Filtering Bleb Size in the Early Postoperative Period Affects the Long-term Surgical Outcome after Trabeculectomy.

PURPOSE: To investigate whether postoperative filtering bleb size affects the surgical outcome after trabeculectomy. METHODS: In this study, we retrospectively reviewed 145 medically uncontrolled glaucoma patients with intraocular pressure (IOP) values >21 mmHg before surgery and data from ≥2 years of follow-up. Postoperative IOP, filtering bleb size including extent and height, and other clinical factors were measured after trabeculectomy. We divided bleb extent into quadrants and bleb height by 0.5 intervals of corneal thickness. The main outcome measure was surgical success. We confirmed complete success when the IOP was ≤21 mmHg and decreased by >20% from baseline without medication or additional procedures. Qualified success used the same criteria but allowed for medication or additional procedures. Cases with reoperation or two consecutive IOP measurements <6 mmHg were considered failures. RESULTS: A total of 145 eyes of 145 patients was included. The average observation period was 30.8 ± 10.9 months. During multivariate Cox regression analysis, a larger extent of filtering bleb revealed significantly low hazard ratios in both complete and surgical success (0.509 and 0.494, respectively); however, there was no significant relationship between bleb height and surgical outcome. CONCLUSIONS: The extent of the filtering bleb was associated with surgical outcomes of trabeculectomy in glaucoma patients.

Border tissue morphology is associated with the pattern of visual field progression in open-angle glaucoma.

The etiology of open-angle glaucoma (OAG) is yet unclear. This study investigated possible risk factors, such as the morphology of the border tissue that affect the pattern of visual field (VF) progression in eyes with OAG. 166 eyes of 166 OAG patients with an externally oblique border tissue (EOBT) at least in one direction were included. EOBT was obtained by analyzing enhanced depth imaging spectral-domain optical coherence tomography images. A pointwise linear regression was used to determine VF progression by measuring the deterioration rate of each point in the VF. The odds ratio of VF progression for each risk factor was estimated using logistic regression analysis. Seventy (42.2%) eyes showed VF deterioration. In multivariate analysis, longer follow-up period, higher baseline intraocular pressure (IOP), lower mean ocular perfusion pressure (MOPP), and smaller angular location of the longest EOBT were associated with VF progression (all p values were below 0.05). In the multivariate analysis, the location of the longest EOBT was significantly associated with inferior (p = 0.002) and central (p = 0.017) VF progression. In conclusion, VF progression pattern in OAG eyes is associated with the location of the longest EOBT as well as other known risk factors.

Mortality prediction of patients in intensive care units using machine learning algorithms based on electronic health records.

Improving predictive models for intensive care unit (ICU) inpatients requires a new strategy that periodically includes the latest clinical data and can be updated to reflect local characteristics. We extracted data from all adult patients admitted to the ICUs of two university hospitals with different characteristics from 2006 to 2020, and a total of 85,146 patients were included in this study. Machine learning algorithms were trained to predict in-hospital mortality. The predictive performance of conventional scoring models and machine learning algorithms was assessed by the area under the receiver operating characteristic curve (AUROC). The conventional scoring models had various predictive powers, with the SAPS III (AUROC 0.773 [0.766-0.779] for hospital S) and APACHE III (AUROC 0.803 [0.795-0.810] for hospital G) showing the highest AUROC among them. The best performing machine learning models achieved an AUROC of 0.977 (0.973-0.980) in hospital S and 0.955 (0.950-0.961) in hospital G. The use of ML models in conjunction with conventional scoring systems can provide more useful information for predicting the prognosis of critically ill patients. In this study, we suggest that the predictive model can be made more robust by training with the individual data of each hospital.