CLINICAL OUTCOMES AND TREATMENT COURSE OF EYES WITH NEOVASCULAR AGE-RELATED MACULAR DEGENERATION FOLLOWING THE DEVELOPMENT OF ENDOPHTHALMITIS.

PURPOSE: To evaluate the clinical course of patients with neovascular age-related macular degeneration (nAMD) after developing endophthalmitis during their treatment with intravitreal injections. METHODS: Multicenter, retrospective series. RESULTS: From April 2013 to October 2018, 196,598 intravitreal anti-vascular endothelial growth factor (VEGF) injections were performed, with 75 cases of endophthalmitis (incidence 0.0381%). There was no association between intravitreal anti-VEGF drug (P = 0.29), anesthetic method (P = 0.26), povidone concentration (P = 0.22), or any intraprocedure variable and endophthalmitis incidence. Seventy-two patients (96%) were treated with intravitreal tap and inject , while 3 underwent immediate pars plana vitrectomy. After endophthalmitis resolution, 17 patients (22.7%) were not re-treated for nAMD (in 10 cases due to inactive disease; follow-up, 115 ± 8.4 weeks). Patients required less frequent anti-VEGF injections after infection (7.4 ± 0.61 weeks vs. 11.5 ± 1.8 weeks; P = 0.004). Preinfection logarithm of the minimum angle of resolution visual acuity was 0.585 ± 0.053 (∼20/77). It worsened with endophthalmitis (1.67 ± 0.08, ∼20/935; P < 0.001) and again on postendophthalmitis treatment day 1 (1.94 ± 0.064; count fingers; P < 0.001), but improved after reinitiating nAMD therapy (1.02 ± 0.11; ∼20/209; P < 0.001). Better visual acuity on postendophthalmitis week 1 (P = 0.002) and reinitiation of nAMD treatment (P = 0.008) were associated with better final visual acuity, and streptococcal culture with worse visual acuity (P = 0.028). The postendophthalmitis treatment interval was associated with the anti-VEGF drug used (aflibercept = ranibizumab > bevacizumab; P < 0.001). CONCLUSION: Patients with nAMD required fewer injections after endophthalmitis, suggesting a biological change in disease activity. Neovascular age-related macular degeneration became quiescent in 13.3% of eyes. Most achieved better outcomes with anti-VEGF reinitiation.

Predictors of Endophthalmitis after Intravitreal Injection: A Multivariable Analysis Based on Injection Protocol and Povidone Iodine Strength.

PURPOSE: To determine the incidence of endophthalmitis after anti-vascular endothelial growth factor (VEGF) therapy at our institution and to identify potential risk factors for endophthalmitis occurring after injection. DESIGN: Retrospective, single-center cohort study. PARTICIPANTS: All patients who received an intravitreal injection of an anti-VEGF medication between January 1, 2014, and March 31, 2017. METHODS: Current Procedural Terminology and International Classification of Diseases billing codes were used to identify instances of anti-VEGF administration and cases of endophthalmitis. Medical records and injection technique were reviewed carefully in each case. Multivariable logistic regression analysis was performed in a stepwise fashion to determine independent predictors of endophthalmitis based on injection protocol. MAIN OUTCOME MEASURES: Incidence of endophthalmitis after injection and odds of endophthalmitis by injection technique with 95% confidence intervals (CIs). RESULTS: A total of 154 198 anti-VEGF injections were performed during the period of interest, resulting in 58 cases of endophthalmitis (0.038% [1:2659]). After adjustment for confounders, both 2% lidocaine jelly (odds ratio [OR], 11.28; 95% CI, 3.39-37.46; P < 0.001) and 0.5% Tetravisc (Ocusoft, Richmond, TX; OR, 3.95; 95% CI, 1.15-13.50; P = 0.03) use were independent risk factors for endophthalmitis after injection. Lid speculum use, povidone iodine strength (5% vs. 10%), injection location (superior or inferior), conjunctival displacement, use of provider gloves, use of a strict no-talking policy, use of subconjunctival lidocaine, and topical antibiotic use were not statistically significant predictors of endophthalmitis after injection. There was no difference in endophthalmitis rate among the anti-VEGF agents (bevacizumab, ranibizumab 0.3 mg, ranibizumab 0.5 mg, and aflibercept). CONCLUSIONS: The incidence of endophthalmitis after anti-VEGF injections is low. Use of lidocaine jelly or Tetravisc may increase the risk of endophthalmitis after injection.

Efficacy and safety of two or more dexamethasone intravitreal implant injections for treatment of macular edema related to retinal vein occlusion (Shasta study).

PURPOSE: To evaluate the efficacy, safety, and reinjection interval of dexamethasone intravitreal implant (DEX implant) in branch retinal vein occlusion and central retinal vein occlusion patients receiving ≥ 2 DEX implant treatments. METHODS: Multicenter (26-site), retrospective chart review study. Data were collected from baseline (at first DEX implant) through 3 months to 6 months after last DEX implant. RESULTS: Patients (n = 289) received 2 to 9 (mean, 3.2) DEX implants as monotherapy (29.1% of patients) or with adjunctive treatments/procedures. Mean duration of macular edema before first DEX implant was 18.4 months. Mean reinjection interval was 5.6 months. Mean peak change in best-corrected visual acuity from baseline through 4 weeks to 20 weeks after final DEX implant was +1.0 line (P < 0.001). Best-corrected visual acuity and central retinal thickness improved significantly from baseline after each of the first 6 DEX implant injections (P ≤ 0.037); 59.7% of branch retinal vein occlusion and 66.7% of central retinal vein occlusion patients achieved ≥ 2-line best-corrected visual acuity improvement. Intraocular pressure increase (≥ 10 mmHg) occurred in 32.6% of patients; 29.1% used intraocular pressure-lowering medication to treat increases associated with DEX implant. Only 1.7% of patients required incisional glaucoma surgery. CONCLUSION: Retinal vein occlusion patients treated with multiple DEX implant injections, either alone or combined with other therapies, had improved central retinal thickness and visual acuity with each subsequent injection. No new safety concerns developed with multiple implants.

Central cone dysfunction in autosomal dominant vitreoretino choroidopathy (ADVIRC).

PURPOSE: To describe a patient with clinically documented autosomal dominant vitreoretinochoroidopathy who has had 11 years of progression from initial description and now demonstrates evidence of central cone dysfunction. DESIGN: Case report. METHODS: The patient is a member of a pedigree described in the literature. This is a case report format that follows standard clinical studies. RESULTS: The patient had normal full field electroretinography results but focally reduced multifocal electroretinography results and evidence of macular atrophy on optical coherence tomography. CONCLUSION: Autosomal dominant vitreoretinochoroidopathy may result in central cone dysfunction because of macular atrophy late in the course of the disease, although electroretinography and visual fields remain stable over extended follow-up periods.

Late development of vitelliform lesions and flecks in a patient with best disease: clinicopathologic correlation.

OBJECTIVE: To provide the clinicopathologic findings of a patient who developed the clinical characteristics of Best disease (typically considered a juvenile macular degeneration) at the age of 75 years after being documented to be ophthalmoscopically normal at the age of 51 years. DESIGN: A member of a large family with Best disease, possessing a Y227N mutation in the VMD2 gene (the gene responsible for the disease, which encodes the bestrophin protein), developed small vitelliform lesions in both eyes at the age of 75 years and later developed yellow flecklike depositions at the level of the retinal pigment epithelium (RPE), which were also identified in fundus photographs of family members. The patient died at the age of 93 years, and the histological features of the macular lesion and peripheral flecks were examined. RESULTS: Histopathologically, the retinal outer nuclear layer was attenuated, particularly in the macula. This attenuation was frequently associated with normal RPE. A large area of photoreceptor degeneration was present in the central macula, with loss of the underlying RPE cells. Outside of this region, the RPE density was within normal limits. The peripheral flecks were clusters of basal laminar deposits and drusen. Bestrophin immunohistochemistry revealed labeling along both the basolateral and apical membranes of the RPE. CONCLUSIONS: Findings characteristic of Best disease may not manifest in a molecularly affected individual until late in life. Mutations in bestrophin appear to lead to extracellular deposit formation outside the macula in some families. The distribution of bestrophin in the RPE suggests that the protein may be mistargeted in those with Best disease who have the Y227N mutation, and that this may be a cause of the associated RPE and photoreceptor dysfunction.

Systemic non-Hodgkin's lymphoma simulating primary intraocular lymphoma.

PURPOSE: To describe a case of an unusual presentation of systemic non-Hodgkin's lymphoma with clinical and histopathologic findings closely resembling that of primary intraocular lymphoma. DESIGN: Observational case report. METHODS: A 58-year-old woman with a history of treated systemic non-Hodgkin's lymphoma presented 2 years later with a subretinal lesion and intraocular inflammation in her left eye. RESULTS: Diagnostic enucleation and histopathologic studies revealed findings consistent with primary intraocular lymphoma including intraretinal, subretinal, and subretinal pigment epithelial tumor cells without involvement of the choroid. CONCLUSIONS: We conclude that systemic non-Hodgkin's lymphoma can present remotely with findings resembling primary intraocular lymphoma and should be included in the differential diagnosis of intraocular inflammation and subretinal infiltrates.

Electroretinographic findings in patients with Stargardt disease and fundus flavimaculatus.

PURPOSE: To characterize the clinical and electroretinogram (ERG) features of our cohort of patients with Stargardt disease (STGD) exhibiting coding sequence variations in the ABCA4 gene. METHODS: Review of 76 patients with the clinical diagnosis of Stargardt disease/fundus flavimaculatus (STGD/FF) from the University of Iowa Department of Ophthalmology and Visual Sciences (41 patients) and the Casey Eye Institute (35 patients). Clinical examination, Goldmann perimetry, and electroretinography were performed on all 76 patients. Patients were divided into three groups on the basis of their funduscopic and electroretinographic features: (1) a normal ERG by the standards of the laboratory; (2) minimal rod or cone abnormalities; (3) severe ERG dysfunction. The latter category was further subdivided on the basis of a cone-dominated loss of function (C > R or "cone-rod dystrophy") or diffuse depression of rods and cones (C = R). Mutational analysis of the coding sequence of the ABCA4 gene was performed by single strand conformation polymorphism analysis followed by automated DNA sequencing. Each electroretinographic group was analyzed for the presence of disease causing changes using exact tests of binomial proportions corrected for multiple comparisons by Bonferroni method. Quantitative polymerase chain reaction (QPCR) was performed on patients who were homozygous for disease causing changes in the ABCA4 gene to rule out the possibility of deletions. RESULTS: Overall, 56 of 76 patients (and 77 of 152 alleles) exhibited coding sequence variations that were compatible with high-penetrance disease-causing mutations. The most common of these were His423Arg (9), frameshift mutations (7), Ala1038Val (7), and Pro1380Leu (6). Although no patients with His423Arg presented with normal ERGs, no significant correlation was observed between specific sequence variations and the electroretinographic characteristics or fundus appearance. However, a significantly greater fraction of patients with normal ERG studies failed to exhibit detectable disease-causing coding sequence variations in the ABCA4 gene identified on either allele (P = 0.0006). CONCLUSION: STGD/FF patients in our cohort exhibit a wide range of electroretinographic abnormalities, some of which are more prevalent than previously suspected. No direct correlation between clinical appearance, electrophysiologic characteristics and specific ABCA4 alleles could be identified, although a significantly lower number of our cohort with a normal ERG exhibited detectable coding sequence variations in the ABCA4 gene. However, four patients with ERG dysfunction were homozygous for a His423Arg change proven by QPCR not to be an artifact of a deletion. The presence of electrophysiologic dysfunction is not uncommon in our cohort of patients with STGD. Thus, the ERG provides clinically important information of retinal function for STGD/FF and, as such, is still indicated as part of the evaluation of these patients.

Clinical phenotype as a prognostic factor in Stargardt disease.

PURPOSE: To determine the prognostic significance of widespread flecks, described as fundus flavimaculatus, in patients with Stargardt disease. DESIGN: Historical cohort study. SUBJECTS AND METHODS: Patients with Stargardt disease were identified by searching preexisting databases at the University of Iowa and Oregon Health Sciences University. The medical records of these patients were evaluated for the validity of the diagnosis, initial and final visual acuity, length of follow-up, and initial and final phenotype. Phenotype was graded as I, II, or III on the basis of the distribution of flecks and the extent of atrophy. The relationship between final visual acuity and final clinical appearance (phenotype I versus II versus III) was evaluated controlling for length of follow-up and age using a Cochran-Mantel-Haenszel test. The clinical progression of visual acuity loss for each phenotype was analyzed using life tables and Kaplan-Meier survival analysis for age and length of follow-up effects. RESULTS: A total of 214 patients were confirmed to have Stargardt disease; 131 patients were seen at multiple visits. Eighty-two patients were identified with phenotype I, 62 with phenotype II, and 70 with phenotype III. The final visual outcome was significantly better for patients whose ophthalmoscopic abnormalities were limited to the macula (phenotype I): 80 of 82 patients with phenotype I (97.6%) maintained 20/200 or better visual acuity in at least one eye as compared to 40/62 (64.5%) patients with phenotype II and 13/70 (18.6%) patients with phenotype III (P < 0.0001). Survival analysis showed a similarly significant difference in the survival probabilities (likelihood of maintaining 20/200 or better vision) for the three phenotypes considered over age and over follow-up length (P < 0.0001), with phenotypes II and III demonstrating significantly more deterioration in vision over time. CONCLUSIONS: The presence of midperipheral flecks, especially early in life, carries a poorer visual prognosis for patients with Stargardt disease than when disease is limited to the macula. The development of midperipheral flecks is an indicator of more extensive fundus involvement and thus poorer long-term visual prognosis.

Comparison of the clinical expression of retinitis pigmentosa associated with rhodopsin mutations at codon 347 and codon 23.

PURPOSE: To examine the difference in expression of retinitis pigmentosa from mutations at codon 23 and codon 347 or rhodopsin; to report a novel mutation in rhodopsin. METHODS: Goldmann perimetry (solid angle of I4e isopter) and electroretinographic amplitudes (square root transform of a response ratio) were analyzed for 24 patients with mutations at codon 347 (15 with Pro347Ala, 2 with Pro347Gln, 6 with Pro347Leu, and 1 with a novel Pro347Cys change) and 41 patients with mutations at codon 23 (6 with Pro23Ala; 35 with Pro23His). RESULTS: When all patients with mutations at codons 347 and 23 were compared, loss of visual fields was significantly worse in patients with codon 347 changes (P =.0003). Only rod responses of the electroretinograms were significantly different between the two groups (P =.048). Specific comparison of Pro347Ala with Pro23Ala using regression analysis demonstrated significant differences in severity between codon 23 and codon 347 patients for b-wave amplitudes of rod (P =.0069), cone (P =.039) and maximum combined response (P =.049). The solid angle of the I4e isopter was also significantly different (P =.025) between the groups after controlling for age. Modeling age by group for Pro347Ala comparison produced an R(2) of.44. CONCLUSION: We reconfirmed that rhodopsin-related retinitis pigmentosa from mutations involving codon 347 produces a more severe phenotype than that involving codon 23. Accurate modeling of disease was shown to be possible by incorporating the effects of a patient's age and specific genotype. Therefore, both of these variables must be considered in prognostic counseling and subject recruitment for future therapeutic trials.

Long-term follow-up of patients with serpinginous choroiditis.

PURPOSE: To evaluate the long-term clinical course of serpiginous choroiditis-a recurrent inflammatory disease that causes progressive visual loss-and to determine the efficacy of immunosuppressive therapy. METHODS: A retrospective study of patients who met inclusion criteria for serpiginous choroiditis at The University of Iowa Hospitals and Clinics was performed. Information collected included duration of follow-up, number of recurrences of inflammation, visual acuities, and development of choroidal neovascularization. The number of recurrences of inflammation in patients treated with immunosuppressive agents was compared with that in patients treated only with corticosteroids or observation. RESULTS: Seventeen patients were identified who had a mean age at presentation of 39.6 years. The mean duration of follow-up was 149.2 months, with 13 patients who were followed up for >60 months. Twelve eyes either presented with or developed macular choroidal neovascularization during follow-up. Thirteen of the patients were followed up for >12 months. Of these 13 patients, 6 received treatment with immunosuppressive agents along with corticosteroids. Four of the six patients developed no further inflammatory recurrence. Seven of the 13 patients were treated with corticosteroids or observation. All of these patients developed recurrences (P = 0.021, Fisher exact test). CONCLUSIONS: Immunosuppressive agents appear to reduce the rate of recurrent disease in serpiginous choroiditis compared with corticosteroids.

Birdshot retinochoroiditis: long term follow-up of a chronically progressive disease.

PURPOSE: To determine the long-term course of birdshot retinochoroiditis by reviewing patient records from The University of Iowa Hospitals and Clinics. DESIGN: A descriptive case series. METHODS: We conducted a retrospective review of 19 patients seen at The University of Iowa for birdshot retinochoroiditis. Inclusion criteria were set before review. Goldmann perimetry isopters were converted to an area measurement in steradians and the I2e and I4e isopters were evaluated at each time point. The visual acuity, electroretinography (ERG), and visual field findings were compared to the clinical appearance of the fundus. RESULTS: Follow-up ranged from one visit to 220 months. Of the 14 patients who were tested, all were HLA-A29-positive. Seven patients were followed for >or=60 months. Eleven patients were followed for >or=30 months. The initial visual acuity was 20/50 or better in 36 of 38 eyes and 20/60 and 20/80 in the remaining two. Visual acuity was worse than 20/50 in three of 22 eyes followed for more than 30 months. Visual field data demonstrated progressive loss of area for either the I4e or I2e isopters in six of seven patients who were followed for >or=60 months. Multiple ERGs were performed over time on eight of 19 patients; seven of eight patients demonstrated progressive loss of electrophysiologic indices. CONCLUSION: Retinal function in birdshot retinochoroiditis deteriorated progressively over a period of years despite stable visual acuity. Late in the course of disease, visual acuity may be lost due to chorioretinal atrophy in the posterior pole. Visual acuity alone is not an adequate parameter with which to monitor disease activity and may falsely suggest that a patient is stable or doing well. Intermittent treatment of the inflammatory exacerbations did not prevent progressive visual loss. Other treatment strategies such as prolonged corticosteroid or immunosuppressive treatment should be investigated for patients with birdshot retinochoroiditis.