab in diabetic macular edema (RESOLVE Study): a 12-month, randomized, controlled, double-masked, multicenter phase II study.
      • Mitchell P.
      • Bandello F.
      • Schmidt-Erfurth U.
      • et al.
      The RESTORE study: ranibizumab monotherapy or combined with laser versus laser monotherapy for diabetic macular edema.
      • Nguyen Q.D.
      • Brown D.M.
      • Marcus D.M.
      • et al.
      Ranibizumab for diabetic macular edema: results from 2 phase III randomized trials: RISE and RIDE.
      • Elman M.J.
      • Aiello L.P.
      • Beck R.W.
      • et al.
      Randomized trial evaluating ranibizumab plus prompt or deferred laser or triamcinolone plus prompt laser for diabetic macular edema.
      • Michaelides M.
      • Kaines A.
      • Hamilton R.D.
      • et al.
      A prospective randomized trial of intravitreal bevacizumab or laser therapy in the management of diabetic macular edema (BOLT study) 12-month data: report 2.
      • Korobelnik J.F.
      • Do D.V.
      • Schmidt-Erfurth U.
      • et al.
      Intravitreal aflibercept for diabetic macular edema.
      • The Diabetic Retinopathy Clinical Research Network
      Aflibercept, bevacizumab, or ranibizumab for diabetic macular edema.
      assuming numerical parity between CVA and BCVA indices.
      Several factors may account for the irregular ophthalmologic monitoring and low frequency of anti-VEGF administration among DME patients in clinical practice. Demands on patients' time, particularly in the case of younger patients of working age, and the health burden of other comorbidities may prevent regular attendance at ophthalmology appointments. Patients with DME are at increased risk of diabetes-related complications compared with those without DME. Administrative claims data indicate that patients of working age with newly diagnosed DME make on average more than 25 outpatient visits each year, with most of these being for nonophthalmologic reasons.
      • Wallick C.J.
      • Hansen R.N.
      • Campbell J.
      • Kiss S.
      • Kowalski J.W.
      • Sullivan S.D.
      Comorbidity and health care resource use among commercially insured non-elderly patients with diabetic macular edema.
      In addition, patient copayments, lack of patient motivation, transport costs, and extended travel times may discourage clinic attendance. Given these constraints, ophthalmologists and/or patients may choose to extend the interval between successive intravitreal injections.
      This pattern of infrequent treatment and monitoring may compromise clinical outcomes. Phase 3 clinical trials of anti-VEGF agents in DME, typically employing monthly monitoring and near-monthly treatment over the first 12 months, have achieved seemingly greater improvements in visual acuity (Table 3). For example, in the RISE and RIDE trials, in which a mean of ∼11 intravitreal ranibizumab injections were administered over the first 12 months, BCVA improved by an average of 11.9 and 12.0 ETDRS letters, respectively, at 24 months, and 39% and 46%, respectively, of study eyes showed ≥15-letter BCVA gain.
      • Nguyen Q.D.
      • Brown D.M.
      • Marcus D.M.
      • et al.
      Ranibizumab for diabetic macular edema: results from 2 phase III randomized trials: RISE and RIDE.
      Likewise, in the RESOLVE, PROTOCOL I, BOLT, VISTA, and VIVID trials,
      • Massin P.
      • Bandello F.
      • Garweg J.G.
      • et al.
      Safety and efficacy of ranibizumab in diabetic macular edema (RESOLVE Study): a 12-month, randomized, controlled, double-masked, multicenter phase II study.
      • Elman M.J.
      • Aiello L.P.
      • Beck R.W.
      • et al.
      Randomized trial evaluating ranibizumab plus prompt or deferred laser or triamcinolone plus prompt laser for diabetic macular edema.
      • Michaelides M.
      • Kaines A.
      • Hamilton R.D.
      • et al.
      A prospective randomized trial of intravitreal bevacizumab or laser therapy in the management of diabetic macular edema (BOLT study) 12-month data: report 2.
      • Korobelnik J.F.
      • Do D.V.
      • Schmidt-Erfurth U.
      • et al.
      Intravitreal aflibercept for diabetic macular edema.
      in which patients received a mean of ∼9 to 12 intravitreal anti-VEGF injections over the first 12 months, BCVA improved on average by ∼8 to 12 ETDRS letters at 12 months, and ∼30% to 60% of study eyes achieved ≥10-letter gain in BCVA. Visual acuity outcomes in the present analysis appeared to be substantially less favorable, as reflected in a CVA gain of (mean) 4.7 approxETDRS letters at 12 months and response rates (proportions of study eyes gaining ≥10 or ≥15 approxETDRS letters) of 31.4% and 24.0%, respectively. In addition, the proportions of eyes losing ≥10 or ≥15 approxETDRS letters (10.8% and 8.3%, respectively) were ∼2-fold higher than in the phase 3 clinical trials of ranibizumab (typically <5%). Comparisons of visual acuity outcomes between the 2 settings should, however, be interpreted with caution. ApproxETDRS letter scores obtained from conversion of Snellen fractions cannot be assumed to be equivalent to “true” ETDRS letter scores obtained using a standardized visual acuity test protocol and ETDRS chart.
      • Gregori N.Z.
      • Feuer W.
      • Rosenfeld P.J.
      Novel method for analyzing Snellen visual acuity measurements.
      Moreover, patients in real-world clinical practice are demographically and clinically more diverse than the defined, homogenous patient populations of randomized clinical trials. Although the visual acuity inclusion criterion selected for this study (baseline CVA of 20/40 to 20/320) was intended to mirror that of the major randomized clinical trials, it is unclear how closely the CVA values from clinical practice approximate to the BCVA values from clinical trials. Apart from the inherent imprecision of Snellen chart measurements compared to visual acuity measurements obtained with an ETDRS chart,
      • Dong L.M.
      • Marsh M.J.
      • Hawkins B.S.
      Measurement and analysis of visual acuity in multicenter randomized clinical trials in the United States: findings from a survey.
      the degree of visual acuity correction afforded by patients' habitual glasses and contact lenses is likely to fall short of their current refraction and the standardized refraction of a clinical trial. Moreover, an efficacy trial can often overestimate treatment effect in real-world practice.
      Table 3Visual Acuity Gains and Response Rates in Landmark Clinical Trials of Anti–Vascular Endothelial Growth Factor Therapy in Diabetic Macular Edema
      StudyMean Injections at 12 MonthsMonitoring FrequencyMean ETDRS Letter GainPatients With ≥10* or ≥15** ETDRS Letter Gain, %Patients With ≥10* or ≥15** ETDRS Letter Loss, %
      RISE
      • Nguyen Q.D.
      • Brown D.M.
      • Marcus D.M.
      • et al.
      Ranibizumab for diabetic macular edema: results from 2 phase III randomized trials: RISE and RIDE.
      (24 months)
      ∼11 (RAN)Monthly11.939**2.4**
      RIDE
      • Nguyen Q.D.
      • Brown D.M.
      • Marcus D.M.
      • et al.
      Ranibizumab for diabetic macular edema: results from 2 phase III randomized trials: RISE and RIDE.
      (24 months)
      ∼11 (RAN)Monthly12.046**3.9**
      PROTOCOL I
      • Elman M.J.
      • Aiello L.P.
      • Beck R.W.
      • et al.
      Randomized trial evaluating ranibizumab plus prompt or deferred laser or triamcinolone plus prompt laser for diabetic macular edema.
      (12 months)
      9 (RAN)Monthly950*4*
      REVEAL
      • Ishibashi T.
      • Li X.
      • Koh A.
      • et al.
      The REVEAL Study: ranibizumab monotherapy or combined with laser versus laser monotherapy in Asian patients with diabetic macular edema.
      (12 months)
      7.8 (RAN)Monthly6.6NANA
      RESTORE
      • Mitchell P.
      • Bandello F.
      • Schmidt-Erfurth U.
      • et al.
      The RESTORE study: ranibizumab monotherapy or combined with laser versus laser monotherapy for diabetic macular edema.
      (12 months)
      7.0 (RAN)Monthly6.837*3.5*
      RESOLVE
      • Massin P.
      • Bandello F.
      • Garweg J.G.
      • et al.
      Safety and efficacy of ranibizumab in diabetic macular edema (RESOLVE Study): a 12-month, randomized, controlled, double-masked, multicenter phase II study.
      (12 months)
      10.2 (RAN)Monthly10.261*4.9*
      BOLT
      • Michaelides M.
      • Kaines A.
      • Hamilton R.D.
      • et al.
      A prospective randomized trial of intravitreal bevacizumab or laser therapy in the management of diabetic macular edema (BOLT study) 12-month data: report 2.
      (12 months)
      9 (BEV)Every 6 weeks831*2.4**
      VISTA
      • Korobelnik J.F.
      • Do D.V.
      • Schmidt-Erfurth U.
      • et al.
      Intravitreal aflibercept for diabetic macular edema.
      (12 months)
      11.8 (AFL)Monthly12.541.6**0.6**
      8.4 (AFL)Monthly (first 5 months), then bimonthly10.731.1**0.7**
      VIVID
      • Korobelnik J.F.
      • Do D.V.
      • Schmidt-Erfurth U.
      • et al.
      Intravitreal aflibercept for diabetic macular edema.
      (12 months)
      12.2 (AFL)Monthly10.532.4**0.7**
      8.7 (AFL)Monthly (first 5 months), then bimonthly10.733.3**0**
      PROTOCOL T
      • The Diabetic Retinopathy Clinical Research Network
      Aflibercept, bevacizumab, or ranibizumab for diabetic macular edema.
      (12 months)
      10 (BEV)
      Median.


      10 (RAN)
      Median.


      9 (AFL)
      Median.
      Monthly7.5
      Patients with baseline BCVA 69–78 ETDRS letters.
      16
      Patients with baseline BCVA 69–78 ETDRS letters.
      **
      1
      Patients with baseline BCVA 69–78 ETDRS letters.
      **
      11.8
      Patients with baseline BCVA <69 ETDRS letters.
      41
      Patients with baseline BCVA <69 ETDRS letters.
      **
      2
      Patients with baseline BCVA <69 ETDRS letters.
      **
      8.3
      Patients with baseline BCVA 69–78 ETDRS letters.
      15
      Patients with baseline BCVA 69–78 ETDRS letters.
      **
      1
      Patients with baseline BCVA 69–78 ETDRS letters.
      **
      14.2
      Patients with baseline BCVA <69 ETDRS letters.
      50
      Patients with baseline BCVA <69 ETDRS letters.
      **
      2
      Patients with baseline BCVA <69 ETDRS letters.
      **
      8.0
      Patients with baseline BCVA 69–78 ETDRS letters.
      18
      Patients with baseline BCVA 69–78 ETDRS letters.
      **
      2
      Patients with baseline BCVA 69–78 ETDRS letters.
      **
      18.9
      Patients with baseline BCVA <69 ETDRS letters.
      67
      Patients with baseline BCVA <69 ETDRS letters.
      **
      1
      Patients with baseline BCVA <69 ETDRS letters.
      **
      AFL = aflibercept; BCVA = best-corrected visual acuity; BEV = bevacizumab; ETDRS = Early Treatment Diabetic Retinopathy Study; NA = not available; RAN = ranibizumab.
      a Median.
      b Patients with baseline BCVA 69–78 ETDRS letters.
      c Patients with baseline BCVA <69 ETDRS letters.
      A strength of this study is its use of EMR data from a large integrated health system, albeit one confined to a specific geographic region. Despite this geographic restriction, the treatment patterns recorded in the Geisinger patient sample are consistent with those revealed in EMR and administrative claims analyses of clinical-practice patients receiving anti-VEGF therapy for DME across the United States (Kiss S, et al. JMCP 2014;20[10a Suppl]:S21–22, Abstract E1).
      • Kiss S.
      • Liu Y.
      • Brown J.
      • et al.
      Clinical utilization of anti-vascular endothelial growth-factor agents and patient monitoring in retinal vein occlusion and diabetic macular edema.
      • VanderBeek B.L.
      • Shah N.
      • Parikh P.C.
      • Ma L.
      Trends in the care of diabetic macular edema: analysis of a national cohort.
      • Fong D.S.
      • Luong T.Q.
      • Contreras R.
      • et al.
      Treatment patterns and two-year vision outcomes with bevacizumab in diabetic macular edema: an analysis from a large U.S. integrated health care system.
      To facilitate comparisons with the populations enrolled in landmark randomized clinical trials of anti-VEGF agents, the study included only a proportion of the DME patients in the Geisinger database, and may therefore have been subject to selection bias. In addition, the study sample was relatively small; this, combined with the heavily skewed distribution of eyes toward the lower end of the injection frequency range, complicated attempts to establish an association between injection frequency and visual acuity outcome. As with all administrative claims analyses, errors may arise through coding inaccuracies. However, assuming that the claims data submitted by physicians are accurate, patients with DME can be reliably identified using ICD-9-CM diagnosis codes from administrative claims databases. The combination of ICD-9-CM codes 250.xx (diabetes mellitus) and 362.53 (cystoid macular degeneration) reportedly shows high specificity and sensitivity in identifying DME, as well as close agreement with medical records.
      • Bearelly S.
      • Mruthyunjaya P.
      • Tzeng J.P.
      • et al.
      Identification of patients with diabetic macular edema from claims data: a validation study.
      Other potential limitations include the lack of information about factors such as DME status (duration of edema, presence of ischemia, hard exudates, microvascular abnormalities), diabetes control, and prior DME treatments that may influence vision outcomes. As with previous real-world studies of anti-VEGF use in DME (Kiss S, et al. JMCP 2014;20[10a Suppl]:S21–22, Abstract E1),
      • Kiss S.
      • Liu Y.
      • Brown J.
      • et al.
      Clinical utilization of anti-vascular endothelial growth-factor agents and patient monitoring in retinal vein occlusion and diabetic macular edema.
      • VanderBeek B.L.
      • Shah N.
      • Parikh P.C.
      • Ma L.
      Trends in the care of diabetic macular edema: analysis of a national cohort.
      • Fong D.S.
      • Luong T.Q.
      • Contreras R.
      • et al.
      Treatment patterns and two-year vision outcomes with bevacizumab in diabetic macular edema: an analysis from a large U.S. integrated health care system.
      the present study describes treatment practices operating prior to publication of the landmark phase 3 clinical trials of monthly ranibizumab in DME (the RIDE/RISE paper was published in 2012)
      • Nguyen Q.D.
      • Brown D.M.
      • Marcus D.M.
      • et al.
      Ranibizumab for diabetic macular edema: results from 2 phase III randomized trials: RISE and RIDE.
      and FDA approval (2015) of aflibercept. Subsequent treatment practice may have altered in light of evidence from these randomized clinical trials and the availability of alternative anti-VEGF agents.
      In conclusion, this analysis suggests that the less frequent anti-VEGF injection schedules used during the first year of treatment of DME in clinical practice result in less favorable visual outcomes compared with those seen over the same period in randomized clinical trials. Further research is needed to assess factors influencing utilization of anti-VEGF therapy for DME in clinical settings, and real-world patient characteristics that may affect vision outcomes.
      Funding/Support: This study was sponsored by Allergan plc (Dublin, Ireland). Allergan plc (Irvine, California, USA) participated in the design of the study, data analysis, and data interpretation, and also supervised the preparation, review, and approval of the manuscript. Financial Disclosures: Nancy M. Holekamp and Szilárd Kiss have received grant support and consultancy fees, and Arghavan Almony has received consultancy and speaker fees, from Allergan plc (Irvine, California, USA). Joanna Campbell and Hitesh Chandwani are employees of Allergan plc (Irvine, California, USA). Ashley L. Cole was an employee of Allergan plc at the time of this work, and is currently an employee of Truven Health Analytics (Durham, North Carolina, USA). Herbert Ingraham and Steven Marks are employees of Geisinger Health System (Danville, Pennsylvania, USA). All authors attest that they meet the current ICMJE criteria for authorship.

      Supplemental Data

      Figure thumbnail figs1
      Dr Nancy M. Holekamp, MD is Professor of Clinical Ophthalmology and Visual Sciences at the Washington University School of Mediscine in St. Louis, Missouri, and the Director of Retina Services at Pepose Vision Institute in St. Louis. She completed her ophthalmology residency at the Washington University School of Medicine, and vitreoretinal surgery fellowship with the Retina Consultants in St. Louis. She is actively involved in clinical research, and has 70 peer-reviewed publications, and 21 book chapters.

      References

        • Centers for Disease Control and Prevention
        National Diabetes Statistics Report: Estimates of Diabetes and Its Burden in the United States, 2014.
        (Available at:)
        • Liew G.
        • Michaelides M.
        • Bunce C.
        A comparison of the causes of blindness certifications in England and Wales in working age adults (16-64 years), 1999-2000 with 2009-2010.
        BMJ Open. 2014; 4: e004015
        • Buch H.
        • Vinding T.
        • La Cour M.
        • Appleyard M.
        • Jensen G.B.
        • Nielsen N.V.
        Prevalence and causes of visual impairment and blindness among 9980 Scandinavian adults: the Copenhagen City Eye Study.
        Ophthalmology. 2004; 111: 53-61
        • Antonetti D.A.
        • Klein R.
        • Gardner T.W.
        Diabetic retinopathy.
        N Engl J Med. 2012; 366: 1227-1239
        • Yau J.W.
        • Rogers S.L.
        • Kawasaki R.
        • et al.
        Global prevalence and major risk factors of diabetic retinopathy.
        Diabetes Care. 2012; 35: 556-564
        • National Eye Institute
        Diabetic retinopathy.
        (Available at:)
        • Varma R.
        • Bressler N.M.
        • Doan Q.V.
        • et al.
        Prevalence of and risk factors for diabetic macular edema in the United States.
        JAMA Ophthalmol. 2014; 132: 1334-1340
        • Antcliff R.J.
        • Marshall J.
        The pathogenesis of edema in diabetic maculopathy.
        Semin Ophthalmol. 1999; 14: 223-232
        • Zhang X.
        • Zeng H.
        • Bao S.
        • Wang N.
        • Gillies M.C.
        Diabetic macular edema: new concepts in patho-physiology and treatment.
        Cell Biosci. 2014; 4: 27
        • Ehrlich R.
        • Harris A.
        • Ciulla T.A.
        • Kheradiya N.
        • Winston D.M.
        • Wirostko B.
        Diabetic macular oedema: physical, physiological and molecular factors contribute to this pathological process.
        Acta Ophthalmol. 2010; 88: 279-291
        • Ferrara N.
        • Gerber H.P.
        • LeCouter J.
        The biology of VEGF and its receptors.
        Nature Med. 2003; 9: 669-676
        • Qaum T.
        • Xu Q.
        • Joussen A.M.
        • et al.
        VEGF-initiated blood-retinal barrier breakdown in early diabetes.
        Invest Ophthalmol Vis Sci. 2001; 42: 2408-2413
        • Ford J.A.
        • Lois N.
        • Royle P.
        • Clar C.
        • Shyangdan D.
        • Waugh N.
        Current treatments in diabetic macular oedema: systematic review and meta-analysis.
        BMJ Open. 2013; 3: e002269
        • Bandello F.
        • Cunha-Vaz J.
        • Chong N.V.
        • et al.
        New approaches for the treatment of diabetic macular oedema: recommendations by an expert panel.
        Eye (Lond). 2012; 26: 485-493
        • Massin P.
        • Bandello F.
        • Garweg J.G.
        • et al.
        Safety and efficacy of ranibizumab in diabetic macular edema (RESOLVE Study): a 12-month, randomized, controlled, double-masked, multicenter phase II study.
        Diabetes Care. 2010; 33: 2399-2405
        • Mitchell P.
        • Bandello F.
        • Schmidt-Erfurth U.
        • et al.
        The RESTORE study: ranibizumab monotherapy or combined with laser versus laser monotherapy for diabetic macular edema.
        Ophthalmology. 2011; 118: 615-625
        • Nguyen Q.D.
        • Brown D.M.
        • Marcus D.M.
        • et al.
        Ranibizumab for diabetic macular edema: results from 2 phase III randomized trials: RISE and RIDE.
        Ophthalmology. 2012; 119: 789-801
        • Elman M.J.
        • Aiello L.P.
        • Beck R.W.
        • et al.
        Randomized trial evaluating ranibizumab plus prompt or deferred laser or triamcinolone plus prompt laser for diabetic macular edema.
        Ophthalmology. 2010; 117: 1064-1077 e1035
        • Elman M.J.
        • Bressler N.M.
        • Qin H.
        • et al.
        Expanded 2-year follow-up of ranibizumab plus prompt or deferred laser or triamcinolone plus prompt laser for diabetic macular edema.
        Ophthalmology. 2011; 118: 609-614
        • Michaelides M.
        • Kaines A.
        • Hamilton R.D.
        • et al.
        A prospective randomized trial of intravitreal bevacizumab or laser therapy in the management of diabetic macular edema (BOLT study) 12-month data: report 2.
        Ophthalmology. 2010; 117: 1078-1086 e1072
        • Korobelnik J.F.
        • Do D.V.
        • Schmidt-Erfurth U.
        • et al.
        Intravitreal aflibercept for diabetic macular edema.
        Ophthalmology. 2014; 121: 2247-2254
        • Ishibashi T.
        • Li X.
        • Koh A.
        • et al.
        The REVEAL Study: ranibizumab monotherapy or combined with laser versus laser monotherapy in Asian patients with diabetic macular edema.
        Ophthalmology. 2015; 122: 1402-1415
        • Kiss S.
        • Liu Y.
        • Brown J.
        • et al.
        Clinical utilization of anti-vascular endothelial growth-factor agents and patient monitoring in retinal vein occlusion and diabetic macular edema.
        Clin Ophthalmol. 2014; 8: 1611-1621
        • VanderBeek B.L.
        • Shah N.
        • Parikh P.C.
        • Ma L.
        Trends in the care of diabetic macular edema: analysis of a national cohort.
        PloS One. 2016; 11: e0149450
        • Gregori N.Z.
        • Feuer W.
        • Rosenfeld P.J.
        Novel method for analyzing Snellen visual acuity measurements.
        Retina. 2010; 30: 1046-1050
        • Rezaei K.A.
        • Stone T.W.
        2014 Global trends in retina survey.
        American Society of Retina Specialists, Chicago, IL2014 (Available at:)
        • The Diabetic Retinopathy Clinical Research Network
        Aflibercept, bevacizumab, or ranibizumab for diabetic macular edema.
        N Engl J Med. 2015; 372: 1193-1203
        • Wallick C.J.
        • Hansen R.N.
        • Campbell J.
        • Kiss S.
        • Kowalski J.W.
        • Sullivan S.D.
        Comorbidity and health care resource use among commercially insured non-elderly patients with diabetic macular edema.
        OSLI Retina. 2015; 46: 744-751
        • Dong L.M.
        • Marsh M.J.
        • Hawkins B.S.
        Measurement and analysis of visual acuity in multicenter randomized clinical trials in the United States: findings from a survey.
        Ophthalmic Epidemiol. 2003; 10: 149-165
        • Fong D.S.
        • Luong T.Q.
        • Contreras R.
        • et al.
        Treatment patterns and two-year vision outcomes with bevacizumab in diabetic macular edema: an analysis from a large U.S. integrated health care system.
        Retina. 2017; https://doi.org/10.1097/IAE.0000000000001790
        • Bearelly S.
        • Mruthyunjaya P.
        • Tzeng J.P.
        • et al.
        Identification of patients with diabetic macular edema from claims data: a validation study.
        Arch Ophthalmol. 2008; 126: 986-989

      Linked Article