Purpose To develop and validate neural network (NN) vs logistic regression (LR) diagnostic prediction models in patients with suspected giant cell arteritis (GCA). Design: Multicenter retrospective chart review. Methods An audit of consecutive patients undergoing temporal artery biopsy (TABx) for suspected GCA was conducted at 14 international medical centers. The outcome variable was biopsy-proven GCA. The predictor variables were age, gender, headache, clinical temporal artery abnormality, jaw claudication, vision loss, diplopia, erythrocyte sedimentation rate, C-reactive protein, and platelet level. The data were divided into three groups to train, validate, and test the models. The NN model with the lowest false-negative rate was chosen. Internal and external validations were performed. Results Of 1,833 patients who underwent TABx, there was complete information on 1,201 patients, 300 (25%) of whom had a positive TABx. On multivariable LR age, platelets, jaw claudication, vision loss, log C-reactive protein, log erythrocyte sedimentation rate, headache, and clinical temporal artery abnormality were statistically significant predictors of a positive TABx ( P ≤0.05). The area under the receiver operating characteristic curve/Hosmer–Lemeshow P for LR was 0.867 (95% CI, 0.794, 0.917)/0.119 vs NN 0.860 (95% CI, 0.786, 0.911)/0.805, with no statistically significant difference of the area under the curves ( P =0.316). The misclassification rate/false-negative rate of LR was 20.6%/47.5% vs 18.1%/30.5% for NN. Missing data analysis did not change the results. Conclusion Statistical models can aid in the triage of patients with suspected GCA. Misclassification remains a concern, but cutoff values for 95% and 99% sensitivities are provided ( https://goo.gl/THCnuU ).

To the best of our knowledge this is the largest study that has investigated the association between postoperative topical PGA or topical beta-blocker use and the incidence of pseudophakic CME. Postoperative use of both topical PGAs and topical beta-blockers was found to be associated with the incidence of pseudophakic CME.

Purpose: To investigate the relationship between corneal collagen cross-linking (CXL) and the number of corneal transplants required for the treatment of keratoconus (KCN) in 2 major Canadian provinces. Methods: This is a retrospective review of all corneal transplantation performed in Ontario and British Columbia over an 18-year period (1998–2016). Data were collected at the Eye Bank of Canada—Ontario/British Columbia Divisions. The primary outcome was to determine the change in proportion and absolute number of corneal transplants required for treatment of KCN since the introduction of CXL in Canada in 2008. Results: A total of 31,943 grafts were included. Overall, the mean age of participants was 39.3 ± 2.2 years, with our cohort being composed of 28% of women and 72% of men. The results showed a significant decrease in the proportion of total transplants required for KCN between 1998 and 2016 [1998–2008 (pre-CXL), range: 14.77%–12.63%; 2009–2016 (post-CXL), range: 12.98%–5.50%, P < 0.001]. However, there was no change in the absolute number of grafts performed for KCN over this time (pre-CXL: 179 ± 26 grafts; post-CXL: 198 ± 27 grafts; P = 0.5), whereas the total number of grafts (pre-CXL: 1318 ± 183 grafts; post-CXL: 2181 ± 404; P < 0.001) and endothelial keratoplasties (pre-CXL: 59 ± 108; post-CXL: 966 ± 431 grafts; P < 0.001) increased significantly. In addition, there were no changes in penetrating keratoplasty/deep anterior lamellar keratoplasty (DALK) performed for indications other than KCN (pre-CXL: 1080 ± 157; post-CXL: 1017 ± 92; P > 0.5). Conclusions: Although there has been a significant decrease in the proportion of corneal graft rates for KCN since the introduction of CXL as a factor of all transplants performed for all indications, this result is most likely because of an increase in endothelial keratoplasties rather than decreased transplants performed for definitive treatment.

No abstract