RELAY INTRASCLERAL 6-0 POLYPROPYLENE-ASSISTED INTRAOCULAR LENS FIXATION: A Retrospective Comparison With Modified Yamane Technique.

PURPOSE: To describe a new surgical technique that uses a relay suture to optimize sutureless scleral fixation of intraocular lens in eyes with capsular insufficiency and to evaluate the outcomes of this technique versus Flanged, double needle-guided, scleral fixation of intraocular lens with haptic trimming (modified Yamane approach). METHODS: Relay intrascleral 6-0 polypropylene sutures with flanges generated at the intraocular ends were used to fixate and securely hold a rigid intraocular lens with haptic eyelets against the inner scleral wall. The results were analyzed at 3-, 6-, and 12-month follow-up. RESULTS: This retrospective cohort study found that compared with the control group (n = 27), the relay-sutured group (n = 26) had greater mean changes in corneal astigmatism (0.44 vs. -0.52 diopters [D]) and fewer mean degrees of intraocular lens astigmatism (0.62 vs. 1.1 D). The mean intraocular lens decentration was comparable between both the groups; however, in the control group, there was a significant increase in intraocular lens tilt degrees from 3 to 12 months. Moreover, significantly higher proportions of patients with persistent macular edema and iris-optic capture were seen in the control group. CONCLUSION: The relay-sutured technique may be an alternative to flapless scleral fixation of intraocular lens and provides a stable intraocular lens position with acceptable complication rates.

Coexistence of choroidal melanoma and rhegmatogenous retinal detachment: a case report and literature review.

Introduction and importance: The simultaneous occurrence of rhegmatogenous retinal detachment (RRD) and choroidal melanoma is extremely rare, and diagnosis of choroidal melanoma in RRD patients is challenging. As a result, choroidal masses in RRD patients tend to be overlooked, resulting in delayed treatment. The authors report a rare case presenting with simultaneous choroidal melanoma and RRD, and the authors review the related literature. Case presentation: A 45-year-old Thai man who presented with inferior RRD and choroidal elevation in the left eye was examined using fundoscopy and ultrasonography. The presumptive diagnosis was simultaneous RRD and haemorrhagic choroidal detachment (CD). Vitrectomy and external drainage were attempted but were unsuccessful in draining fluid from the presumed haemorrhagic CD. Although the retina was reattached, the choroidal elevation remained unchanged. After being lost to follow-up, the patient returned later with severe proptosis in the left eye. MRI findings suggested a presumptive diagnosis including choroidal melanoma and choroidal metastasis, which choroidal melanoma was later confirmed through histopathological examination after exenteration. Clinical discussion: Important clinical clues were provided for distinguishing between choroidal melanoma and haemorrhagic CD in cases of coexisting RRD. Conclusions: In the differential diagnosis of RRD with suspicious choroidal elevation, the possibility of the presence of choroidal melanoma should be considered. In particular, in cases where haemorrhagic CD fails to drain during surgery, the possibility of underlying choroidal melanoma should be investigated.

SIMPLE REFIXATION OF A DISLOCATED HAPTIC OF FLANGED INTRAOCULAR LENS USING A 25-GAUGE NEEDLE.

PURPOSE: To report the development of a novel, simple-to-use technique for the salvage of a haptic of dislocated flanged intraocular lens during a retinal detachment surgery. METHODS: A 25-gauge regular-wall needle was prepared and threaded by the dislocated flanged haptic before the flanged haptic was heated and enlarged by low-temperature cautery. RESULTS: The dislocated flanged haptic was probably the result of too small of a diameter of the previously flanged haptic and was triggered by scleral indentation. The rescue technique required only a 25-gauge regular-wall needle to conduct the conventional procedure of flanged haptic intraocular lens fixation before being enlarged by low-temperature cautery. The intraocular lens remained centered and stable at a 3-month postoperative follow-up examination. CONCLUSION: The application of a 25-gauge needle in the salvage of a haptic of dislocated flanged intraocular lens is simple and effective without the need for trimming the flanged tip.

The Classification of Common Macular Diseases Using Deep Learning on Optical Coherence Tomography Images with and without Prior Automated Segmentation.

We compared the performance of deep learning (DL) in the classification of optical coherence tomography (OCT) images of macular diseases between automated classification alone and in combination with automated segmentation. OCT images were collected from patients with neovascular age-related macular degeneration, polypoidal choroidal vasculopathy, diabetic macular edema, retinal vein occlusion, cystoid macular edema in Irvine-Gass syndrome, and other macular diseases, along with the normal fellow eyes. A total of 14,327 OCT images were used to train DL models. Three experiments were conducted: classification alone (CA), use of automated segmentation of the OCT images by RelayNet, and the graph-cut technique before the classification (combination method 1 (CM1) and 2 (CM2), respectively). For validation of classification of the macular diseases, the sensitivity, specificity, and accuracy of CA were found at 62.55%, 95.16%, and 93.14%, respectively, whereas the sensitivity, specificity, and accuracy of CM1 were found at 72.90%, 96.20%, and 93.92%, respectively, and of CM2 at 71.36%, 96.42%, and 94.80%, respectively. The accuracy of CM2 was statistically higher than that of CA (p = 0.05878). All three methods achieved AUC at 97%. Applying DL for segmentation of OCT images prior to classification of the images by another DL model may improve the performance of the classification.

Appraisal of vitreous syphilis antibody as a novel biomarker for the diagnosis of syphilitic uveitis: a prospective case-control study.

PURPOSE: To determine the sensitivity and specificity of syphilis antibody tests in vitreous samples and to propose an algorithm using vitreous syphilis antibody as a supplementary test to confirm syphilitic uveitis (SU). METHODS: A prospective case-control study was conducted at the Retina and Uveitis Clinic from May 2017 to January 2020. Initially, patients were classified based on syphilis serology into group 1 (positive testing) and group 2 (negative testing). Group 1 was further divided into 2 subgroups (group 1A and 1B) depending on their relevant clinical manifestations and clinical improvement. Group 2 served as a control group. RESULTS: Thirty-eight patients were enrolled in the study: 14 in group 1A, 5 in group 1B, and 19 in group 2B. No patient was assigned to group 2A. All patients in group 1A, representing definite SU, completed syphilis test (rapid plasma reagin [RPR], enzyme immunoassay [EIA], and fluorescent treponemal antibody-absorption [FTA-ABS]) for vitreous, and all vitreous samples yielded positive results. Of the 5 subjects in group 1B, 3 cases were considered to be not SU with different conditions, and 2 were indeterminate for SU. They presented with different features not typical of SU, and they had variable and fewer positive syphilis antibody responses. The most sensitive test for detecting syphilis antibodies in vitreous was EIA (90.9%), followed by RPR (80.0%) and FTA-ABS IgG (78.9%). EIA and FTA-ABS had the highest specificity, detecting 100% of the syphilis antibody. CONCLUSIONS: Vitreous analysis of syphilis antibody can serve as a supplementary test to confirm SU in selected cases as the proposed algorithm.

Sutureless and Equipment-Free Technique for Contact Lens Viewing System During Vitreoretinal Surgery.

PURPOSE: To report a sutureless and equipment-free technique for contact lens wide-angle viewing system during vitreoretinal surgery. METHODS: A step-by-step procedure and a surgical video using this technique were proposed. RESULTS: This technique needs only three-finger control of contact lens either by the left hand (for the right-handed surgeon) or right hand (for the left-handed surgeon). The contact lens is not fixed to the environment (lid speculums, cannulas, corneoscleral limbus); therefore, it allows the surgeon to move and tilt the contact lens freely while keeping a clear view of the retina. CONCLUSION: The sutureless and equipment-free technique for contact lens viewing system during vitreoretinal surgery has been tried and tested, and it is practical and effective in the real world.

Recalcitrant infectious uveoscleritis: A case report of a coinfection by Epstein-Barr virus and Talaromyces marneffei.

Introduction and importance: In an immunocompromised patient, there has never been a report of Epstein-Barr virus (EBV) and Talaromyces marneffei co-infection. Treatment with intravenous ganciclovir for EBV-associated uveitis has only been explored in a few cases. Case presentation: A 47-year-old HIV-positive female presented with scleral nodule and vascularized iris tumor in the left eye. After the ancillary laboratory workup, EBV was detected by polymerase chain reaction on aqueous humor and scleral tissue analysis. The lesion subsided completely with systemic and local ganciclovir therapy. Subsequently, recurrent scleral nodule with iris plaque developed and was pathologically diagnosed as Talaromyces marneffei. The employed treatment, including intravenous amphotericin B and intracameral amphotericin B injection, was successful in attaining a significant regression of the lesion. Clinical discussion: Although EBV treatment is debatable, our combination of systemic and local ganciclovir therapy demonstrated regression of EBV-caused scleral nodule and iris plaque. The organism's identification was complicated by Talaromyces marneffei, but it was easily treated. Conclusions: In HIV-infected patients, EBV-associated sclerouveitis can be successfully treated with systemic and local ganciclovir therapy, and there is a possibility of coinfection with Talaromyces marneffei.

ISOLATED RETINAL VASCULITIS: Prognostic Factors and Expanding the Role of Immunosuppressive Treatment in Retinal Vasculitis Associated With Positive QuantiFERON-TB Gold Test.

PURPOSE: To identify prognostic factors for poor visual outcomes in patients with isolated retinal vasculitis and to elucidate the outcome of immunosuppressive treatment without the use of antituberculosis drugs for patients with retinal vasculitis associated with a positive QuantiFERON-TB Gold In-Tube (QFT) test. METHODS: A retrospective chart review was performed of patients presenting with retinal vasculitis. After the diagnosis of active retinal vasculitis had been confirmed by fluorescein angiography and other possible causes of retinal vasculitis had been excluded, patients were categorized into two groups by their QFT result. Potential associated factors between the poor and good visual outcome groups were statistically analyzed by the chi-square test and logistic regression model with generalized estimating equations. RESULTS: Seventy-three eyes (48 patients) were enrolled in this study. After univariate analysis, multivariate logistic regression analysis was performed and revealed that logMAR visual acuity at the initial visit ( P = 0.01) and outer retinal disruption ( P = 0.03) were the two factors significantly associated with poor visual outcomes. Systemic corticosteroids were administered without the use of antituberculosis drugs to all 16 cases of presumed tuberculous retinal vasculitis associated with positive QFT (26 eyes), 10 (63%) of whom were given nonsteroidal immunosuppressive drugs and achieved inflammatory control and treatment success. CONCLUSION: Risk factors leading to poor visual outcome in patients with isolated retinal vasculitis have been identified. Immunosuppressive treatment without antituberculosis drugs seems to be a promising regimen for selected patients with presumed tuberculous retinal vasculitis under vigilant care.

Effects of Ketoconazole on the Clinical Recovery in Central Serous Chorioretinopathy.

Purpose: Patients with hypercortisolism have been associated with a higher prevalence of the pachychoroid spectrum including central serous chorioretinopathy (CSCR), which may explain the inconsistency of therapeutic responses of the mineralocorticoid receptor antagonist because hyperaldosteronism has rarely been detected in patients with CSCR. Therefore, this study aimed to evaluate the effects of ketoconazole, the first-line cortisol inhibitor, on the resolution of subretinal fluid (SRF) in CSCR and to analyze correlations between choroidal thickness and steroid hormones. Patients and Methods: This retrospective cohort study included 41 naïve CSCR eyes of 41 patients categorized into control (20 eyes) and treatment (21 eyes) groups. Patients in the treatment group were administered oral ketoconazole at a daily dose of 400 or 600 mg for 3-6 weeks. At week 12, rescue laser therapy was applied to patients exhibiting persistent SRF. Thus, a survival analysis was performed to determine the time interval from presentation to clinical resolution of SRF. Secondary outcomes consisted of eyes with persistent SRF and factors affecting the therapeutic response. Results: The mean 24-hour urinary free cortisol (UFC) levels were elevated at 181 ± 70 and 150 ± 68 µg/day (range: 20-150) in the treatment and control groups, respectively (p = 0.21). After controlling for age and gender, baseline UFC levels were significantly associated with choroidal thickness in both eyes (p < 0.05). Ketoconazole significantly increased the CSCR resolution with the median time to resolution of 7 vs 16 weeks (p < 0.01) and decreased the proportion of eyes receiving rescue therapy at 12 weeks (23.8% vs 50%; p = 0.01). Prolonged CSCR durations were likely found in elderly patients with thick choroids in fellow eyes. Conclusion: Patients with CSCR showed elevated glucocorticoids, which further correlated with their choroidal thickness. Using cortisol blockers may shorten the duration of existing SRF.

Treatment for Epstein-Barr Virus-associated uveitis confirmed by polymerase chain reaction: Efficacy of Anti-Viral Agents and a literature review.

BACKGROUND: There are still many research challenges and unanswered questions in relation to Epstein-Barr virus-associated uveitis. These include the presence of Epstein-Barr virus (EBV) DNA in asymptomatic patients, its pathogenicity in the uveitis eye, and the role of antiviral therapy for EBV-associated intraocular inflammation. METHODS: This was a retrospective review of prospectively collected data from the Ophthalmology Department, Rajavithi Hospital between 2015 and 2020. A qualitative assay using multiplex real-time PCR was performed to detect pathogen genes from specimens obtained from a total of 344 patients. The main outcome measure was treatment success defined by clinical improvement and absence of viral DNA confirmed by PCR. RESULTS: Of the 35 cases, 24 with complete data were enrolled in the study, including 22 with post-treatment PCR results. Sixty-seven percent were HIV-infected, and other plausible causes or coinfection with other pathogens were found in 75% of patients. Cytomegalovirus (38%) was the most common co-infecting pathogen. The most commonly employed regimen was a combination of systemic acyclovir and intravitreal ganciclovir injection (58%). Of the 22 cases who had post-treatment PCR results, absence of detection of the virus by PCR in the intraocular fluid after treatment was demonstrated in 73% of patients. CONCLUSION: Patients with EBV infection can be simultaneously co-infected with other pathogens. Systemic acyclovir and ganciclovir achieved clinical improvement in most cases, and EBV infection was cured in the majority of patients.

The incidence, risk factor, and time to develop rheumatologic diseases after isolated inflammatory eye diseases: a 12-year cohort study.

OBJECTIVE: The incidence, risk factors, and time to diagnosis of rheumatologic disease (RD) in patients with isolated inflammatory eye diseases (IED) were investigated. MATERIAL AND METHODS: A 12-year bidirectional cohort study was conducted in IED patients who were tested for antinuclear antibody (ANA) and rheumatoid factor (RF). Patients with prior RD were excluded. Impacts of relevant symptoms, signs, and laboratory investigations were analyzed. RESULTS: Seventy-five patients presented with IED including scleritis, anterior uveitis (AU), retinal vasculitis (RV), keratopathy, and optic neuritis (OP). AU, RV, keratopathy, and OP were associated with RD development. The incidence of RD was 36% during 12 years. RD developed most frequently in AU (55.5%) and RV (22.2%). The longest duration for RD development was 5.5 years. Prevalence of positive ANA and RF were 57.3% and 13.3%, respectively. The three most common RDs developed after IEDs were spondyloarthropathy (44.4%), systemic lupus erythematosus (SLE) (18.5%), and Sjogren's syndrome (pSS) (11.1%). The risk factors of RD from univariate analysis were age below 35 years old at onset of IED [relative risk (RR) 3.45; 95% CI (1.13-10.55), p =0.026], positive pertinent findings from history [RR 2.125; 95% CI (1.39-3.25), p<0.001], and physical examination [RR 3.23; 95% CI (1.84-5.66), p<0.001]. Multivariable logistic regression showed that bilateral eye involvement of IED was the significant risk of RD [RR 4.33; 95% CI (1.57-11.96), p=0.004]. IED patients with bilateral eye involvement had shorter time to develop RD (p=0.018). Positive ANA was the predictor for SLE and pSS in RV or OP patients [RR 2.00; 95% CI (1.14-3.52), p=0.04]. CONCLUSIONS: IED with bilateral eye involvement was the risk of RD development. Positive ANA was the predictor of RD in RV and OP patients, but not for the other IEDs. Follow-up period of at least 5 years was required to identify RD. KEY POINTS: • Bilateral involvement of isolated inflammatory eye disease was a significant risk factor of rheumatologic disease development.

Capacity building in screening and treatment of diabetic retinopathy in Asia-Pacific region.

The focus of capacity building for screening and treatment of diabetic retinopathy (DR) is on health professionals who are nonophthalmologists. Both physicians and nonphysicians are recruited for screening DR. Although there is no standardization of the course syllabus for the capacity building, it is generally accepted to keep their sensitivity >80%, specificity >95%, and clinical failure rate <5% for the nonophthalmologists, if possible. A systematic literature search was performed using the PubMed database and the following search terms: diabetic retinopathy, diabetic retinopathy screening, Asia, diabetic retinopathy treatment, age-related macular degeneration, capacity building, deep learning, artificial intelligence (AI), nurse-led clinic, and intravitreal injection (IVI). AI may be a tool for improving their capacity. Capacity building on IVIs of antivascular endothelial growth factors for DR is focused on nurses. There is evidence that, after a supervision of an average of 100 initial injections, the trained nurses can do the injections effectively and safely, the rate of endophthalmitis ranges from 0.03 to 0.07%, comparable to ophthalmologists. However, laws and regulations, which are different among countries, are challenges and barriers for nonophthalmologists, particularly for nonphysicians, for both screening and treatment of DR. Even if nonphysicians or physicians who are nonophthalmologists are legally approved for these tasks, sustainability of the capacity is another important challenge, this may be achieved if the capacity building can be part of their career development. Patient acceptability is another important barrier for initiating care provided by nonophthalmologists, particularly in Asia. There are also collaborations between national eye institutes of high-income countries, nongovernment organizations, and local eye institutes to improve both the quality and quantity of ophthalmologists and retinal specialists in low-income countries in Asia. This approach may require more labor, cost, and time consuming than training nonophthalmologists.

Endophthalmitis after pars plana vitrectomy with reused single-use devices: a 13-year retrospective study.

BACKGROUND: To describe the incidence, clinical characteristics, and treatment outcomes of endophthalmitis after pars plana vitrectomy (PPV) with recycled single-use devices. The recommended sterilization process as well as safety measures are discussed. METHODS: Medical charts of patients who developed endophthalmitis after PPV were retrospectively reviewed and reported in a descriptive manner. Cases undergoing PPV for preexisting endophthalmitis or open globe injury were excluded. Data collection included patient demographics, operative details, ocular findings, microbiological profiles, treatment modalities, and visual outcomes. RESULTS: Over the past thirteen years, a total of 12,989 pars plana vitrectomy operations were included. In total, 13 eyes of 13 cases (0.10%) experienced endophthalmitis after vitrectomy. These occurred in 3 cases (0.11%) using 20-gauge vitrectomy compared to 8 cases (0.09%) using 23-gauge vitrectomy and 2 cases (0.18%) using 25-gauge vitrectomy. There were no statistically significant differences between the 20-gauge and microincisional vitrectomy surgery (MIVS) group (P = 0.64), and the 23- and 25-gauge approach (P = 0.34). Causative pathogens were positive by culture in 5 cases (45%): 3 g-positive cases, 1 g-negative case, and 1 fungus case. CONCLUSIONS: The rate of endophthalmitis in patients who underwent 23-gauge PPV was comparable to those who underwent 25-gauge PPV. With our standardized protocol for instrument sterilization, endophthalmitis rates in those undergoing PPV using recycled single-use instruments were within the range of previously published results in which vitrectomy tools were disposed of after one use.

Analysis of Three-Dimensional Choroidal Volume with Enhanced Depth Imaging Findings in Patients with Recurrent Vogt-Koyanagi-Harada Disease.

Purpose: To demonstrate changes in three-dimensional choroidal volume with enhanced depth imaging optical coherence tomography (EDI-OCT) in patients with recurrent stage of Vogt-Koyanagi-Harada disease (VKH).Materials and Methods: This prospective comparative case series included 9 patients with recurrent VKH, 10 patients with quiet VKH, and 15 healthy controls after sample size was calculated. All VKH cases with recurrences underwent raster scanning with EDI-OCT at active and inactive stages of the disease.Results: All choroidal parameters in the active stage significantly reduced when the inflammation subsided: total choroidal volume (P = .02), central choroidal volume (P = .01), central choroidal thickness (P = .03). The changes in central choroidal volume over the resolution phase were more pronounced than the changes in central choroidal thickness in 56% of cases. Two cases presenting with only subclinical posterior segment recurrence had their choroidal parameters recovered after prompt treatment.Conclusions: In the recurrent stage of VKH, alteration in choroidal volume was evident by EDI-OCT even in an absence of anterior segment inflammation. Central choroidal volume may serve as a biomarker for detecting choroidal morphological change.

Poor prognostic factors in post-traumatic endophthalmitis following open globe injury.

AIM: To demonstrate prognostic factors for poor visual outcome in patients with post-traumatic endophthalmitis (PTE) following open globe injury. METHODS: A retrospective study was conducted on 66 patients (66 eyes) with PTE following open globe injury from 2005 to 2015. Potential factors accounting for good and poor visual outcome were statistically analyzed by Chi-square test and Logistic regression model. RESULTS: In 66 cases, 39 cases (59%) had a poor visual outcome. Univariate and multivariate Logistic regression analysis identified retained intraocular foreign body (IOFB) as the only factor significantly associated with poor visual outcome [adjusted odds ratio, 4.62; 95% confidence interval (1.04-20.53); P=0.04]. The most common causative agents were gram-positive organisms (83%), of which Bacillus cereus (33%), was the most common pathogen. All cases received intravitreal antibiotic injections. Oral ciprofloxacin was the most used systemic antibiotic (33%). Pars plana vitrectomy was performed in 83% (55/66) of cases. At 6mo follow-up, mean BCVA was 1.74±0.72 logMAR units. CONCLUSION: In patients with PTE following open globe injury, the only predictor of poor visual outcome is the presence of IOFB. Bacillus cereus is the most isolated microorganism.

Predicting optical coherence tomography-derived diabetic macular edema grades from fundus photographs using deep learning.

Center-involved diabetic macular edema (ci-DME) is a major cause of vision loss. Although the gold standard for diagnosis involves 3D imaging, 2D imaging by fundus photography is usually used in screening settings, resulting in high false-positive and false-negative calls. To address this, we train a deep learning model to predict ci-DME from fundus photographs, with an ROC-AUC of 0.89 (95% CI: 0.87-0.91), corresponding to 85% sensitivity at 80% specificity. In comparison, retinal specialists have similar sensitivities (82-85%), but only half the specificity (45-50%, p < 0.001). Our model can also detect the presence of intraretinal fluid (AUC: 0.81; 95% CI: 0.81-0.86) and subretinal fluid (AUC 0.88; 95% CI: 0.85-0.91). Using deep learning to make predictions via simple 2D images without sophisticated 3D-imaging equipment and with better than specialist performance, has broad relevance to many other applications in medical imaging.

VITREOUS TREPONEMAL ANTIBODY AS A SUPPLEMENTARY TEST TO SEROLOGY FOR THE CONFIRMATION OF SYPHILITIC CHORIORETINITIS.

PURPOSE: To report the novel application of nontreponemal and treponemal antibody to confirm diagnosis of ocular syphilis from vitreous samples. METHODS: Two distinct case reports emphasizing the importance of confirmatory vitreous treponemal antibody. Multimodal imaging of patients was also applied. RESULTS: We report two distinct cases with positive serum treponemal antibody but opposing vitreous treponemal antibody results. One case with a positive vitreous test responded well to antisyphilitic treatment. By contrast, a case with a negative vitreous result was changed to serpiginous choroiditis, eventually cured by immunomodulatory treatment. CONCLUSION: Intraocular fluid analysis of nontreponemal and treponemal antibody may play an important role in ruling out suspected ocular syphilis in settings without a polymerase chain reaction facility, especially immunocompromised patients who are at risk of multiple infections. Further studies are needed to establish the sensitivity and specificity of nontreponemal and treponemal antibody test on vitreous samples.

Deep learning versus human graders for classifying diabetic retinopathy severity in a nationwide screening program.

Deep learning algorithms have been used to detect diabetic retinopathy (DR) with specialist-level accuracy. This study aims to validate one such algorithm on a large-scale clinical population, and compare the algorithm performance with that of human graders. A total of 25,326 gradable retinal images of patients with diabetes from the community-based, nationwide screening program of DR in Thailand were analyzed for DR severity and referable diabetic macular edema (DME). Grades adjudicated by a panel of international retinal specialists served as the reference standard. Relative to human graders, for detecting referable DR (moderate NPDR or worse), the deep learning algorithm had significantly higher sensitivity (0.97 vs. 0.74, p < 0.001), and a slightly lower specificity (0.96 vs. 0.98, p < 0.001). Higher sensitivity of the algorithm was also observed for each of the categories of severe or worse NPDR, PDR, and DME (p < 0.001 for all comparisons). The quadratic-weighted kappa for determination of DR severity levels by the algorithm and human graders was 0.85 and 0.78 respectively (p < 0.001 for the difference). Across different severity levels of DR for determining referable disease, deep learning significantly reduced the false negative rate (by 23%) at the cost of slightly higher false positive rates (2%). Deep learning algorithms may serve as a valuable tool for DR screening.

Erratum: Author Correction: Deep learning versus human graders for classifying diabetic retinopathy severity in a nationwide screening program.

[This corrects the article DOI: 10.1038/s41746-019-0099-8.].

Risk factors for endophthalmitis after cataract surgery in diabetic patients: a case control study.

AIM: To identify risk factors associated with post-cataract surgery endophthalmitis (PCE) in type 2 diabetic patients. METHODS: A hospital-based retrospective case-control study was conducted on 194 type 2 diabetic patients undergoing cataract surgery in Rajavithi Hospital from January 2007 to December 2015. Fifteen patients with PCE were included as the case group and 179 patients without PCE were included as the control group. Potential factors associated with PCE among both groups including demographics, pre-operative characteristics, surgical settings and complications, were statistically analyzed using Chi-square testing and a logistic regression model. RESULTS: Within the case group, 53% were females and the median age was 68y. Univariate analysis of pre-operative characteristics, surgical settings and complications revealed that recent pre-operative fasting plasma glucose, insulin therapy, presence of diabetic retinopathy, and severe non-proliferative or proliferative diabetic retinopathy were significantly associated with PCE. In a multivariate analysis adjusting for blood glucose level, insulin treatment was the only significant factor associated with an increased risk of PCE (OR 3.9, 95%CI 1.0-15.0, P=0.04) compared to patients without insulin treatment. The most common causative organisms were gram-positive bacteria (89%). Staphylococcus species represented the most common group (67%). Median best corrected visual acuity at 1-month and 3-month follow-up was equal at 0.7 logMAR (20/100). CONCLUSION: The authors identify insulin treatment as the only risk factor associated with endophthalmitis after cataract surgery in type 2 diabetic patients. Further studies with serum levels of pre-operative glycated hemoglobin (HbA1c) and post-operative fasting plasma glucose level are essential to truly demonstrate the role of peri-operative glycemic markers as a risk factor for PCE.