Purpose: To develop a machine learning (ML) classifier for predicting post-induction hypotension (PIH) in non-cardiac surgeries. Materials and Methods: Preoperative data and early vital signs were obtained from 3669 cases in the VitalDB database, an opensource registry. PIH was defined as sustained mean arterial pressure (MAP) <65 mm Hg within 20 minutes since induction or from induction to incision. Six different ML algorithms were used to create binary classifiers to predict PIH. The primary outcome was the area under the receiver operating characteristic curve (AUROC) of ML classifiers. Results: A total of 2321 (63.3%) cases exhibited PIH. Among ML classifiers, the random forest regressor and extremely gradient boosting regressor showed the highest AUROC, both recording a value of 0.772. Excluding these models, the light gradient boosting machine regressor showed the second highest AUROC [0.769; 95% confidence interval (CI), 0.767–0.771], followed by the gradient boosting regressor (0.768; 95% CI, 0.763–0.772), AdaBoost regressor (0.752; 95% CI, 0.743–0.761), and automatic relevance determination regression (0.685; 95% CI, 0.669–0.701). The top three important features were mean diastolic blood pressure (DBP), minimum MAP, and minimum DBP from anesthetic induction to tracheal intubation, and these features were lower in cases with PIH (all p<0.001). Conclusion ML classifiers exhibited moderate performance in predicting PIH, and have the potential for real-time prediction.

Purpose: To develop a machine learning (ML) classifier for predicting post-induction hypotension (PIH) in non-cardiac surgeries. Materials and Methods: Preoperative data and early vital signs were obtained from 3669 cases in the VitalDB database, an opensource registry. PIH was defined as sustained mean arterial pressure (MAP) <65 mm Hg within 20 minutes since induction or from induction to incision. Six different ML algorithms were used to create binary classifiers to predict PIH. The primary outcome was the area under the receiver operating characteristic curve (AUROC) of ML classifiers. Results: A total of 2321 (63.3%) cases exhibited PIH. Among ML classifiers, the random forest regressor and extremely gradient boosting regressor showed the highest AUROC, both recording a value of 0.772. Excluding these models, the light gradient boosting machine regressor showed the second highest AUROC [0.769; 95% confidence interval (CI), 0.767–0.771], followed by the gradient boosting regressor (0.768; 95% CI, 0.763–0.772), AdaBoost regressor (0.752; 95% CI, 0.743–0.761), and automatic relevance determination regression (0.685; 95% CI, 0.669–0.701). The top three important features were mean diastolic blood pressure (DBP), minimum MAP, and minimum DBP from anesthetic induction to tracheal intubation, and these features were lower in cases with PIH (all p<0.001). Conclusion ML classifiers exhibited moderate performance in predicting PIH, and have the potential for real-time prediction.

Purpose: To develop a machine learning (ML) classifier for predicting post-induction hypotension (PIH) in non-cardiac surgeries. Materials and Methods: Preoperative data and early vital signs were obtained from 3669 cases in the VitalDB database, an opensource registry. PIH was defined as sustained mean arterial pressure (MAP) <65 mm Hg within 20 minutes since induction or from induction to incision. Six different ML algorithms were used to create binary classifiers to predict PIH. The primary outcome was the area under the receiver operating characteristic curve (AUROC) of ML classifiers. Results: A total of 2321 (63.3%) cases exhibited PIH. Among ML classifiers, the random forest regressor and extremely gradient boosting regressor showed the highest AUROC, both recording a value of 0.772. Excluding these models, the light gradient boosting machine regressor showed the second highest AUROC [0.769; 95% confidence interval (CI), 0.767–0.771], followed by the gradient boosting regressor (0.768; 95% CI, 0.763–0.772), AdaBoost regressor (0.752; 95% CI, 0.743–0.761), and automatic relevance determination regression (0.685; 95% CI, 0.669–0.701). The top three important features were mean diastolic blood pressure (DBP), minimum MAP, and minimum DBP from anesthetic induction to tracheal intubation, and these features were lower in cases with PIH (all p<0.001). Conclusion ML classifiers exhibited moderate performance in predicting PIH, and have the potential for real-time prediction.

Purpose: To develop a machine learning (ML) classifier for predicting post-induction hypotension (PIH) in non-cardiac surgeries. Materials and Methods: Preoperative data and early vital signs were obtained from 3669 cases in the VitalDB database, an opensource registry. PIH was defined as sustained mean arterial pressure (MAP) <65 mm Hg within 20 minutes since induction or from induction to incision. Six different ML algorithms were used to create binary classifiers to predict PIH. The primary outcome was the area under the receiver operating characteristic curve (AUROC) of ML classifiers. Results: A total of 2321 (63.3%) cases exhibited PIH. Among ML classifiers, the random forest regressor and extremely gradient boosting regressor showed the highest AUROC, both recording a value of 0.772. Excluding these models, the light gradient boosting machine regressor showed the second highest AUROC [0.769; 95% confidence interval (CI), 0.767–0.771], followed by the gradient boosting regressor (0.768; 95% CI, 0.763–0.772), AdaBoost regressor (0.752; 95% CI, 0.743–0.761), and automatic relevance determination regression (0.685; 95% CI, 0.669–0.701). The top three important features were mean diastolic blood pressure (DBP), minimum MAP, and minimum DBP from anesthetic induction to tracheal intubation, and these features were lower in cases with PIH (all p<0.001). Conclusion ML classifiers exhibited moderate performance in predicting PIH, and have the potential for real-time prediction.

Purpose: To develop a machine learning (ML) classifier for predicting post-induction hypotension (PIH) in non-cardiac surgeries. Materials and Methods: Preoperative data and early vital signs were obtained from 3669 cases in the VitalDB database, an opensource registry. PIH was defined as sustained mean arterial pressure (MAP) <65 mm Hg within 20 minutes since induction or from induction to incision. Six different ML algorithms were used to create binary classifiers to predict PIH. The primary outcome was the area under the receiver operating characteristic curve (AUROC) of ML classifiers. Results: A total of 2321 (63.3%) cases exhibited PIH. Among ML classifiers, the random forest regressor and extremely gradient boosting regressor showed the highest AUROC, both recording a value of 0.772. Excluding these models, the light gradient boosting machine regressor showed the second highest AUROC [0.769; 95% confidence interval (CI), 0.767–0.771], followed by the gradient boosting regressor (0.768; 95% CI, 0.763–0.772), AdaBoost regressor (0.752; 95% CI, 0.743–0.761), and automatic relevance determination regression (0.685; 95% CI, 0.669–0.701). The top three important features were mean diastolic blood pressure (DBP), minimum MAP, and minimum DBP from anesthetic induction to tracheal intubation, and these features were lower in cases with PIH (all p<0.001). Conclusion ML classifiers exhibited moderate performance in predicting PIH, and have the potential for real-time prediction.

Objective: We compared the osteoblastogenesis by serially administrating recombinant human bone morphogenetic protein-2 (rhBMP-2) and osteoprotegerin-immunoglobulin Fc segment complex (OPG-Fc). Methods: The MC3T3-E1 preosteoblast cell line was differentiated for 1, 3, and 7 days with a treatment of OPG-Fc in 10~200 ng/mL concentration and the cell viability was evaluated by Cell Counting Kit-8 analysis. The level of differentiation from MC3T3-E1 cells to osteoblasts was determined by alkaline phosphatase activity. The level of runt domain-containing transcription factor 2 (Runx2) and osteopontin (OPN) manifestation, involved in osteoblast differentiation, was examined by real-time polymerase chain reaction and western blotting. Results: During MC3T3-E1 cell differentiation, the differentiation level was high with 1-day treatment using 100 ng/mL OPGFc. The treatment with 50 ng/mL rhBMP-2 for 7 days, followed by 1-day treatment with 100 ng/mL OPG-Fc produced the highest differentiation level, which was approximately 5.3 times that of the control group (p<0.05). The expression of Runx2 mRNA significantly increased, reaching 2.5 times the level of the control group under the condition of 7-day treatment with rhBMP-2 and 1-day treatment with OPG-Fc (p<0.001). The expression of Runx2 protein significantly increased to approximately 5.7 times that of the control group under the condition of 7-day treatment with rhBMP-2, followed by 1-day treatment with OPG-Fc (p<0.01).The expression of OPN protein showed no change from that of the control group under various conditions of rhBMP-2 and OPGFc combinations. Conclusion These results imply that the treating preosteoblasts with rhBMP-2 first and then with OPG-Fc increased osteoblast differentiation efficacy.

Background/Aims: The Gout Impact Scale (GIS), a part of the Gout Assessment Questionnaire 2.0, is used to measure gout-specific health-related quality of life (HRQOL). Although several studies have been conducted on the factors affecting the HRQOL of patients with gout, few have focused on lifestyle factors. This study aimed to investigate the correlation between lifestyle habits and HRQOL using the GIS in patients with gout. Methods: We used data from the Urate-Lowering TheRApy in Gout (ULTRA) registry, a prospective cohort of Korean patients with gout treated at multiple centers nationwide. The patients were aged ≥18 years and met the 2015 American College of Rheumatology/European League Against Rheumatism gout classification criteria. They were asked to complete a GIS and questions regarding their lifestyle habits at enrollment. Results: The study included 232 patients. ‘Gout concern overall’ scores in the GIS were significantly lower in patients who exercised more frequently and consumed soft drinks and meat less, and ‘well-being during attack’ scores were significantly lower in patients who consumed vegetables and exercised more frequently. The frequency of vegetable consumption had a negative linear relationship with the ‘well-being during attack’ and ‘gout concern during attack’ scores (p = 0.01, p = 0.001, respectively). The frequency of exercise had a negative linear relationship with the ‘gout concern overall’ and ‘gout concern during attack’ scores (p = 0.04 and p = 0.002, respectively). Conclusions Patients with gout who frequently consumed vegetables and exercised regularly experienced less impact of gout, exhibiting a better GIS that represented HRQOL.

Purpose: To evaluate the reliability and validity of the Cataract-related Visual Function Questionnaire (CVFQ). Methods: A prospective cross-sectional study of 141 cataract patients was conducted from March 2022 to June 2022. The questionnaire was created based on a literature review and advice from an expert panel. This study determined its construct validity, criterion validity, internal consistency, and test-retest reliability. Results: The CVFQ consists of 15 items distributed among five categories: overall visual quality, overall visual function, distance vision, near vision, and glare. In the exploratory factor analysis of validity, the first three principal components explained 77.8% of the variance. The p-values in the Spearman correlation test comparing the pre- and postoperative total CVFQ score and best-corrected visual acuity (BCVA) were 0.006 and 0.004, respectively. In the reliability analysis, Cronbach’s alpha was > 0.9 for internal consistency and the p-values of each subcategory were all significant in the analysis of test-retest reliability. Conclusions Our results indicate that the CVFQ is useful for measuring the visual quality and visual function of cataract patients in Korea.

Objective: : Stent-assisted coil embolization (SAC) has been increasingly used to treat various types of intracranial aneurysms. Delayed thromboembolic complications are major concerns regarding this procedure, so dual antiplatelet therapy with aspirin and clopidogrel is needed. However, clinicians vary the duration of dual antiplatelet therapy after SAC, and no randomized study has been performed. This study aims to compare the safety and efficacy of long-term (12 months) dual antiplatelet therapy and shortterm dual antiplatelet therapy (6 months) after SAC for patients with unruptured intracranial aneurysms (UIAs). Methods: : This is a prospective, randomized and multicenter trial to investigate the optimal duration of dual antiplatelet therapy after SAC in patients with UIAs. Subjects will receive dual antiplatelet therapy for 6 months (short-term group) or 12 months (longterm group) after SAC. The primary endpoint is the assessment of thromboembolic complications between 1 and 18 months after SAC. We will enroll 528 subjects (264 subjects in each group) and perform 1 : 1 randomization. This study will involve 14 topperforming, high-volume Korean institutions specializing in coil embolization. Results: : The trial will begin enrollment in 2022, and clinical data will be available after enrollment and follow-up. Conclusion : This article describes that the aim of this prospective randomized multicenter trial is to compare the effect of short-term (6 months) and long-term (12 months) dual antiplatelet therapy on UIAs in patients undergoing SAC, and to find the optimal duration.

Objective: High disease activity of ankylosing spondylitis (AS) is associated with poor sleep quality. The purpose of this study was to identify which of the representative tools for evaluating the disease activity of AS best reflect the quality of sleep. Methods: A total of 107 AS patients were enrolled in the study and the sleep quality was assessed using the Pittsburgh Sleep Quality Index (PSQI). Age, sex, concomitant medication, erythrocyte sedimentation rate (ESR), serum C-reactive protein (CRP) level, Beck Depression Inventory second edition (BDI-II), Bath ankylosing spondylitis disease activity index (BASDAI), ankylosing spondylitis disease activity score (ASDAS)-ESR, ASDAS-CRP, pain visual analog scale, Insomnia Severity Index (ISI), and Epworth Sleepiness Scale (ESS) were analyzed as covariates. Results: Overall, 65% (70/107) of subjects reported poor sleep quality (PSQI＞5). There was a positive correlation between the sleep quality and disease activity as measured by the BASDAI, ASDAS-ESR, and ASDAS-CRP. In addition, the BASDAI demonstrated good correlations with ISI, ESS, and BDI-II, respectively. However, only BASDAI showed reliable correlation with PSQI among the disease activity parameters of AS (adjusted odd ratio 5.36, p=0.023). Conclusion BASDAI is the most reliable parameter of disease activity associated with the sleep quality in patients with AS.