INTRODUCTION/OBJECTIVES:We aimed to investigate associations between menopausal characteristics, including age at menopause and cause (natural vs. artificial), with long-term outcomes in women with rheumatoid arthritis (RA), incident Alzheimer's disease and related dementias (ADRD), RA severity, and overall mortality. We hypothesized that atypical menopausal characteristics increase ADRD risk and disease severity. METHODS:Women with incident RA in 1980-2014 who were ≥50 years of age at RA incidence were included. Menopause cause and age at menopause (categorized as <45, 45-54, ≥55 years) were abstracted from medical records. Associations between menopausal characteristics and ADRD, extra-articular manifestations (ExRA), erosions, cardiovascular events, and mortality were evaluated using Cox models. Associations with clinic visits for flares and remissions were assessed using mixed-effects models. Data were collected using the Rochester Epidemiology Project medical records-linkage system. RESULTS:Both early/premature and late menopause were associated with nonsignificant near two-fold increases in ADRD risk. In addition, early/premature (HR 2.40; 95% CI 1.15-5.01) and late menopause (HR 2.17; 95% CI 1.01-4.66) significantly increased severe ExRA risk. Menopausal characteristics were not associated with presence of erosions or mortality. Artificial menopause was associated with fewer cardiovascular events and fewer visits for RA remission. Early/premature menopause was associated with fewer remission visits (OR 0.50; 95% CI 0.29-0.87). CONCLUSION:Women with early/premature and late menopause had nonsignificant, increased risk of incident ADRD. Menopausal characteristics had significant associations with RA severity and cardiovascular events, and may be relevant contextual factors in long-term risk assessment for women with RA.

ABSTRACT Ischemic events after contemporary percutaneous coronary intervention (PCI) are uncommon but carry high morbidity/mortality. Identifying patients at highest risk is critical to guide dual antiplatelet therapy (DAPT) intensity while minimizing bleeding. Current machine learning (ML) model‐developed risk scores do not incorporate pharmacogenetic data. To develop and externally validate ML models that integrate clinical, demographic, and CYP2C19 genetic information to predict 1‐year ischemic outcomes post‐PCI to refine clinical decision making. We analyzed 8317 patients from TAILOR‐PCI trial (n = 4572) and the Precision PCI registry (n = 3745). The outcome was a composite of cardiovascular death, myocardial infarction, stroke, and stent thrombosis at 1 year. Boruta feature selection identified 11 predictors. Multiple ML algorithms were trained in TAILOR‐PCI and validated in Precision PCI using cross‐validation and synthetic minority oversampling (SMOTE). Model performance was assessed by area under the receiver operating characteristic curve (AUC), sensitivity, and specificity. The best external performance was achieved with a support vector machine (SVM, polynomial kernel) (AUC 0.667; sensitivity 0.871; specificity 0.282), while XGBoost provided a more balanced profile (AUC 0.619; sensitivity 0.442; specificity 0.688). Variable importance for the SVM polynomial model demonstrated that all 11 included Boruta feature selected predictors had relatively high importance (> 75). ML models trained on large clinical trial and real‐world registry datasets can help identify the small subset of patients at high ischemic risk after PCI. This distinction is of relevance because ischemic events are rare, and most patients may safely de‐escalate DAPT to reduce bleeding risk while maintaining ischemic protection through a multimodal approach to risk stratification. Trial Registration: TAILOR‐PCI URL: https://clinicaltrials.gov/ct2/show/NCT01742117

Artificial intelligence (AI)-based screening tools show promise for early identification of chronic liver disease (CLD), yet their effectiveness in real-world settings may depend on clinician response to AI-generated recommendations. We performed a post hoc analysis of the intervention arm of the pragmatic, cluster-randomized DULCE trial, in which primary care clinicians received electrocardiogram-based machine learning (ECG-ML) alerts indicating elevated risk for CLD. Clinicians were categorized as high engagement (HE; top quartile) or low engagement (LE), and diagnostic yield was defined as the proportion of ECG-ML-positive cases with confirmed CLD. Among 110 clinicians receiving ≥1 alert (1385 ECG-ML-positive patients), overall engagement was 29.8%. HE was associated with higher detection of advanced CLD (OR 2.12, 95% CI 1.36-3.30; p = 0.001) and any CLD (OR 2.59, 95% CI 1.83-3.68; p < 0.001) compared with LE. Diagnostic yield was 10.6% versus 2.9% for advanced CLD and 22.3% versus 5.0% for any CLD in HE versus LE (OR 2.99, 95% CI 1.73-5.16; p < 0.001 and OR 3.74, 95% CI 2.44-5.75; p < 0.001, respectively). These findings suggest that the effectiveness of AI-based screening may depend not only on algorithm performance but also on clinician engagement with AI recommendations and highlight the importance of accounting for engagement when designing and interpreting AI-enabled clinical trials. ClinicalTrials.gov NCT05782283.

BACKGROUND:In patients with cirrhosis, prolonged international normalized ratio (INR) is primarily driven by liver synthetic dysfunction. Thromboelastography more accurately reflects coagulability and can also predict short-term mortality. Nevertheless, INR is used to define coagulation failure in the European Foundation for the Study of Chronic Liver Failure criteria for acute-on-chronic liver failure (ACLF). AIM:To evaluate the association between TEG parameters and mortality in critically ill patients with cirrhosis and to justify future investigation into TEG as a prognostic tool in ACLF. METHODS:We performed a retrospective study of 52 patients with cirrhosis admitted to the intensive care unit (ICU) who had TEG performed during their ICU admission prior to the administration of any blood products. We assessed the association between TEG parameters and 28-day mortality. RESULTS:Patients who did not survive beyond 28 days generally had more hypocoagulable TEG parameters (R time: 10.0 vs. 7.6, p = 0.03; K time: 3.4 vs. 2.2, p = 0.003; alpha angle: 53.1 vs. 63.5, p = 0.002; MA: 49.2 vs. 60.2, p = 0.001). Although global assessment of TEG demonstrated a state of rebalanced hemostasis among critically ill patients with cirrhosis, patients meeting diagnostic criteria for ACLF tended to have more hypocoagulable TEG parameters as compared to those who did not (coagulation index: 0.2 vs. 2.0, p = 0.004). As proof of concept, MA was incorporated into the definition of ACLF to replace INR as a marker of coagulation failure (TEG-ACLF). The area under the curve (AUC) for TEG-ACLF was 0.83 (0.72-0.95) compared to 0.8 (0.68-0.92) for the current ACLF-CLIF grading, with 9.6% (5/52) of patients being reclassified into a different ACLF grade. CONCLUSION:Hypocoagulable TEG parameters are associated with increased 28-day mortality in critically ill patients with cirrhosis, and incorporation of TEG parameters into a modified definition for ACLF may result in improved prediction of short-term mortality.

Background:Cirrhosis is a leading cause of morbidity and mortality worldwide, yet preventable at early stages. Currently, effective approaches for early diagnosis are lacking. A novel electrocardiogram (ECG)-enabled deep learning model trained for detection of advanced chronic liver disease (CLD) has demonstrated promising results and it may be used for screening of advanced CLD in primary care. Design:A pragmatic, cluster randomized trial (NCT05782283) in 45 Mayo Clinic primary care practices will be conducted over a period of 6 months with 6 months of follow up. Care teams will be randomized 1:1 to intervention or usual care, stratified by region and patient volume. Patients from providers enrolled in the trial who undergo an ECG during the study period will be included. In the intervention arm, consenting providers to patients identified as higher risk of advanced CLD based on their ECG will be notified with a recommendation for noninvasive fibrosis assessment. The primary endpoint will be detection of advanced CLD (defined as stage 3-4 on blood- or imaging-based noninvasive liver disease assessment or liver biopsy). Secondary outcomes will include completion of fibrosis assessment tests within 180 days of ECG, new diagnosis of liver disease stratified by etiology and risk factors for CLD, and detection of any liver fibrosis (stages 1-4). Post-study surveys to participating clinicians will be conducted. Summary:Preliminary findings suggest outstanding potential for the use of an ECG-enabled machine learning algorithm for detection of advanced CLD in the primary care community.

BACKGROUND AND AIMS:Patients with inflammatory bowel disease (IBD) face increased risk of colorectal cancer (CRC). While the natural history of conventional dysplastic precursor lesions has been well-studied, the neoplastic potential of recently described nonconventional (NC) IBD-associated colonic mucosal lesions is unclear. We aimed to assess the incidence of antecedent NC lesions in patients with IBD who developed CRC. METHODS:A case-cohort study was performed to include patients with a diagnosis of IBD with or without CRC who underwent at least 2 surveillance endoscopic procedures at our institution between 1/1/2007 and 5/31/2023. NC lesions included serrated change and indefinite for dysplasia. Detection rates pre- and post-introduction of high-definition (HD) surveillance colonoscopy were compared. RESULTS:In total, 87 patients with IBD and CRC and 200 patients with IBD without CRC were identified. Of the cases, a majority had ulcerative colitis (n = 52, 60%), most commonly with extensive involvement (n = 46, 89%). Conventional (hazard ratio [HR] 2.18, 95% confidence interval [CI] 1.34-3.52) and NC (HR 2.28, 95% CI 1.59-3.26) lesions were associated with increased risk of CRC. Conventional lesions in the post-HD era appeared to have a stronger association with CRC (HR 2.79, 95% CI 1.62-4.77) than NC lesions (HR 1.62, 95% CI 0.86-3.06). CONCLUSIONS:Both conventional and NC lesions seem to be associated with increased risk of CRC. Conventional lesions are more strongly associated with CRC than NC lesions in the post-HD era, but misclassifications in the pre-HD era may have resulted in a biased increased risk estimate for NC lesions.

Advanced chronic liver disease (CLD) affects 2-5% of the general population, and accessible screening tools are needed in primary care. Here we conducted a pragmatic trial to assess whether an electrocardiogram (ECG)-based machine learning (ECG-ML) model enables early detection of advanced CLD. In this trial, 98 primary care teams were cluster randomized to intervention (access to ECG-ML results; 123 clinicians) or usual care (122 clinicians). Clinicians in the intervention arm were notified of a positive ECG-ML result, indicating higher risk of advanced CLD. The primary endpoint was new diagnosis of CLD with advanced fibrosis within 180 days of ECG, confirmed by sequential liver disease assessments. A total of 15,596 adults underwent 12-lead ECGs as part of routine care and met inclusion criteria (N = 8,034 intervention and N = 7,562 control). The intervention significantly increased new diagnoses of advanced CLD in the overall cohort (1.0% versus 0.5% in the control arm; odds ratio (OR) 2.09, 95% confidence interval (CI) 1.22-3.55, P = 0.007). Among ECG-ML-positive patients, advanced CLD was more frequent in the intervention arm (4.4% versus 1.1%; OR 4.37, 95% CI 1.94-9.88, P < 0.001). The intervention also increased the detection of any fibrosis (secondary endpoint) in the overall cohort (1.7% versus 0.5%; OR 3.17, 95% CI 1.86-5.40, P < 0.001) and among ECG-ML-positive patients (8.4% versus 1.1%; OR 8.03, 95% CI 3.50-18.4, P < 0.001). The diagnostic yield below epidemiological estimates probably reflects variable clinician adherence to artificial intelligence-driven recommendations. These results demonstrate that an ECG-based machine learning model, followed by targeted testing based on risk factors, may aid case finding of advanced CLD in routine primary care. ClinicalTrials.gov registration: NCT05782283 .

BACKGROUND:Functional lumen imaging probe (FLIP) utility is established in treatment-naïve achalasia but less clear following lower esophageal sphincter (LES) directed therapy. METHODS:Achalasia patients with LES directed therapy across three tertiary care centers between 2017 and 2024 with post-treatment FLIP and timed barium esophagram (TBE) were retrospectively identified. Reduced esophagogastric junction (EGJ) opening was defined by distensibility index (DI) < 2 mm2/mmHg and diameter < 12 mm. Abnormal emptying on TBE was defined as column height ≥ 5 cm at 5 min and/or retained tablet. Eckardt scores ≤ 3 defined clinical response. KEY RESULTS:The study included 222 patients (46% peroral endoscopic myotomy, 46% laparoscopic Heller myotomy, 8% pneumatic dilation) with a median of 1.4 years to post-treatment TBE/FLIP. Abnormal emptying on TBE was associated with a narrower median EGJ diameter (13.2 vs. 14.8 mm, p = 0.008), a greater frequency of EGJ diameter < 12 mm (36% vs. 21%, p = 0.012), and a smaller change in EGJ diameter (+4.6 vs. +8.6 mm, p = 0.002). Abnormal emptying on TBE occurred more frequently in patients with EGJ DI < 2 mm2/mmHg (8.5% vs. 2.6%, p = 0.052), but was not associated with median EGJ DI (4.5 vs. 5.1 mm2/mmHg, p = 0.29) nor median change in EGJ DI (+2.9 vs. +3.9 mm2/mmHg, p = 0.25). Patients with reduced EGJ DI or EGJ diameter more often had abnormal TBE (37% vs. 22%, p = 0.012). Only the change in DI (+3.8 vs. +1.5 mm2/mmHg, p = 0.012) and diameter (+8.2 vs. +1.6 mm, p = 0.002) on FLIP was associated with a clinical response based on Eckardt ≤ 3. CONCLUSIONS AND INFERENCES:FLIP following achalasia therapy generally correlates with TBE, although discrepant findings are not uncommon. In particular, FLIP EGJ-diameter has a strong association with esophageal emptying on TBE. Both TBE and FLIP have limited association with clinical response based on Eckardt, with change in DI and diameter on FLIP most strongly associated. Consequently, FLIP as part of multimodal assessment appears useful in the longitudinal follow-up of treated achalasia.

BACKGROUND/OBJECTIVE:We aimed to examine the incidence of sleep disorders (SD) in individuals with rheumatoid arthritis (RA) vs. non-RA comparators, evaluate risk factors for SD, and assess the association between incident SD and dementia in RA. METHODS:This retrospective cohort study included residents aged ≥50 years within an 8-county region of Minnesota who first met the 1987 ACR criteria for RA in 1980-2014. Individuals with RA were matched 1:1 with non-RA individuals on age, sex, and calendar year of RA incidence. Data on SD, cardiovascular disease (CVD) risk factors, CVD and other comorbidities were collected from the medical records. RESULTS:Nine hundred thirteen individuals with RA and 913 non-RA comparators were included (mean age: 65 years, 65 % female in both cohorts). During the median follow-up of 10.4 years in RA and 11.0 years in non-RA cohort, SD developed in 234 and 206 individuals, respectively. RA patients experienced an increased risk for any incident SD (HR 1.34; 95 % CI:1.11-1.61) and insomnia (HR 1.34; 95 % CI:1.03-1.73). Obesity, dyslipidemia, presence of CVD, depression, anxiety, and more recent calendar year of RA incidence were associated with increased risk of any SD in RA. There were no significant association between SD overall and by subtype with dementia in RA. CONCLUSION:Individuals with RA (vs non-RA) experienced a significantly increased risk for any SD, particularly insomnia. CVD and CVD risk factors, as well as depression and anxiety increased the risk for incident SD in RA. There was no significant association between SD and dementia in RA.

Introduction: Identifying patients at risk for ischemic events after percutaneous coronary intervention (PCI) relies on traditional analysis of limited clinical and imaging variables. Machine learning (ML) has shown promise in effectively predicting cardiovascular risk in population studies. While existing ML models mainly predict mortality and incorporate clinical variables, there is a lack of tools that have utilized genetic data and that predict ischemic events. Aims: This study aims to develop and validate a ML model incorporating genotyping and clinical data to enhance prediction of ischemic outcomes for PCI patients utilizing large prospectively derived diverse datasets. Methods: Patients from the TAILOR-PCI trial (n=5302) were utilized for model development. 50% of the sample was utilized for Boruta feature selection and 50% for training and testing using cross validation. Features included demographics, medical history, medications, PCI characteristics, and genetic data (specifically, CYP2C19 *2, *3, *17 alleles). The primary endpoint was a composite of cardiovascular death, myocardial infarction, stroke, stent thrombosis, and severe recurrent ischemia at 12 months. Multiple ML classification algorithms, including Support Vector Machine (SVM) polynomial, Random Forest, Light gradient boost, XG Boost, among others, were benchmarked for their prediction performance on rare events. The top performing classifiers were externally validated on an independent dataset from the PRECISION PCI study (n=3,745). Results: Mean participant age of the training set was 64.2 ± 11.0 years, with 75.4% being male. During follow-up of 12 months, among 4,572 patients in the entire cohort 343 (7.5%) met the primary outcome. The SVM polynomial model demonstrated the highest area under the curve (AUC) of 0.67 for predicting the primary outcome with test dataset. The sensitivity, specificity, precision, and recall were 0.87, 0.28, 0.07, and 0.87 respectively (Figure). Peripheral arterial disease, body mass index, and age were among the top variables by feature importance. Conclusion: ML models incorporating both clinical and genetic data are feasible and highly promising in predicting major adverse cardiac events that may help guide use of anti-platelet drug therapy. The AUC values are reasonable given imbalances and misclassifications in datasets, and further model optimization with prospective utilization of the model will be paramount.

CYP2C19 loss of function (LOF) carriers undergoing percutaneous coronary intervention (PCI) have an increased risk of ischemic events when treated with clopidogrel. PCI patients in TAILOR-PCI were randomized to clopidogrel or genotype-guided (GG) therapy in which LOF carriers received ticagrelor and non-carriers clopidogrel. Direct medical costs associated with a GG approach have not been described before. TAILOR-PCI participants for whom direct medical costs were available for the duration from the date of PCI to one-year post PCI were included. Primary cost estimates were obtained from the Mayo Clinic Cost Data Warehouse. There were no differences in direct medical costs between the GG and clopidogrel groups (mean $20,682 versus $19,747, p = 0.11) however total costs were greater in the GG group (mean $21,245 versus $19,891, p = 0.02) which was primarily driven by ticagrelor costs. In conclusion the increased expense of a GG strategy post PCI as compared to clopidogrel for all is primarily driven by the cost of ticagrelor.

BACKGROUND AND AIMS:Irritable bowel syndrome (IBS) is a pain disorder classified by bowel habits, disregarding other factors that may influence the clinical course. The aim of this study was to determine if IBS patients can be clustered based on clinical, dietary, lifestyle, and psychosocial factors. METHODS:Between 2013 and 2020, the Mayo Clinic Biobank surveyed and received 40,291 responses to a questionnaire incorporating Rome III criteria. Factors associated with IBS were determined and latent class analysis, a model-based clustering, was performed on IBS cases. RESULTS:We identified 4021 IBS patients (mean 64 years; 75% women) and 12,063 controls. Using 26 variables separating cases from controls, the optimal clustering revealed 7 latent clusters. These were characterized by perceived health impairment (moderate or severe), psychoneurological factors, and bowel dysfunction (diarrhea or constipation predominance). Health impairment clusters demonstrated more pain, with the severe cluster also having more psychiatric comorbidities. The next 3 clusters had unique enrichment of psychiatric, neurological, or both comorbidities. The bowel dysfunction clusters demonstrated less abdominal pain, with diarrhea cluster most likely to report pain improvement with defecation. The constipation cluster had the highest exercise score and consumption of fruits, vegetables, and alcohol. The distribution of clusters remained similar when Rome IV criteria were applied. Physiologic tests were available on a limited subset (6%), and there were no significant differences between clusters. CONCLUSIONS:In this cohort of older IBS patients, 7 distinct clusters were identified demonstrating varying degrees of gastrointestinal symptoms, comorbidities, dietary, and lifestyle factors. Further research is required to assess whether these unique clusters could be used to direct clinical trials and individualize patient management.

Background and study aims Chronically inflamed colonic mucosa is primed to develop dysplasia identified at surveillance colonoscopy by targeted or random biopsies. We aimed to explore the effect of mucosal inflammation on detection of visible and "invisible" dysplasia and the concordance between the degree of endoscopic and histologic inflammation. Patients and methods This was a 6-year cross-sectional analysis of endoscopic and histologic data from IBD. A multinomial model was created to estimate the odds for a specific lesion type as well as the odds of random dysplasia relative to the degree of inflammation. Kappa statistics were used to measure concordance between endoscopic and histologic inflammation. Results A total of 3437 IBD surveillance colonoscopies between 2016-2021 were reviewed with 970 procedures from 721 patients containing 1603 visible lesions. Kappa agreement between histologic and endoscopic degree of inflammation was low at 0.4. There was a positive association between increased endoscopic inflammation and presence of tubulovillous adenomas (TVAs) (odds ratio [OR] 2.18; 95% confidence interval [CI] 1.03-4.62; P =0.04). Among cases with visible lesions, the yield of concomitant random dysplasia was 2.7% and 1.9% for random indefinite dysplasia. The odds of random dysplasia significantly increased as the degree of endoscopic and histologic inflammation increased (OR 2.18, 95%CI 1.46-3.26; P <0.001 and OR 2.75; 95%CI 1.65-4.57, P <0.001, respectively. The odds of indefinite random dysplasia also significantly increased as endoscopic and histologic inflammation increased (OR 2.90; 95%CI 1.85, 4.55, P <0.001 and OR 1.98; 95%CI 1.08, 3.62, P <0.035, respectively. Conclusions Endoscopic and histologic inflammation are associated with higher odds of finding TVAs and random low-grade, high-grade, and indefinite dysplasia. Concordance between histologic and endoscopic inflammation severity is low.