Prediction of Visual Outcomes After Diabetic Vitrectomy Using Clinical Factors From Common Data Warehouse.

Purpose: We sought to analyze the visual outcome and systemic prognostic factors for diabetic vitrectomy and predicted outcomes using these factors. Methods: This was a multicenter electronic medical records (EMRs) review study of 1504 eyes with type 2 diabetes that underwent vitrectomy for proliferative diabetic retinopathy at 6 university hospitals. Demographics, laboratory results, intra-operative findings, and visual acuity (VA) values were analyzed and correlated with visual outcomes at 1 year after the vitrectomy. Prediction models for visual outcomes were obtained using machine learning. Results: At 1 year, VA was 1.0 logarithm of minimal angle resolution (logMAR) or greater (poor visual outcome group) in 456 eyes (30%). Baseline visual acuity, duration of diabetes treatment, tractional membrane, silicone oil tamponade, smoking, and vitreous hemorrhage correlated with logMAR VA at 1 year (r = 0.450, -0.159, 0.221, 0.280, 0.067, and -0.105; all P ≤ 0.036). An ensemble decision tree model trained using all variables generated accuracy, specificity, F1 score (the harmonic means of which precision and sensitivity), and receiver-operating characteristic curve area under curve values of 0.77, 0.66, 0.85, and 0.84 for the prediction of poor visual outcomes at 1 year after vitrectomy. Conclusions: Visual outcome after diabetic vitrectomy is associated with pre- and intra-operative findings and systemic factors. Poor visual outcome after diabetic vitrectomy was predictable using clinical factors. Intensive care in patients who are predicted to result in poor vision may limit vision loss resulting from type 2 diabetes. Translational Relevance: This study demonstrates that a real world EMR big data could predict outcome after diabetic vitrectomy using clinical factors.

Long-term prediction models for vision-threatening diabetic retinopathy using medical features from data warehouse.

We sought to evaluate the performance of machine learning prediction models for identifying vision-threatening diabetic retinopathy (VTDR) in patients with type 2 diabetes mellitus using only medical data from data warehouse. This is a multicenter electronic medical records review study. Patients with type 2 diabetes screened for diabetic retinopathy and followed-up for 10 years were included from six referral hospitals sharing same electronic medical record system (n = 9,102). Patient demographics, laboratory results, visual acuities (VAs), and occurrence of VTDR were collected. Prediction models for VTDR were developed using machine learning models. F1 score, accuracy, specificity, and area under the receiver operating characteristic curve (AUC) were analyzed. Machine learning models revealed F1 score, accuracy, specificity, and AUC values of up 0.89, 0.89.0.95, and 0.96 during training. The trained models predicted the occurrence of VTDR at 10-year with F1 score, accuracy, and specificity up to 0.81, 0.70, and 0.66, respectively, on test set. Important predictors included baseline VA, duration of diabetes treatment, serum level of glycated hemoglobin and creatinine, estimated glomerular filtration rate and blood pressure. The models could predict the long-term occurrence of VTDR with fair performance. Although there might be limitation due to lack of funduscopic findings, prediction models trained using medical data can facilitate proper referral of subjects at high risk for VTDR to an ophthalmologist from primary care.

Effectiveness of Toric Orthokeratology in the Treatment of Patients with Combined Myopia and Astigmatism.

PURPOSE: The purpose of this multi-institute, single-group clinical trial was to evaluate the effectiveness and safety of toric orthokeratology lenses for the treatment of patients with combined myopia and astigmatism. METHODS: A total of 44 patients were included in this clinical trial. The patients ranged in age from 7 to 49 years, with myopia of -0.75 to -6.0 diopters (D) and astigmatism of 1.25 to 4.0 D. After excluding 21 subjects, 23 subjects (39 eyes) were analyzed after toric orthokeratology lens use. The subjects underwent ophthalmologic examination after 1 day and 1, 2, 3, and 4 weeks of wearing overnight toric orthokeratology lenses. RESULTS: A total of 19 subjects (31 eyes) completed the trial after five subjects (eight eyes) dropped out. In the patients who completed the study by wearing lenses for 4 weeks, the myopic refractive error decreased significantly by 2.60 ± 2.21 D (p < 0.001), from -3.65 ± 1.62 to -1.05 ± 1.64 D. The astigmatic refractive error were also significantly decreased by 0.63 ± 0.98 D (p = 0.001), from 2.07 ± 0.83 to 1.44 ± 0.99 D. The mean uncorrected and corrected visual acuities before wearing the lenses were 2.14 ± 0.80 logarithm of the logMAR (logMAR) and 0.05 ± 0.13 logMAR, respectively, which changed to 0.12 ± 0.30 logarithm of the logMAR (p < 0.001) and 0.01 ± 0.04 logMAR (p = 0.156) after 4 weeks. No serious adverse reactions were reported during the clinical trial. CONCLUSIONS: Our results suggest that toric orthokeratology is an effective and safe treatment for correcting visual acuity in patients with combined myopia and astigmatism.

Lower Lid Laxity is Negatively Correlated with Improvement of the Ocular Surface Disease Index in Dry Eye Treatment.

PURPOSE: To compare the responses to dry eye treatment of patients sorted by the degree of lower lid laxity. METHODS: Sixty patients were grouped into three groups according to the degree of lower lid laxity. Tear break-up time (TBUT), Schirmer test (ST) scores, ocular surface disease index (OSDI) scores, and changes in OSDI score in each group were compared, before and at 3 months after treatment. RESULTS: TBUT, ST, and OSDI scores were not different among the three groups at baseline. TBUT improved in each group at 3 months after treatment, and no differences between groups were found. ST scores were not increased after treatment, while OSDI were improved to 22.57 ± 5.243, 31.16 ± 11.353, and 37.85 ± 13.342 in the no, moderate, and high laxity groups, respectively; these improvements were statistically significant (p = 0.003, <0.001, <0.001, respectively). Patients with greater than moderate lower lid laxity saw the smallest improvement in response to dry eye treatment, as assessed by change in OSDI score (p = 0.005 versus moderate laxity group, p = 0.005 versus no laxity group). CONCLUSIONS: Lower lid laxity is one of the factors contributing to the responses to dry eye treatment assessed by change in OSDI score, independent of TBUT and ST scores.

Endothelial Cell Loss after Phacoemulsification according to Different Anterior Chamber Depths.

Purpose. To compare the loss of corneal endothelial cells after phacoemulsification according to different anterior chamber depths (ACDs). Methods. We conducted a prospective study on 135 eyes with senile cataracts. Eyes with nuclear density grades of 2 to 4 were divided into three groups according to ACD: ACD I, 1.5 < ACD ≤ 2.5 mm; ACD II, 2.5 < ACD ≤ 3.5 mm; or ACD III, 3.5 < ACD ≤ 4.5 mm. Intraoperative mean cumulative dissipated energy (CDE) was measured. Clinical examinations included central corneal thickness (CCT) and endothelial cell count (ECC) preoperatively and 2 months postoperatively. Results. There were no significant differences in CDE among the ACD groups (P > 0.05). Endothelial cell loss was significantly higher in ACD I than in ACD III in grades 3 and 4 cataract density groups 2 months after phacoemulsification (P < 0.05). There were also more changes in CCT in all of the cataract density groups in the ACD I group compared to the ACD II and III groups 2 months postoperatively, but the difference was not statistically significant. Conclusions. Eyes with shallow ACDs, especially those with relatively hard cataract densities, can be vulnerable to more corneal endothelial cell loss in phacoemulsification surgery.