EyeArt artificial intelligence analysis of diabetic retinopathy in retinal screening events.

PURPOSE: Early detection and treatment of diabetic retinopathy (DR) are critical for decreasing the risk of vision loss and preventing blindness. Community vision screenings may play an important role, especially in communities at higher risk for diabetes. To address the need for increased DR detection and referrals, we evaluated the use of artificial intelligence (AI) for screening DR. METHODS: Patient images of 124 eyes were obtained using a 45° Canon Non-Mydriatic CR-2 Plus AF retinal camera in the Department of Endocrinology Clinic (Newark, NJ) and in a community screening event (Newark, NJ). Images were initially classified by an onsite grader and uploaded for analysis by EyeArt, a cloud-based AI software developed by Eyenuk (California, USA). The images were also graded by an off-site retina specialist. Using Fleiss kappa analysis, a correlation was investigated between the three grading systems, the AI, onsite grader, and a US board-certified retina specialist, for a diagnosis of DR and referral pattern. RESULTS: The EyeArt results, onsite grader, and the retina specialist had a 79% overall agreement on the diagnosis of DR: 86 eyes with full agreement, 37 eyes with agreement between two graders, 1 eye with full disagreement. The kappa value for concordance on a diagnosis was 0.69 (95% CI 0.61-0.77), indicating substantial agreement. Referral patterns by EyeArt, the onsite grader, and the ophthalmologist had an 85% overall agreement: 96 eyes with full agreement, 28 eyes with disagreement. The kappa value for concordance on "whether to refer" was 0.70 (95% CI 0.60-0.80), indicating substantial agreement. Using the board-certified retina specialist as the gold standard, EyeArt had an 81% accuracy (101/124 eyes) for diagnosis and 83% accuracy (103/124 eyes) in referrals. For referrals, the sensitivity of EyeArt was 74%, specificity was 87%, positive predictive value was 72%, and negative predictive value was 88%. CONCLUSIONS: This retrospective cross-sectional analysis offers insights into use of AI in diabetic screenings and the significant role it will play in automated detection of DR. The EyeArt readings were beneficial with some limitations in a community screening environment. These limitations included a decreased accuracy in the presence of cataracts and the functional cost of EyeArt uploads in a community setting.

Educational quality and content of YouTube videos on diabetic macular edema.

PURPOSE: Diabetic macular edema (DME) is a vision-threatening complication of diabetes mellitus due to increased vascular permeability. Patients are increasingly using YouTube videos to educate themselves about DME. This study analyzes the content and quality of YouTube videos about DME. METHODS: Videos were searched in December 2021 for "diabetic macular edema." The first 100 videos sorted by both relevance and view count were reviewed (n = 200). Quantitative metrics and content were collected. Two reviewers assessed videos using the JAMA (0-4), modified DISCERN (1-5), and Global Quality Scale (GQS, 1-5). Videos were sorted into author groups: 1 (academic institutions/organizations), 2 (private practices/organizations), and 3 (independent users; ophthalmologist users noted). Statistical analyses were deemed significant at a = 0.05. RESULTS: One hundred four videos were included after applying exclusion criteria. Overall mean + standard deviations were 2.25 ± 0.83 (JAMA), 3.47 ± 0.55 (DISCERN), and 3.95 ± 0.95 (GQS). 51.9% of videos stated a definition, 32.7% mentioned screening, and 50% mentioned any DME risk factor. Healthcare professional-targeted videos had higher JAMA and DISCERN scores than patient-targeted videos (p < 0.05). Videos using ophthalmologists had higher JAMA and DISCERN scores than those lacking their presence (p < 0.05). JAMA scores significantly varied between author groups; within group 3, ophthalmologist-authored videos had higher DISCERN scores (p < 0.05). CONCLUSION: Videos without ophthalmologists or targeted toward patients had poor quality and content coverage. The rising prevalence of diabetes, coupled with increased internet use for acquiring medical information, creates a strong need for high-quality information about DME.

Risk factors for ischemic cerebral stroke in patients with acute amaurosis fugax.

OBJECTIVE: The National Stroke Association and the American Heart Association consider retinal ischemia, as in the case of amaurosis fugax (AF), to be a stroke event. The purpose of this study was to evaluate the risk factors for ischemic cerebral stroke in patients hospitalized for acute AF. METHODS: The National Inpatient Sample Database from 2002 to 2014 was used to identify patients 21 years of age and older with a primary admission diagnosis of AF with the ICD-9 code 362.34. Comorbidity measures and in-hospital events were extracted using relevant ICD-9 codes. Statistical analyses were performed using IBM SPSS 25 and R package. RESULTS: A weighted total of 12,142 patients was identified. The most common comorbidities in this cohort with AF included hypertension, dyslipidemia, tobacco use, coronary artery disease (CAD), and diabetes mellitus. Multivariable regression analysis showed comorbidities of hypercoagulable state, systemic vasculitis, CAD, and atherosclerosis to be independent risk factors for ischemic stroke in patients with AF. In contrast, dyslipidemia was associated with a decreased risk. Asian/Pacific Islander race conferred a 5-fold increased risk compared with Whites. CONCLUSION: Ischemic stroke and myocardial infarction were diagnosed in 0.3%-0.9% of hospitalized acute AF cases. Presence of hypercoagulable state, systemic vasculitis, CAD, and atherosclerosis each individually increased the risk of ischemic stroke by more than 3-fold; patients with these risk factors and acute AF should be closely monitored for developing acute systemic thrombotic events.

Endogenous endophthalmitis in patients with intravenous opioid use: demographics and associated comorbidities.

PURPOSE: To identify risk factors for endogenous endophthalmitis (EE) in hospitalized adults, under 65 years of age, with a history of intravenous opioid use and non-ocular infection. METHODS: The National Inpatient Sample Database was used to identify cases of EE with a recent history of intravenous opioid use disorder with associated non-ocular infection. Systemic and ocular comorbidities were identified using codes from the International Classification of Diseases, Ninth Revision (ICD-9). Descriptive and regression analyses were performed to evaluate the risk factors for EE using IBM SPSS 23. RESULTS: Of the 605,859 inpatients, 21-65 years age, who had a history of recent opioid-IVDU and an associated IVDU-associated systemic infection, 363 (0.1%) had EE. Systemic comorbidities such as diabetes mellitus, mitral valve disease, aortic valve disease, history of cardiac valve transplantation, chronic kidney disease/renal failure, cirrhosis, active or previous radiation therapy, and history of solid organ transplantation were significantly more prevalent in patients with EE. A significantly increased risk of EE in intravenous opioid users was noted if they were of male gender (OR = 1.84), Asian/Pacific Islander ethnicity (OR = 4.41), had history of cirrhosis (OR = 2.33), active or history of radiation therapy (OR = 14.74), history of solid organ transplantation (OR = 5.91), candidemia (OR = 15.22), and infectious endocarditis (OR = 4.83). Conversely, concurrent alcohol use disorder (OR = 0.35) decreased the risk of EE. CONCLUSION: Various demographic variables and systemic comorbidities increased the risk of developing EE in inpatients with a history of intravenous opioid use with associated non-ocular infection.

Risk Factors for Endogenous Endophthalmitis in Hospitalized Patients with Candida Fungemia.

PURPOSE: To use the 2002 through 2014 National Inpatient Sample (NIS) database to identify risk factors for endogenous endophthalmitis (EE) in hospitalized patients with candidemia. DESIGN: Retrospective, cross-sectional study. PARTICIPANTS: Hospitalized patients sampled in the NIS database. METHODS: The NIS database (2002-2014) was used to identify patients with candidemia and EE and their comorbidities. Descriptive analysis was performed with chi-square testing, and risk factors for EE were identified using logistic regression analysis. Chi-square testing and regression analysis were performed using IBM SPSS software version 23 (IBM Corp, Armonk, NY) and R package software version 3.4.3 (R Foundation for Statistical Computing, Vienna, Austria), respectively. MAIN OUTCOME MEASURE: Diagnosis of EE in hospitalized patients with candidemia. RESULTS: We identified 98 783 hospitalized patients with candidemia; 529 patients (0.5%) had concurrent EE. Men constituted 48.0% of patients who did not demonstrate EE and 45.1% of those who did (P = 0.186). The average age of fungemia patients with EE was 54.6 years and of those without EE was 58.2 years (P < 0.001). Most EE cases (58.6%) occurred in patients 21 to 64 years of age. Hispanic (odds ratio [OR], 1.58), Asian or Pacific Islander (OR, 3.51), and Native American (OR, 5.22) patients with candidemia were at an increased risk of EE developing compared with White patients. Candida endocarditis (OR, 1.84), cirrhosis (OR, 1.93), diabetes with chronic complications (OR, 1.96), intravenous drug use (OR, 3.12), radiation therapy (OR, 5.28), and solid organ transplantation (OR, 2.48) increased the risk of seeding the infection into the eye. Conversely, chronic kidney disease (OR, 0.53) and invasive mechanical intubation (OR, 0.43) were associated with a decreased risk of EE. The mortality of inpatients with candidemia was significantly lower in the EE group (2.8% vs. 15.6%; P < 0.001). CONCLUSIONS: Systemic comorbidities that increased the risk of EE in candidemia included endocarditis, cirrhosis, diabetes with chronic complications, intravenous drug use, radiation therapy, and solid organ transplantation. Racial disparity was observed with Hispanics, Asians and Pacific Islanders, and Native Americans at a higher risk than Whites of being diagnosed with EE in the setting of Candida fungemia.

Risk factors for central retinal vein occlusion in young adults.

PURPOSE: Several risk factors have been identified for central retinal vein occlusion (CRVO) in older population. CRVO in young is uncommon, and the risk factors for this group are unclear. This large retrospective, cross-sectional study used the National Inpatient Sample (NIS) database to evaluate the risk factors for CRVO in patients 18 to 40 years of age. METHODS: The 2002 to 2014 NIS database was used. All patients 18 to 40 years of age with a primary diagnosis of CRVO were identified. Age- and gender-matched non-CRVO controls were randomly selected. The primary outcome was identification of risk factors for CRVO. Chi-square analysis and Firth logistic regression were performed with IBM SPSS 23 and R packages versions 3.4.3, respectively. p < 0.05 was considered significant. RESULTS: A total of 95 weighted young CRVO patients were identified. The average age was 31.44 ± 6.41 years with no gender predilection. Systemic and ocular conditions found to have statistically significant associations with CRVO included primary open-angle glaucoma (POAG) (OR 836.72, p < 0.001), retinal vasculitis (OR 705.82, p < 0.001), pseudotumor cerebri (OR 35.94, p < 0.001), hypercoagulable state (OR 25.25, p < 0.001), history of deep vein thrombosis/pulmonary embolism (DVT/PE) (OR 21.88, p < 0.001), and hyperlipidemia (OR 3.60, p = 0.003). CONCLUSION: The most significant risk factors for CRVO in young adults were POAG, retinal vasculitis, and pseudotumor cerebri. Hypercoagulable states and DVT/PE were also associated with CRVO in this population. Systemic inflammatory conditions were not associated with CRVO. Traditional risk factors such as hypertension and diabetes did not pose significant risks, whereas hyperlipidemia was deemed a significant risk factor.

Predictive factors of enucleation after open globe injuries.

BACKGROUND: Trauma is the leading cause of enucleations in the USA. Current information regarding open globe injuries (OGI) is based mainly on data from individual tertiary care centers across the country which might skew the findings towards the population served by these level-one trauma centers. The aim of this study is to evaluate the demographics, characteristics, and risk factors of traumatic enucleations in a large data sample. METHODS: Descriptive cross-sectional observational study using the National Inpatient Sample (NIS) Database from 2002 to 2013. Inpatients with traumatic enucleations were identified using ICD-9 codes. Chi-square and logistic regression analyses were used to identify differences between the enucleated and non-enucleated cohorts and to evaluate the predictive factors of enucleation in OGIs. RESULTS: Enucleations were performed in 3020 (6.2%) of 48,563 OGIs identified. The average age in the enucleated cohort for males vs. females was 44.7 vs. 62.2 years. In the USA, the highest number of traumatic enucleations occurred in the 21-40 group (41.8%) and the fewest in the 80+ age group (11.8%). The risk of enucleation decreased across the age groups significantly. Compared with the 21-40 age group, the risk of undergoing enucleation was 15% lower in patients 41 to 60 years of age, 35% in patients 61 to 80, and 40% lower in patients over 80. In total, 5.1% OGIs in women and 6.7% of OGIs in men were enucleated. The risk of enucleation was 29% higher in men than in women. The highest absolute number of enucleations was seen in Whites. Compared with Whites, Blacks had a 63% higher risk of enucleation following an OGI. OGIs with rupture-type injury, endophthalmitis, or phthisis were significantly higher odds to be enucleated. CONCLUSIONS: The risk of enucleation following traumatic OGI significantly increased for patients who were in the 21-40 age group, of Black race, or of male gender; the risk also increased if the injury was a rupture-type or associated with endophthalmitis or phthisis. The risk of depression was 75% higher in enucleated patients versus non-enucleated patients.

Update on the Use of Anti-VEGF Drugs in the Treatment of Retinopathy of Prematurity.

Retinopathy of prematurity (ROP) is one of the many significant consequences of premature birth and remains one of the leading causes of visual impairment in infants. Originally, cryotherapy was used to prevent the complications of vitreous hemorrhage and retinal detachment. Subsequently, laser photocoagulation, which is at least as effective and possibly safer than cryoretinopexy, was adopted as the primary treatment for type 1 ROP (stage 2 or 3 disease in zone II with plus disease or any stage disease in zone I with plus disease or stage 3 disease in zone I without plus disease). Laser therapy has been proven effective, and has a degree of permanence that is yet to be matched by alternative treatments, but can be associated with significant ocular side effects such as myopia. Treatment of type 1 ROP with anti-vascular endothelial growth factor (VEGF) agents seems to have fewer ocular side effects than laser ablation of the retina, particularly if used to treat type 1 ROP in zone I. However, ROP recurrence is a real threat after anti-VEGF therapy and long-term systemic side effects of this therapy remain under evaluation. This review focuses on the ophthalmic and systemic benefits and risks of anti-VEGF therapies for ROP as compared to retinal photocoagulation. Anti-VEGF therapies have dramatically altered the management of ROP and have also been shown to be beneficial with regard to the visual prognosis of patients with ROP, but patients so treated require frequent short- and long-term follow-up to detect and manage potential complications associated with this form of treatment. Such information also will allow clinicians to characterize the efficacy, side effect profile, and utility of intravitreal anti-VEGF agents for this condition. Prospective studies are needed to identify the optimum anti-VEGF drug and dose. [J Pediatr Ophthalmol Strabismus. 2020;57(6):351-362.].

Assessing the risk of stroke development following retinal artery occlusion.

BACKGROUND: Retinal artery occlusion (RAO) can cause acute, painless, and irreversible loss of vision. Using a large sample population database, we investigated the possible impact of RAO and numerous other clinical and non-clinical factors on the risk of developing a subsequent cerebrovascular accident. METHODS: Cases of RAO were obtained from the National Inpatient Sample (NIS) database between 2002 and 2013 using ICD-9 codes. Associated morbidities and procedures were assessed in these cases with a primary hospital admitting diagnosis of stroke. Univariate and multivariate logistic regression analyses were carried out in RAO cases to determine risk factors for stroke. The Bonferroni correction method was applied. RESULTS: The RAO group consisted of 19,809 cases that were separated into stroke (n=1,157, 55% male, mean age: 69±0.4 years) and non-stroke (n=18,652, 55% male, mean age: 68±0.1) cohorts. Age groups associated with reduced risk of stroke with respect to were 20-39 (OR: 0.391), 40-59 (OR: 0.842), and 60-79 (OR: 0.837). No cases of stroke were present for ages < 20. Other factors associated with a reduced stroke risk were carotid stenosis (OR: 0.187), transient ischemic attack (OR: 0.064), coronary artery disease (OR: 0.788), cardiac catheterization (OR: 0.481), and septicemia (OR: 0.333). Factors associated with an increased risk of stroke included hypertension (OR: 1.418), tobacco use (OR: 1.568), valvular disease (OR: 1.359), hyperlipidemia (OR: 1.298), and non-stroke cerebrovascular disease (OR: 2.985). CONCLUSIONS: A large patient population was used to determine that RAO patients with a history of hypertension, hyperlipidemia, tobacco usage, valvular disease, or non-stroke cerebrovascular disease had an increased risk of stroke. Patients below the age of 40 had significantly reduced the odds of stroke. Carotid stenosis, coronary artery disease, transient ischemic attacks, cardiac catheterization, and septicemia were all independently associated with a decreased risk of stroke development in RAO patients.

Silicone oil removal: post-operative complications.

PURPOSE: To describe the characteristics, outcomes and complications of eyes after silicone oil removal. METHODS: Retrospective case series of eyes that underwent oil removal between 2012 and 2016 at The Institute of Ophthalmology and Visual Science. Visual acuity (VA), intraocular pressure (IOP) and rates of retinal re-detachment, hypotony, ocular hypertension, corneal decompensation, cystoid macular edema (CME) and cataract progression were evaluated. RESULTS: Totally, 101 eyes of 99 patients (65% male, average age 47.2 years) were identified. Oil tamponade had been used for retinal detachment (RD) repair in all eyes; 15 eyes had also undergone an open globe repair previously. The most common vitreous substitutes used after oil removal were balanced salt solution (BSS) and air in 90% of eyes. The average time of oil tamponade before removal was 9.46 months. The average logMAR VA before oil removal was 1.7 which improved to an average of 1.4 post-operatively. The average IOP pre-operatively was 16.1 mm Hg, which decreased to an average of 14.8 mm Hg post-operatively. Complications after oil removal, included retinal re-detachment (6.9%), hypotony (7.9%), ocular hypertension (12.9%), corneal decompensation (9.9%), CME (2%) and cataract progression (68%). CONCLUSION: This study showed an overall improvement in VA and decrease in IOP after oil removal. Cataract progression was the most common complication.

Practical Lessons from Protocol I for the Management of Diabetic Macular Edema.

Protocol I, a multicenter randomized clinical trial, compared the visual outcomes of patients treated with 0.5 mg intravitreal ranibizumab with either prompt or deferred (by 24 weeks laser), 4 mg intravitreal triamcinolone with prompt laser, or sham injection with prompt laser for the treatment of center-involving diabetic macular edema (DME). A total of 854 adult patients with type I or II diabetes and any level of non-proliferative diabetic retinopathy or proliferative retinopathy with adequate panretinal photocoagulation, with best-corrected visual acuity (BCVA) of 78 to 24 ETDRS letters (Snellen equivalent of 20/32 to 20/320) and visual loss attributed to macular edema, or retinal thickening with central subfield thickness of at least 250 µm by OCT were enrolled. The main outcomes relevant for practicing clinicians are as follows. (1) Intravitreal ranibizumab treatment provides superior visual outcomes compared to conventional laser treatment. (2) Adjunctive laser treatment does not appear to provide substantial visual benefit compared to ranibizumab treatment alone, but may reduce the number of injections required to resolve DME. Deferral of laser is likely beneficial in patients with worse initial visual acuity. (3) Intravitreal triamcinolone provides similar visual outcomes compared to intravitreal ranibizumab in pseudophakic patients but is associated with a clinically important increased risk of increased intraocular pressure (IOP), need for glaucoma medications, and need for glaucoma surgery. (4) Delayed initiation of intravitreal ranibizumab therapy provides improved visual outcome among patients initially treated with conventional laser photocoagulation or triamcinolone, but the magnitude of the benefit is not as great as is observed when ranibizumab treatment is initiated promptly. (5) The number of ranibizumab injections required to achieve the desired visual outcome decreases substantially after the first year, with the majority of patients not requiring further treatment after 3 years. (6) Patients who do not have a rapid response to ranibizumab still display long-term benefit to continued therapy, although perhaps less than those with immediate improvement. (7) Intravitreal ranibizumab is not only effective in reducing retinal edema and improving BCVA among patients with DME, it is also a disease modifying therapy and induces improvement of the diabetic retinopathy severity score by 2 or more steps in approximately one third of patients. Triamcinolone injection also induces improvement in diabetic retinopathy severity in DME patients, but perhaps to a lesser degree. (8) No increased risk of systemic adverse events was observed among patients treated with intravitreal ranibizumab compared to sham-injected controls or triamcinolone-treated patients, but the low frequency of adverse events, restrictive enrollment criteria, and specific posology employed in this study limit the generalization of this conclusion to patients routinely encountered in clinical practice. (9) There was no clinically important increased risk of major ocular complications among patients treated with intravitreal ranibizumab (including the risk of glaucoma), although endophthalmitis is a potentially devastating outcome should it occur. In addition to the risk of endophthalmitis, intravitreal triamcinolone injection was associated with clinically important increased risk of cataract progression and increased IOP.

Practical Lessons from Protocol T for the Management of Diabetic Macular Edema.

PURPOSE: To review the results of Diabetic Retinopathy Clinical Research Network Protocol T, as applied to clinical practice. METHODS: Review of major publications reporting the results of Protocol T, a randomized single-masked (in year-1 only), multicenter clinical trial comparing aflibercept, bevacizumab, and ranibizumab as treatment option for center-involving diabetic macular edema (DME). The main outcome measures were change in visual acuity (VA), central subfield thickness (CST) on optical coherence tomography, cost effectiveness, burden of care, and safety. RESULTS: A total of 660 participants (mean age 61 ± 10 years, 47% women, 65% Caucasian) were randomized to treatment with aflibercept (n = 224), ranibizumab (n = 218), or bevacizumab (n = 218). The majority of patients (90%) had type II diabetes, with an average duration of 17 ± 11 years. About half the patients had baseline ETDRS VA of 20/32 to 20/40, and half had ETDRS VA of 20/50 to 20/320 in all 3 cohorts. Patients in all 3 cohorts received a similar number of injections during the study period (9-10 in year-1; 5-6 in year-2). The year-1 improvement in ETDRS letters was significantly higher for aflibercept than for ranibizumab and bevacizumab in patients with baseline VA 20/50 or worse (p = 0.003 and p < 0.001, respectively), but was no different in patients with better baseline VA of 20/32 to 20/40 (p = 0.69). By year-2, among patients with poorer baseline VA, there was a difference in mean letters gained between aflibercept and bevacizumab (p = 0.02), but no difference between aflibercept and ranibizumab (p = 0.18). At year-2, there was no clinically meaningful difference in VA improvement (i.e., gain or loss of ≥10 or ≥15 letters) among any of the agents (p > 0.74). Bevacizumab was less effective than the other agents in decreasing CST at years-1 and -2 in the overall cohort of patients (p < 0.001). However, bevacizumab is substantially cheaper and much more cost-effective (when comparing expense and quality of life measures) than aflibercept and ranibizumab. The cost of other agents would have to decrease by 80-90% to be cost-effective relative to bevacizumab. Intravitreal administration of anti-VEGF therapy has relatively few ocular and systemic side effects, but caution may be warranted for patients with a recent history or high risk of myocardial infarction or stroke. CONCLUSIONS: Aflibercept, bevacizumab, and ranibizumab are highly effective treatments for DME. Bevacizumab is more cost-effective than aflibercept and ranibizumab. Intravitreal administration of drugs is relatively safe; however, intravitreal administration may be associated with severe systemic side effects in a small percentage of patients, particularly in those with a prior history of or high risk of Anti-Platelet Trialists' Collaboration events.

Anti-Vascular Endothelial Growth Factor Injections: The New Standard of Care in Proliferative Diabetic Retinopathy?

For decades, panretinal photocoagulation (PRP) has been the standard of care for the treatment of proliferative diabetic retinopathy (PDR). The relatively recent advent of anti-vascular endothelial growth factor (VEGF) formulations for intravitreal injection has provided a fresh perspective on PDR treatment, especially in eyes with concurrent diabetic macular edema (DME). The anti-VEGF agent ranibizumab has demonstrated a potentially protective effect on eyes with DME in terms of progression to PDR in the RIDE/RISE trials, as has aflibercept in the VIVID/VISTA trials. In 2015, these 2 agents were approved by the Food and Drug Administration for the treatment of PDR with DME, though PRP still remains the standard of care for eyes without baseline DME. Published results from Protocol S illustrate the non-inferiority of ranibizumab versus PRP in the treatment of PDR, the first prospective study to do so in eyes with and without baseline DME. These results also reveal that treatment with ranibizumab, when compared to standard treatment with PRP, may also lead to less peripheral visual field loss, reduced need for vitrectomy, and reduced chance for developing DME. Both PRP and intravitreal ranibizumab have very low rates of adverse events. However, treatment with anti-VEGF agents generally is associated with higher costs, increased need for follow-up, and the risk of potentially catastrophic ocular complications (e.g., endophthalmitis) and systemic side effects. Anti-VEGF agents should be considered in cases of media opacity preventing completion of PRP in compliant patients without recent cerebrovascular accident or myocardial infarction, though the long-term efficacy of these agents remains to be studied, especially after the discontinuation of injections.

Application of Clinical Trial Results to Clinical Practice.

Two critical questions one must answer as one applies the results of a clinical trial to clinical practice are: (1) Regardless of whether the trial result is likely to be replicated or reproduced in a second large-scale trial, are the results likely to be reproduced in one's practice? (2) Regardless of whether the experimental treatment was better than the alternative on average for a population of patients, are the results clinically important for a given patient in one's practice? To determine if a study result is likely to be reproduced in one's clinical practice, it may be helpful to answer 5 questions: (1) Have steps been taken to minimize bias? (2) Is the result likely due to the treatment? (3) Is the result unlikely due to chance? (4) Is the study population representative of one's patients? (5) Is the totality of evidence consistent? If the answer to all 5 questions is "yes," then we posit that the trial result is likely to be reproduced in one's practice. If not, the likelihood of reproducibility is low. If the answer is yes to all questions except the last, then reproducibility in one's practice is not clear and depends on the strength of the prior versus the current evidence. If the prior evidence is strong, such as multiple pivotal randomized clinical trials, and if the current trial result is not consistent with the previous studies, then the current result may not be reproduced in one's practice. To determine if a study result is clinically important, a 3-step approach is suggested. Step 1. Decide, a priori, what a clinically meaningful difference between 2 treatments would be. This choice defines regions of beneficial, harmful, and trivial outcomes. Step 2. Identify the confidence intervals (CIs). Determine whether the 95% CI mostly includes the range of clinically beneficial outcomes and lies outside the range of clinically harmful outcomes. If these conditions are met, the result is probably clinically important, but the result may or may not be statistically significant. Put the CIs and the regions of benefit/harm together to make a decision about clinically important effects. Step 3. Assess the proportion of eyes with clinically meaningful changes in vision. The proportion of "responders" among patients receiving a given treatment reflects the likelihood of one's patient having a clinically meaningful response to the treatment. In summary, not all statistically significant results are reproduced, even those of carefully designed clinical trials. Determining if a study result is likely to be reproduced in one's practice is even more problematic. The 5-question test may help in this regard. The 5-question test attempts to assess whether steps have been taken to: minimize bias; avoid confounding; ensure adequate statistical power to support precision in the estimates of population parameters; insure external validity of the trial result; and determine whether there is a convergence of evidence consistent with the trial's major findings. To determine if a statistically significant result is likely to be clinically important, a 3-step approach may be useful, focusing on CIs and the proportion of eyes with clinically meaningful changes in vision. Application of clinical trial results to clinical practice requires critical analysis of the extant literature and good clinical judgment.

INFECTIOUS KERATITIS-ASSOCIATED ENDOPHTHALMITIS: A 14-Year Study.

PURPOSE: To describe the demographics, characteristics, management, and outcomes of eyes with endophthalmitis related to infectious keratitis. METHODS: Retrospective chart review of all patients treated for infectious keratitis-associated infectious endophthalmitis between 2001 and 2014 at University Hospital, Rutgers New Jersey Medical School. RESULTS: Thirty-eight cases with infectious keratitis-associated infectious endophthalmitis were identified (21 men [55%], mean age: 66.2 ± 20.7 years), with average time from the beginning of ulcer symptoms to endophthalmitis of 11.0 days. Associated systemic conditions (diabetes, HIV, immunosuppressive therapy, cirrhosis, or dementia) were present in 57.9%; 60.5% had previous intraocular surgery. Etiology showed gram-positive bacteria in 14 cases (36.9%), gram-negative bacteria in 7 (18.4%), fungi in 4 (10.5%), and no growth/unknown in 12 (31.6%). Nineteen cases (50%) presented with no light perception and were primarily enucleated. The remaining 19 eyes each received intravitreal antibiotics (mean: 1.5 injections); 8 (42.0%) underwent pars plans vitrectomy with vitreous biopsy, whereas 5 (26.3%) received emergency corneal transplantation. Final visual acuity was no light perception in 6 eyes (3 secondarily enucleated), light perception in 2, hand motion in 7, counting fingers in 2, and ≥20/50 in 2. CONCLUSION: Our study of 38 eyes with infectious keratitis-associated infectious endophthalmitis revealed generally poor visual outcomes and a high rate of systemic conditions and previous intraocular surgery.

Posterior Segment Intraocular Foreign Bodies: A 10-Year Review.

PURPOSE: To describe the characteristics of open-globe injuries with posterior segment intraocular foreign bodies (IOFBs). DESIGN: Retrospective chart review study. PARTICIPANTS: Patients treated for posterior segment IOFB injuries. METHODS: Retrospective analysis of all patients with posterior segment IOFBs from 2003 to 2014 was conducted. Data including demographics, mechanism of injury, type of IOFB, method of diagnosis, presenting examination, medical and surgical treatment, visual outcomes, and complications were recorded. MAIN OUTCOME MEASURES: Visual acuity (VA); anatomically successful retinal reattachment; need for additional surgery; frequency of post-traumatic complications, such as sympathetic ophthalmia (SO), endophthalmitis, and enucleations; and accuracy of Ocular Trauma Score (OTS). RESULTS: Thirty-one patients (28 male; mean age, 36.6 years; 42% Hispanic) had posterior segment IOFB injuries, 23 (74%) of which were construction work related. Twenty-five IOFBs (81%) were metallic. Twenty-four IOFBs (77%) had Zone I entry. Computed tomography (CT) scan detected an IOFB in 21 of 22 eyes in which it was performed, with 1 scan highly suspicious for an IOFB. Average size of the IOFB was 10 mm3; size or initial VA did not have any correlation with final VA. The OTS had 60% accuracy in predicting final VA (n = 20). The majority of patients had traumatic cataract and vitreous hemorrhage (VH) on presentation (77% and 61%, respectively); 65% had a retinal tear or retinal detachment (RD), and these patients had worse final VA than those with no retinal pathology. Average time from injury to IOFB removal was 3 days because of the delay in presentation to our facility; 27 of 31 patients (87%) had IOFBs removed within 24 hours of presentation with pars plana vitrectomy (PPV) and either gas or silicone oil tamponade. Patients were admitted for an average of 4 days of intravenous antibiotics. The most common complication was recurrent RD in 11 patients (35%), which portended worse final VA. One patient (3%) developed SO. There were no cases of postoperative endophthalmitis or enucleation. CONCLUSIONS: Open-globe injuries with posterior segment IOFBs have a guarded visual prognosis, particularly when associated with RD. Increased awareness of the importance of eye protection can help minimize the occurrence of these injuries.