A Bayesian method to detect drug-drug interaction using external information for spontaneous reporting system.

Due to the insufficiency of safety assessments of clinical trials for drugs, further assessments are required for post-marketed drugs. In addition to adverse drug reactions (ADRs) induced by one drug, drug-drug interaction (DDI)-induced ADR should also be investigated. The spontaneous reporting system (SRS) is a powerful tool for evaluating the safety of drugs continually. In this study, we propose a novel Bayesian method for detecting potential DDIs in a database collected by the SRS. By applying a power prior, the proposed method can borrow information from similar drugs for a drug assessed DDI to increase sensitivity of detection. The proposed method can also adjust the amount of the information borrowed by tuning the parameters in power prior. In the simulation study, we demonstrate the aforementioned increase in sensitivity. Depending on the scenarios, approximately 20 points of sensitivity of the proposed method increase from an existing method to a maximum. We also indicate the possibility of early detection of potential DDIs by the proposed method through analysis of the database shared by the Food and Drug Administration. In conclusion, the proposed method has a higher sensitivity and a novel criterion to detect potential DDIs early, provided similar drugs have similar observed-expected ratios to the drug under assessment.

Artificial intelligence-assisted interventions for perioperative anesthetic management: a systematic review and meta-analysis.

BACKGROUND: Integration of artificial intelligence (AI) into medical practice has increased recently. Numerous AI models have been developed in the field of anesthesiology; however, their use in clinical settings remains limited. This study aimed to identify the gap between AI research and its implementation in anesthesiology via a systematic review of randomized controlled trials with meta-analysis (CRD42022353727). METHODS: We searched the databases of Medical Literature Analysis and Retrieval System Online (MEDLINE), Excerpta Medica Database (Embase), Web of Science, Cochrane Central Register of Controlled Trials (CENTRAL), Institute of Electrical and Electronics Engineers Xplore (IEEE), and Google Scholar and retrieved randomized controlled trials comparing conventional and AI-assisted anesthetic management published between the date of inception of the database and August 31, 2023. RESULTS: Eight randomized controlled trials were included in this systematic review (n = 568 patients), including 286 and 282 patients who underwent anesthetic management with and without AI-assisted interventions, respectively. AI-assisted interventions used in the studies included fuzzy logic control for gas concentrations (one study) and the Hypotension Prediction Index (seven studies; adding only one indicator). Seven studies had small sample sizes (n = 30 to 68, except for the largest), and meta-analysis including the study with the largest sample size (n = 213) showed no difference in a hypotension-related outcome (mean difference of the time-weighted average of the area under the threshold 0.22, 95% confidence interval -0.03 to 0.48, P = 0.215, I2 93.8%). CONCLUSIONS: This systematic review and meta-analysis revealed that randomized controlled trials on AI-assisted interventions in anesthesiology are in their infancy, and approaches that take into account complex clinical practice should be investigated in the future. TRIAL REGISTRATION: This study was registered with the International Prospective Register of Systematic Reviews (PROSPERO ID: CRD42022353727).

Relationship between changes in pulmonary function and patient-reported outcomes of lung cancer surgery.

PURPOSE: To investigate the relationship between changes in pulmonary function (PF) and patient-reported outcomes (PROs) of lung cancer surgery. METHODS: We recruited 262 patients who underwent lung resection for lung cancer, to evaluate the PROs, using the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire (EORTC QLQ)-C30 and the Lung Cancer 13-question supplement (LC13). The patients underwent PF tests and PRO assessments preoperatively (Pre) and 1 year after surgery (Y1). Changes were calculated by subtracting the value at Pre from the value at Y1. We set two cohorts: patients under the ongoing protocol (Cohort 1) and patients who were eligible for lobectomy with clinical stage I lung cancer (Cohort 2). RESULTS: Cohorts 1 and 2 comprised 206 and 149 patients, respectively. In addition to dyspnea, changes in PF were also correlated with scores for global health status, physical and role function scores, fatigue, nausea and vomiting, pain, and financial difficulties. Absolute correlation coefficient values ranged from 0.149 to 0.311. Improvement of emotional and social function scores was independent of PF. Sublobar resection preserved PF more than lobectomy did. Wedge resection mitigated dyspnea in both cohorts. CONCLUSION: The correlation between PF and PROs was found to be weak; therefore, further studies are needed to improve the patient's postoperative experience.

Risk of postoperative pneumonia after extubation with the positive pressure versus normal pressure technique: a single-center retrospective observational study.

PURPOSE: A normal pressure extubation technique (no lung inflation before extubation), proposed by the Japanese Society of Anesthesiologists to prevent droplet infection during the coronavirus disease 2019 (COVID-19) pandemic, could theoretically increase postoperative pneumonia incidence compared with a positive pressure extubation technique (lung inflation before extubation). However, the normal pressure extubation technique has not been adequately evaluated. This study compared postoperative pneumonia incidence between positive and normal pressure extubation techniques using a dataset from the University of Tsukuba Hospital. METHODS: In our hospital, the extubation methods changed from positive to normal pressure extubation techniques on March 3, 2020 due to the COVID-19 pandemic. Thus, we compared the risk of postoperative pneumonia between the positive (April 1, 2017 to December 31, 2019) and normal pressure extubation techniques (March 3, 2020 to March 31, 2022) using propensity score analyses. Postoperative pneumonia was defined using the International Classification of Diseases, 10th Edition (ICD-10) codes (J13-J18), and we reviewed the medical records of patients flagged with these ICD-10 codes (preoperative pneumonia and ICD-10 codes for prophylactic antibiotic prescriptions for pneumonia). RESULTS: We identified 20,011 surgeries, including 11,920 in the positive pressure extubation group (mean age 48.2 years, standard deviation [SD] 25.2 years) and 8,091 in the normal pressure extubation group (mean age 47.8 years, SD 25.8 years). The postoperative pneumonia incidences were 0.19% (23/11,920) and 0.17% (14/8,091) in the positive and normal pressure extubation groups, respectively. The propensity score analysis using inverse probability weighting revealed no significant difference in postoperative pneumonia incidence between the two groups (adjusted odds ratio 0.98, 95% confidence interval 0.50 to 1.91, P = 0.94). CONCLUSIONS: These results indicated no increased risk of postoperative pneumonia associated with the normal pressure extubation technique compared with the positive pressure extubation technique. CLINICAL TRIAL NUMBER: Clinical trial number: UMIN000048589 https://center6.umin.ac.jp/cgi-open-bin/ctr/ctr_view.cgi?recptno=R000055364.

Continuous glucose monitoring-derived time in range and CV are associated with altered tissue characteristics of the carotid artery wall in people with type 2 diabetes.

AIMS/HYPOTHESIS: Previous studies have suggested that glucose variability may accelerate atherosclerosis progression in people with type 2 diabetes. Current guidelines recommend assessing glycaemic control using continuous glucose monitoring (CGM), which provides a comprehensive glycaemic profile to supplement HbA1c measurement. However, the association between CGM-derived metrics and atherosclerosis progression is not entirely clear. METHODS: This exploratory study used baseline data and data obtained after 104 weeks from an ongoing prospective, multicentre, observational study. Six hundred study participants with type 2 diabetes and no apparent history of symptomatic cardiovascular disease underwent CGM and ultrasonographic atherosclerosis measurements of the carotid arteries, including the intima-media thickness (IMT) and grey-scale median (GSM), at baseline and 104 weeks. Non-invasive ultrasonic tissue characterisation of the carotid artery wall or plaque using the GSM reflects vascular composition. Multivariate regression models were used to analyse the association between CGM-derived indices, mainly time in range (TIR) and CV, and changes in carotid atherosclerosis index values. RESULTS: Over the 104-week study period, there were modest increases in mean IMT (from 0.759±0.153 to 0.773±0.152 mm, p<0.001) and thickened-lesion GSM (from 43.5±19.5 to 53.9±23.5 units, p<0.001), but no significant changes in common carotid artery maximum-IMT (from 1.109±0.442 to 1.116±0.469 mm, p=0.453) or mean GSM (from 48.7±19.3 to 49.8±20.8 units, p=0.092). In a linear regression model with adjustment for possible atherosclerotic risk factors, including HbA1c, TIR and CV at baseline were significantly associated with the annual change in mean GSM (regression coefficient per 10% increase in TIR 0.52; 95% CI 0.06, 0.98; Hochberg-adjusted p value 0.038; regression coefficient per 1% increase in CV -0.12; 95% CI -0.22, -0.02; Hochberg-adjusted p value 0.038). TIR and CV at baseline were also significantly associated with the annual change in thickened-lesion GSM (regression coefficient per 10% increase in TIR 0.95; 95% CI 0.12, 1.79; Hochberg-adjusted p value 0.038; regression coefficient per 1% increase in CV -0.19; 95% CI -0.36, -0.01; Hochberg-adjusted p value 0.038). Participants who achieved target CGM-derived metrics at baseline, as proposed by an international consensus, showed significant annual changes in mean GSM compared with those who did not (0.94±6.88 vs -0.21±6.19 units/year, p=0.007). CONCLUSIONS/INTERPRETATION: TIR and CV were significantly associated with changes in the tissue characteristics of the carotid artery wall. TRIAL REGISTRATION: University Hospital Medical Information Network Clinical Trials Registry, number UMIN000032325.

A comparison of bias-adjusted generalized estimating equations for sparse binary data in small-sample longitudinal studies.

Using a generalized estimating equation (GEE) can lead to a bias in regression coefficients for a small sample or sparse data. The bias-corrected GEE (BCGEE) and penalized GEE (PGEE) were proposed to resolve the small-sample bias. Moreover, the standard sandwich covariance estimator leads to a bias of standard error for small samples; several modified covariance estimators have been proposed to address this issue. We review the modified GEEs and modified covariance estimators, and evaluate their performance in sparse binary data from small-sample longitudinal studies. The simulation results showed that GEE and BCGEE often failed to achieve convergence, whereas the convergence proportion for PGEE was quite high. The bias for the regression coefficients was generally in the ascending order of PGEE < $$ < $$ BCGEE < $$ < $$ GEE. However, PGEE and BCGEE did not sufficiently remove the bias involving 20-30 subjects with unequal exposure levels with a 5% response rate. The coverage probability (CP) of the confidence interval for BCGEE was relatively poor compared with GEE and PGEE. The CP with the sandwich covariance estimator deteriorated regardless of the GEE methods under the small sample size and low response rate, whereas the CP with the modified covariance estimators-such as Morel's method-was relatively acceptable. PGEE will be the reasonable way for analyzing sparse binary data in small-sample studies. Instead of using the standard sandwich covariance estimator, one should always apply the modified covariance estimators for analyzing these data.

Bias in Odds Ratios From Logistic Regression Methods With Sparse Data Sets.

BACKGROUND: Logistic regression models are widely used to evaluate the association between a binary outcome and a set of covariates. However, when there are few study participants at the outcome and covariate levels, the models lead to bias of the odds ratio (OR) estimated using the maximum likelihood (ML) method. This bias is known as sparse data bias, and the estimated OR can yield impossibly large values because of data sparsity. However, this bias has been ignored in most epidemiological studies. METHODS: We review several methods for reducing sparse data bias in logistic regression. The primary aim is to evaluate the Bayesian methods in comparison with the classical methods, such as the ML, Firth's, and exact methods using a simulation study. We also apply these methods to a real data set. RESULTS: Our simulation results indicate that the bias of the OR from the ML, Firth's, and exact methods is considerable. Furthermore, the Bayesian methods with hyper-É¡ prior modeling of the prior covariance matrix for regression coefficients reduced the bias under the null hypothesis, whereas the Bayesian methods with log F-type priors reduced the bias under the alternative hypothesis. CONCLUSION: The Bayesian methods using log F-type priors and hyper-É¡ prior are superior to the ML, Firth's, and exact methods when fitting logistic models to sparse data sets. The choice of a preferable method depends on the null and alternative hypothesis. Sensitivity analysis is important to understand the robustness of the results in sparse data analysis.

Robot-Assisted Training as Self-Training for Upper-Limb Hemiplegia in Chronic Stroke: A Randomized Controlled Trial.

BACKGROUND: This study aimed to examine whether robotic self-training improved upper-extremity function versus conventional self-training in mild-to-moderate hemiplegic chronic stroke patients. METHODS: Study design was a multi-center, prospective, randomized, parallel-group study comparing three therapist-guided interventions (1-hour sessions, 3×/wk, 10 weeks). We identified 161 prospective patients with chronic, poststroke, upper-limb hemiplegia treated at participating rehabilitation centers. Patients were enrolled between November 29, 2016, and November 12, 2018 in Japan. A blinded web-based allocation system was used to randomly assign 129 qualifying patients into 3 groups: (1) conventional self-training plus conventional therapy (control, N=42); (2) robotic self-training (ReoGo-J) plus conventional therapy (robotic therapy [RT], N=44); or (3) robotic self-training plus constraint-induced movement therapy (N=43). PRIMARY OUTCOME: Fugl-Meyer Assessment for upper-extremity. SECONDARY OUTCOMES: Motor Activity Log-14 amount of use and quality of movement; Fugl-Meyer Assessment shoulder/elbow/forearm, wrist, finger, and coordination scores; Action Research Arm Test Score; Motricity Index; Modified Ashworth Scale; shoulder, elbow, forearm, wrist, and finger range of motion; and Stroke Impact Scale (the assessors were blinded). Safety outcomes were adverse events. RESULTS: Safety was assessed in 127 patients. An intention-to-treat full analysis set (N=121), and a per-protocol set (N=115) of patients who attended 80% of sessions were assessed. One severe adverse event was recorded, unrelated to the robotic device. No significant differences in Fugl-Meyer Assessment for upper-extremity scores were observed between groups (RT versus control: -1.04 [95% CI, -2.79 to 0.71], P=0.40; RT versus movement therapy: -0.33 [95% CI, -2.02 to 1.36], P=0.90). The RT in the per-protocol set improved significantly in the Fugl-Meyer Assessment for upper-extremity shoulder/elbow/forearm score (RT versus control: -1.46 [95% CI, -2.63 to -0.29]; P=0.037). CONCLUSIONS: Robotic self-training did not improve upper-limb function versus usual self-training, but may be effective combined with conventional therapy in some populations (per-protocol set). REGISTRATION: URL: https://www.umin.ac.jp/ctr; Unique identifier: UMIN000022509.

Relationship between blood glucose variability in ambulatory glucose profile and standardized continuous glucose monitoring metrics: Subanalysis of a prospective cohort study.

AIM: To clarify the relationship between ambulatory glucose profile (AGP) indexes and standardized continuous glucose monitoring (CGM) metrics in patients with type 2 diabetes (T2D). METHODS: This is an exploratory, cross-sectional analysis of baseline data collected from a prospective, multicentre, 5-year follow-up observational study conducted and published previously by our group. The study participants were 999 outpatients with T2D who used CGM at baseline, and had no apparent history of cardiovascular disease. We investigated the relationship between average interquartile range (IQR) and time in range (TIR). We also calculated, for the first time, the cutoff values to achieve the TIR target values. RESULTS: In both the TIR more than 70% and TIR more than 90% achievement groups, the average IQR was notably small compared with the non-achievement groups. Particularly in comparison of the TIR quartiles, the average IQR became significantly smaller as the TIR became larger. The average IQR correlated negatively with TIR, and the cutoff values for TIR of more than 70% achievement and TIR of more than 90% achievement were an average IQR (>70%/>90%) of 2.13/1.85 mmol/L. CONCLUSION: Our results showed a negative correlation between TIR and the range of blood glucose variations visually represented in AGP. The results also showed that the range of blood glucose variations in AGP is associated with indices of intraday and interday blood glucose variations and also with hypoglycaemia. Our results may provide new perspectives in the assessment and application of AGP in the clinical setting.

Statistical tests for two-stage adaptive seamless design using short- and long-term binary outcomes.

The adaptive seamless design combining phases II and III into a single trial has been shown growing interest for improving the efficiency of drug development, becoming the most frequent adaptive design type. It typically consists of two stages, the trial objectives being often different in each stage. The primary objectives are to select optimal experimental treatment group(s) in the first stage and compare the efficacy between the selected treatment and control groups in the second stage. In this article, we focus on a two-stage adaptive seamless design, for which treatment selection is based on the short-term binary endpoint and treatment comparison is based on the long-term binary endpoint. We thus propose an exact conditional test as a final analysis, based on the bivariate binomial distribution and given the selected treatment with the most promising short-term endpoint response rate from an interim analysis. Additionally, the mid- p $$ p $$ approach is incorporated to improve conservativeness for an exact test. Simulation studies were conducted to compare the proposed methods with a method based on the combination test. The proposed exact method controlled for type I error rate at the nominal level, regardless of the number of initial treatments or the correlation between short- and long-term endpoints. In terms of the treatment comparison power, the proposed methods are more powerful than that based on the combination test in the scenarios, with only one treatment being effective.

Real-World Investigation on Anticoagulation Management Before and After Catheter Ablation for Atrial Fibrillation in Japan　- Periprocedural and Long-Term Outcomes.

BACKGROUND: Optimal periprocedural oral anticoagulant (OAC) therapy before catheter ablation (CA) for atrial fibrillation (AF) and the safety profile of OAC discontinuation during the remote period (from 31 days and up to 1 year after CA) have not been well defined.Methods and Results: The RYOUMA registry is a prospective multicenter observational study of Japanese patients who underwent CA for AF in 2017-2018. Of the 3,072 patients, 82.3% received minimally interrupted direct-acting OACs (DOACs) and 10.2% received uninterrupted DOACs. Both uninterrupted and minimally interrupted DOACs were associated with an extremely low thromboembolic event rate. Female, long-standing persistent AF, low creatinine clearance, hepatic disorder, and high intraprocedural heparin dose were independent factors associated with periprocedural major bleeding. At 1 year after CA, DOAC was continued in 55.9% of patients and warfarin in 56.4%. The incidence of thromboembolic and major bleeding events for 1 year was 0.3% and 1.2%, respectively. Age ≥73 years, dementia, and AF recurrence were independently associated with major bleeding events. Univariate analyses revealed that warfarin continuation and off-label overdose of DOACs were risk factors for major bleeding after CA. CONCLUSIONS: High intraprocedural dose of heparin was associated with periprocedural major bleeding events. At 1 year after CA, over half of the patients had continued OAC therapy. Thromboembolic events were extremely low; however, major bleeding occurred in 1.2%. Age ≥73 years, dementia, and AF recurrence were independently associated with major bleeding after CA.

Associations between continuous glucose monitoring-derived metrics and arterial stiffness in Japanese patients with type 2 diabetes.

BACKGROUND: Previous studies have suggested that high mean glucose levels and glycemic abnormalities such as glucose fluctuation and hypoglycemia accelerate the progression of atherosclerosis in patients with type 2 diabetes. Although continuous glucose monitoring (CGM) that could evaluate such glycemic abnormalities has been rapidly adopted, the associations between CGM-derived metrics and arterial stiffness are not entirely clear. METHODS: This exploratory cross-sectional study used baseline data from an ongoing prospective, multicenter, observational study with 5 years of follow-up. Study participants included 445 outpatients with type 2 diabetes and no history of apparent cardiovascular disease who underwent CGM and brachial-ankle pulse wave velocity (baPWV) measurement at baseline. Associations between CGM-derived metrics and baPWV were analyzed using multivariate regression models. RESULTS: In a linear regression model, all CGM-derived metrics were significantly associated with baPWV, but HbA1c was not. Some CGM-derived metrics related to intra-day glucose variability, hyperglycemia, and hypoglycemia remained significantly associated with baPWV after adjusting for possible atherosclerotic risk factors, including HbA1c. Based on baPWV ≥ 1800 cm/s as indicative of high arterial stiffness, multivariate logistic regression found that some CGM-derived metrics related to intra-day glucose variability and hyperglycemia are significantly associated with high arterial stiffness even after adjusting for possible atherosclerotic risk factors, including HbA1c. CONCLUSIONS: Multiple CGM-derived metrics are significantly associated with baPWV and high arterial stiffness in patients with type 2 diabetes who have no history of apparent cardiovascular disease. These metrics might be useful for identifying patients at high risk of developing cardiovascular disease.

Associations of continuous glucose monitoring-assessed glucose variability with intima-media thickness and ultrasonic tissue characteristics of the carotid arteries: a cross-sectional analysis in patients with type 2 diabetes.

BACKGROUND: The association between glucose variability and the progression of atherosclerosis is not completely understood. We aimed to evaluate the associations of glucose variability with the progression of atherosclerosis in the early stages. METHODS: We conducted a cross-sectional analysis to investigate the associations of glucose variability, assessed by continuous glucose monitoring, with intima-media thickness (IMT) and gray-scale median (GSM) of the carotid arteries, which are different indicators for the progression of atherosclerosis. We used baseline data from a hospital-based multicenter prospective observational cohort study among Japanese patients with type 2 diabetes without a history of cardiovascular diseases aged between 30 and 80 years. Continuous glucose monitoring was performed by Freestyle Libre Pro, and glucose levels obtained every 15 min for a maximum of eight days were used to calculate the metrics of glucose variability. IMT and GSM were evaluated by ultrasonography, and the former indicates thickening of intima-media complex in the carotid artery wall, while the latter indicates tissue characteristics. RESULTS: Among 600 study participants (age: 64.9 ± 9.2 (mean ± SD) years; 63.2%: men; HbA1c: 7.0 ± 0.8%), participants with a larger intra- and inter-day glucose variability had a lower GSM and most of these associations were statistically significant. No trend based on glucose variability was shown regarding IMT. Standard deviation of glucose (regression coefficient, ß = - 5.822; 95% CI - 8.875 to - 2.768, P < 0.001), glucose coefficient of variation (ß = - 0.418; - 0.685 to - 0.151, P = 0.002), mean amplitude of glycemic excursion (ß = - 1.689; - 2.567 to - 0.811, P < 0.001), mean of daily differences (ß = - 6.500; - 9.758 to - 3.241, P < 0.001), and interquartile range (ß = - 4.289; - 6.964 to - 1.614, P = 0.002) had a statistically significant association with mean-GSM after adjustment for conventional cardiovascular risk factors, including HbA1c. No metrics of glucose variability had a statistically significant association with IMT. CONCLUSIONS: Continuous glucose monitoring-assessed glucose variability was associated with the tissue characteristics of the carotid artery wall in type 2 diabetes patients without cardiovascular diseases.

An application of the mixed-effects model and pattern mixture model to treatment groups with differential missingness suspected not-missing-at-random.

Likelihood-based, mixed-effects models for repeated measures (MMRMs) are occasionally used in primary analyses for group comparisons of incomplete continuous longitudinal data. Although MMRM analysis is generally valid under missing-at-random assumptions, it is invalid under not-missing-at-random (NMAR) assumptions. We consider the possibility of bias of estimated treatment effect using standard MMRM analysis in a motivational case, and propose simple and easily implementable pattern mixture models within the framework of mixed-effects modeling, to handle the NMAR data with differential missingness between treatment groups. The proposed models are a new form of pattern mixture model that employ a categorical time variable when modeling the outcome and a continuous time variable when modeling the missingness-data patterns. The models can directly provide an overall estimate of the treatment effect of interest using the average of the distribution of the missingness indicator and a categorical time variable in the same manner as MMRM analysis. Our simulation results indicate that the bias of the treatment effect for MMRM analysis was considerably larger than that for the pattern mixture model analysis under NMAR assumptions. In the case study, it would be dangerous to interpret only the results of the MMRM analysis, and the proposed pattern mixture model would be useful as a sensitivity analysis for treatment effect evaluation.

Sample size calculations for single-arm survival studies using transformations of the Kaplan-Meier estimator.

In single-arm clinical trials with survival outcomes, the Kaplan-Meier estimator and its confidence interval are widely used to assess survival probability and median survival time. Since the asymptotic normality of the Kaplan-Meier estimator is a common result, the sample size calculation methods have not been studied in depth. An existing sample size calculation method is founded on the asymptotic normality of the Kaplan-Meier estimator using the log transformation. However, the small sample properties of the log transformed estimator are quite poor in small sample sizes (which are typical situations in single-arm trials), and the existing method uses an inappropriate standard normal approximation to calculate sample sizes. These issues can seriously influence the accuracy of results. In this paper, we propose alternative methods to determine sample sizes based on a valid standard normal approximation with several transformations that may give an accurate normal approximation even with small sample sizes. In numerical evaluations via simulations, some of the proposed methods provided more accurate results, and the empirical power of the proposed method with the arcsine square-root transformation tended to be closer to a prescribed power than the other transformations. These results were supported when methods were applied to data from three clinical trials.

A note on the bias of standard errors when orthogonality of mean and variance parameters is not satisfied in the mixed model for repeated measures analysis.

The mixed effect models for repeated measures (MMRM) analysis is sometimes used as a primary analysis in longitudinal randomized clinical trials. The SE for the treatment effect in the MMRM analysis is usually estimated by assuming the orthogonality of the fixed effect and variance-covariance parameters, which is the orthogonality property of a multivariate normal distribution, because of default settings of most standard statistical software. However, this property might be lost when analysis models are misspecified and/or data include missing values with the mechanism of being missing at random. In this study, we investigated the effect of the assumption of the orthogonality property on the estimation of the SE for the MMRM analysis. From simulation and case studies, it was shown that the SE with the assumption of orthogonality property had nonnegligible bias, especially when the analysis models assuming heteroscedasticity between treatment groups were applied. We also introduce the SAS code for the MMRM analysis without assuming the orthogonality property. Assuming the orthogonality property in the MMRM analysis would lead to invalid statistical inference, and it is necessary to be careful when applying the MMRM analysis with most standard software.

Borrowing external information to improve Bayesian confidence propagation neural network.

PURPOSE: A Bayesian confidence propagation neural network (BCPNN) is a signal detection method used by the World Health Organization Uppsala Monitoring Centre to analyze spontaneous reporting system databases. We modify the BCPNN to increase its sensitivity for detecting potential adverse drug reactions (ADRs). METHOD: In a BCPNN, the information component (IC) is defined as an index of disproportionality between the observed and expected number of reported drugs and events. Our proposed method adjusts the IC value by borrowing information about events that have occurred in drugs defined as similar to the target drug. We compare the performance of our method with that of a traditional BCPNN through a simulation study. RESULTS: The false positive rate of the proposed method was lower than that of the traditional BCPNN method and close to the nominal value, 0.025, around the true difference in ICs between the target drug and similar drugs equal to 0. The sensitivity of the proposed method was much higher than that of the traditional BCPNN method in case in which the difference in ICs between the target drug and similar drugs ranges from 0 to 2. When applied to a database managed by Japanese regulatory authority, the proposed method could detect known ADRs earlier than the traditional method. CONCLUSIONS: The proposed method is a novel criterion for early detection of signals if similar drugs have the same tendencies. The proposed BCPNN tends to have higher sensitivity when the true difference is greater than 0.

Endoscopic Detachable Snare Ligation Improves the Treatment for Colonic Diverticular Hemorrhage.

BACKGROUND/AIMS: Recently, endoscopic detachable snare ligation (EDSL) has become increasingly common as treatment for colonic diverticular hemorrhage. This study aimed to evaluate the efficacy and safety of EDSL in comparison with endoscopic clipping (EC) as treatment for colonic diverticular hemorrhage. METHODS: From April 2013 to September 2017, 131 patients were treated with EDSL or EC at the Tokyo Metropolitan Bokutoh Hospital. We retrospectively evaluated patient characteristics and clinical outcomes, including early rebleeding rates (rebleeding within 30 days after initial hemostasis) and complications for each procedure. RESULTS: Of 131 patients, 44 and 87 were treated with EDSL and EC respectively. We initially achieved endoscopic hemostasis in all patients. The early rebleeding rate was significantly lower for EDSL (6.8%, 3 patients) than for EC (23.0%, 20 patients). There were no differences in the total procedure time (43 vs. 45 min, p = 0.84) or time to hemostasis after identification of bleeding site (12 vs. 10 min, p = 0.23). There were no severe complications following EDSL. CONCLUSION: The results of this study suggest that EDSL is superior to EC as treatment for colonic diverticular hemorrhage. EDSL may provide improvements in the clinical course of patients with colonic diverticular hemorrhage.

Multiple imputation for longitudinal data in the presence of heteroscedasticity between treatment groups.

Multiple imputation is a promising approach for handling of missing data. One uncertainty in applications of the multiple imputation to randomized controlled trials with longitudinal data is whether the imputation should be carried out across all subjects simultaneously or by treatment group separately, which leads to two different strategies for building imputation procedures and/or models. Indeed, it has not been sufficiently addressed and well-documented how the two imputation strategies work in the analysis of the longitudinal data. We consider situations in the presence of heteroscedasticity between treatment groups and conducted extensive simulation studies to examine how the choice of imputation strategy had impacts on the estimation of treatment effects under an assumption of missing at random mechanism. The choice of analysis model was also assessed. The simulation studies suggested that in the presence of heteroscedasticity, the separate imputation by treatment group was robust enough to provide unbiased and precise estimation of the treatment effects; in contrast, the simultaneous imputation, which is frequently used in applications, led to serious biases and poor coverage probabilities of 95% confidence interval for the treatment effects. The heteroscedasticity should be dealt with in more careful manners for the longitudinal data analysis, and if it could be the case in hand, we recommend using the separate imputation by treatment group, as well as applying unequal variance analysis methods for complete data with imputed values. The methods were illustrated with data from two real examples of pediatric research and mental health research.

Curved single-plane clear corneal incision made under anterior chamber collapse by aqueous humor aspiration prevents leakage after cataract surgery.

PURPOSE: To evaluate the self-sealing on curved single-plane clear corneal incisions (CCIs) after cataract surgery. The collapse of the anterior chamber was made by aspirating the aqueous humor in order to make curved single-plane CCI. METHODS: A retrospective study was conducted at Asai Hospital in Japan. There were 189 eyes of 132 patients scheduled from February to April 2017 for single-plane CCI. There were 175 eyes of 128 patients from February to April 2018 for curved single-plane CCI. Curved single-plane CCIs were made with aqueous humor aspiration via aqueous humor pipette, and single-plane CCIs were made without the aspiration. In case of leakage of aqueous humor from the CCI at the end of surgery, sutures were used. The main outcome measure was the proportion of sutures for CCI. Secondary outcomes included CCI length and surgically induced astigmatism (SIA) magnitude. RESULTS: The proportions of suture in single-plane CCI and in curved single-plane CCIs were 25.4% (48/189) and 0% (0/175), respectively (p < 0.001). The length of curved single-plane CCI (2.06 ± 0.18 mm) was significantly greater than that of single-plane CCI (1.66 ± 0.12 mm; p < 0.001). The median postoperative SIAs of single-plane and curved single-plane CCIs were 0.64 D (n = 89) and 0.73 D (n = 85) for nasal incision, and 0.46 D (n = 100) and 0.47 D (n = 90) for temporal incision at examination 3 months after surgery, respectively, without significant change (p = 0.12 for nasal incision and p = 0.37 for temporal incision). CONCLUSION: The curved single-plane CCI achieved by only aqueous humor aspiration showed significant effectiveness in self-sealing. TRIAL REGISTRATION: UMIN clinical trials registry UMIN000032480, retrospectively registered on 6 May 2018.