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.

Background: There are few safe effective ways to relieve osteoarthritis (OA) pain; as a result, off-label psychotropic drug prescriptions have increased worldwide. This study examined the change in psychotropic drug prescriptions for patients with OA from 2011 to 2020 using the Korean National Health Insurance Service dataset. Methods: The study population consisted of patients with hip or knee OA aged ≥ 65 years.Psychotropic drugs included opioids, benzodiazepines, non-benzodiazepine hypnotics (Z-drugs), anti-epileptics, tricyclic antidepressants, selective serotonin reuptake inhibitors (SSRIs), serotonin and norepinephrine reuptake inhibitors (SNRIs), typical antipsychotics, atypical antipsychotics, and anxiolytics. The prevalence and long-term (> 3 months) prescription rates of psychotropic drugs in OA patients were calculated. Results: The study included 1,821,158 patients with OA (mean age 71.7 years; 65.32% female).Of the cohort, 49% had comorbidities for which psychotropics were indicated. The prevalence of psychotropic prescriptions decreased from 58.2% to 52.0% in 2018 and then leveled off.The long-term prescription rate remained constant until 2018 and then increased slightly.The most commonly prescribed psychotropics were opioids and long- and short-acting benzodiazepines. The prescription rates of opioids and long-acting benzodiazepines decreased from 2011 to 2020. For those with psychiatric co-morbidities, the prescription rates of anti-epileptics and SNRIs increased, while the prescription rates of anti-epileptics, SSRIs, other antidepressants, and atypical psychotropics increased for those without such co-morbidities. The most commonly prescribed psychotropics were diazepam and alprazolam, excluding tramadol and tramadol–acetaminophen combination. For those with psychiatric co-morbidities, the prescription rates of gabapentin and fentanyl increased, while for those without such co-morbidities, the prescription rates of lorazepam, fentanyl, escitalopram and quetiapine increased. Conclusion A significant number of older Korean patients with OA were prescribed psychotropic drugs in the absence of comorbidities requiring such drugs, including drugs that have little effect on OA and unfavorable safety profiles in older adults.

Background: There are few safe effective ways to relieve osteoarthritis (OA) pain; as a result, off-label psychotropic drug prescriptions have increased worldwide. This study examined the change in psychotropic drug prescriptions for patients with OA from 2011 to 2020 using the Korean National Health Insurance Service dataset. Methods: The study population consisted of patients with hip or knee OA aged ≥ 65 years.Psychotropic drugs included opioids, benzodiazepines, non-benzodiazepine hypnotics (Z-drugs), anti-epileptics, tricyclic antidepressants, selective serotonin reuptake inhibitors (SSRIs), serotonin and norepinephrine reuptake inhibitors (SNRIs), typical antipsychotics, atypical antipsychotics, and anxiolytics. The prevalence and long-term (> 3 months) prescription rates of psychotropic drugs in OA patients were calculated. Results: The study included 1,821,158 patients with OA (mean age 71.7 years; 65.32% female).Of the cohort, 49% had comorbidities for which psychotropics were indicated. The prevalence of psychotropic prescriptions decreased from 58.2% to 52.0% in 2018 and then leveled off.The long-term prescription rate remained constant until 2018 and then increased slightly.The most commonly prescribed psychotropics were opioids and long- and short-acting benzodiazepines. The prescription rates of opioids and long-acting benzodiazepines decreased from 2011 to 2020. For those with psychiatric co-morbidities, the prescription rates of anti-epileptics and SNRIs increased, while the prescription rates of anti-epileptics, SSRIs, other antidepressants, and atypical psychotropics increased for those without such co-morbidities. The most commonly prescribed psychotropics were diazepam and alprazolam, excluding tramadol and tramadol–acetaminophen combination. For those with psychiatric co-morbidities, the prescription rates of gabapentin and fentanyl increased, while for those without such co-morbidities, the prescription rates of lorazepam, fentanyl, escitalopram and quetiapine increased. Conclusion A significant number of older Korean patients with OA were prescribed psychotropic drugs in the absence of comorbidities requiring such drugs, including drugs that have little effect on OA and unfavorable safety profiles in older adults.

Background: There are few safe effective ways to relieve osteoarthritis (OA) pain; as a result, off-label psychotropic drug prescriptions have increased worldwide. This study examined the change in psychotropic drug prescriptions for patients with OA from 2011 to 2020 using the Korean National Health Insurance Service dataset. Methods: The study population consisted of patients with hip or knee OA aged ≥ 65 years.Psychotropic drugs included opioids, benzodiazepines, non-benzodiazepine hypnotics (Z-drugs), anti-epileptics, tricyclic antidepressants, selective serotonin reuptake inhibitors (SSRIs), serotonin and norepinephrine reuptake inhibitors (SNRIs), typical antipsychotics, atypical antipsychotics, and anxiolytics. The prevalence and long-term (> 3 months) prescription rates of psychotropic drugs in OA patients were calculated. Results: The study included 1,821,158 patients with OA (mean age 71.7 years; 65.32% female).Of the cohort, 49% had comorbidities for which psychotropics were indicated. The prevalence of psychotropic prescriptions decreased from 58.2% to 52.0% in 2018 and then leveled off.The long-term prescription rate remained constant until 2018 and then increased slightly.The most commonly prescribed psychotropics were opioids and long- and short-acting benzodiazepines. The prescription rates of opioids and long-acting benzodiazepines decreased from 2011 to 2020. For those with psychiatric co-morbidities, the prescription rates of anti-epileptics and SNRIs increased, while the prescription rates of anti-epileptics, SSRIs, other antidepressants, and atypical psychotropics increased for those without such co-morbidities. The most commonly prescribed psychotropics were diazepam and alprazolam, excluding tramadol and tramadol–acetaminophen combination. For those with psychiatric co-morbidities, the prescription rates of gabapentin and fentanyl increased, while for those without such co-morbidities, the prescription rates of lorazepam, fentanyl, escitalopram and quetiapine increased. Conclusion A significant number of older Korean patients with OA were prescribed psychotropic drugs in the absence of comorbidities requiring such drugs, including drugs that have little effect on OA and unfavorable safety profiles in older adults.

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.

Background: There are few safe effective ways to relieve osteoarthritis (OA) pain; as a result, off-label psychotropic drug prescriptions have increased worldwide. This study examined the change in psychotropic drug prescriptions for patients with OA from 2011 to 2020 using the Korean National Health Insurance Service dataset. Methods: The study population consisted of patients with hip or knee OA aged ≥ 65 years.Psychotropic drugs included opioids, benzodiazepines, non-benzodiazepine hypnotics (Z-drugs), anti-epileptics, tricyclic antidepressants, selective serotonin reuptake inhibitors (SSRIs), serotonin and norepinephrine reuptake inhibitors (SNRIs), typical antipsychotics, atypical antipsychotics, and anxiolytics. The prevalence and long-term (> 3 months) prescription rates of psychotropic drugs in OA patients were calculated. Results: The study included 1,821,158 patients with OA (mean age 71.7 years; 65.32% female).Of the cohort, 49% had comorbidities for which psychotropics were indicated. The prevalence of psychotropic prescriptions decreased from 58.2% to 52.0% in 2018 and then leveled off.The long-term prescription rate remained constant until 2018 and then increased slightly.The most commonly prescribed psychotropics were opioids and long- and short-acting benzodiazepines. The prescription rates of opioids and long-acting benzodiazepines decreased from 2011 to 2020. For those with psychiatric co-morbidities, the prescription rates of anti-epileptics and SNRIs increased, while the prescription rates of anti-epileptics, SSRIs, other antidepressants, and atypical psychotropics increased for those without such co-morbidities. The most commonly prescribed psychotropics were diazepam and alprazolam, excluding tramadol and tramadol–acetaminophen combination. For those with psychiatric co-morbidities, the prescription rates of gabapentin and fentanyl increased, while for those without such co-morbidities, the prescription rates of lorazepam, fentanyl, escitalopram and quetiapine increased. Conclusion A significant number of older Korean patients with OA were prescribed psychotropic drugs in the absence of comorbidities requiring such drugs, including drugs that have little effect on OA and unfavorable safety profiles in older adults.

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.

Background: Remimazolam, a new benzodiazepine, is known for its quick onset of effects and recovery time. Recently, it has been licensed for general anesthesia and sedation in Korea and its use is increasing in other countries. However, less is known about its effect on postoperative recovery. We used a patient-reported outcome questionnaire to examine the effect of remimazolam on postoperative recovery. Methods: Patients who underwent hysteroscopy on day surgery basis were administered an induction dose of remimazolam 6 mg/kg/h followed by a maintenance dose of 1–2 mg/kg/h. After surgery, the translated Korean version of 15-item Quality of Recovery scale (QoR-15K) including post-discharge nausea and vomiting (PDNV) and/or pain, was surveyed 24 h after surgery to evaluate patient recovery. Results: Total of 38 patients were enrolled in this prospective, observational study. All patients successfully completed QoR-15K. Only one patient scored low for moderate pain and PDNV. On average, patients scored 9 and above for all QoR-15K items except for moderate pain (8.66 ± 1.68). When QoR-15K items were grouped into dimensions, all dimensions scored an average of 9 or higher on a 10-point scale. In addition, 19 out of 38 patients gave score range of 148 to 150 out of possible 150. Conclusions Psychometric evaluation based on postoperative QoR-15K among patients receiving remimazolam shows satisfactory patient recovery profiles without significant pain or PDNV. Considering its effectiveness and safety, remimazolam could be one of useful agents for general anesthesia of day surgery in terms of postoperative recovery.