Artificial Intelligence-Enabled Electrocardiography Predicts Future Pacemaker Implantation and Adverse Cardiovascular Events.

Medical advances prolonging life have led to more permanent pacemaker implants. When pacemaker implantation (PMI) is commonly caused by sick sinus syndrome or conduction disorders, predicting PMI is challenging, as patients often experience related symptoms. This study was designed to create a deep learning model (DLM) for predicting future PMI from ECG data and assess its ability to predict future cardiovascular events. In this study, a DLM was trained on a dataset of 158,471 ECGs from 42,903 academic medical center patients, with additional validation involving 25,640 medical center patients and 26,538 community hospital patients. Primary analysis focused on predicting PMI within 90 days, while all-cause mortality, cardiovascular disease (CVD) mortality, and the development of various cardiovascular conditions were addressed with secondary analysis. The study's raw ECG DLM achieved area under the curve (AUC) values of 0.870, 0.878, and 0.883 for PMI prediction within 30, 60, and 90 days, respectively, along with sensitivities exceeding 82.0% and specificities over 81.9% in the internal validation. Significant ECG features included the PR interval, corrected QT interval, heart rate, QRS duration, P-wave axis, T-wave axis, and QRS complex axis. The AI-predicted PMI group had higher risks of PMI after 90 days (hazard ratio [HR]: 7.49, 95% CI: 5.40-10.39), all-cause mortality (HR: 1.91, 95% CI: 1.74-2.10), CVD mortality (HR: 3.53, 95% CI: 2.73-4.57), and new-onset adverse cardiovascular events. External validation confirmed the model's accuracy. Through ECG analyses, our AI DLM can alert clinicians and patients to the possibility of future PMI and related mortality and cardiovascular risks, aiding in timely patient intervention.

Changes of antithrombotic prescription in atrial fibrillation patients with acute coronary syndrome or percutaneous coronary intervention and the subsequent impact on long-term outcomes: a longitudinal cohort study.

OBJECTIVES: The choice of optimal antithrombotic therapy in atrial fibrillation (AF) patients with acute coronary syndrome (ACS) or percutaneous coronary intervention (PCI) remains controversial. The aim of this longitudinal cohort study is to investigate the prescribing pattern of antithrombotic regimen in different cohorts and its subsequent impact. SETTING AND DESIGN: Longitudinal data from the Tri-Service General Hospital-Coronary Heart Disease (TSGH-CHD) registry, between January 2016 and August 2018 was screened. PARTICIPANTS AND METHOD: Patients with prior history of nonvalvular AF, who had ACS presentation or underwent PCI were selected, and these patients were divided into cohort 1 and cohort 2, according to the index date of antithrombotic prescription before and after the PIONEER AF-PCI study. PRIMARY AND SECONDARY OUTCOMES: The primary safety endpoints were composites of major bleeding and/or clinically relevant non-major bleeding. The secondary efficacy endpoints included the occurrence of all-cause mortality, stroke/systemic embolization, nonfatal myocardial infarction (MI), and >30-days coronary revascularization. RESULTS: A total of 121 patients were included into analysis (cohort 1=35; cohort 2=86). Comparing with cohort 1, the prescription rate of triple antithrombotic therapy (TAT) increased from 17.1 to 38.4%, especially the regimen with dual antiplatelet therapy (DAPT) plus low-dose non-vitamin-K dependent oral anticoagulation (NOAC). However, the prescription rate of dual antithrombotic therapy (DAT) decreased (14.3-10.5%), as well as the prescription rate of DAPT (68.6-51.2%). These changes of antithrombotic prescription across different cohorts were not associated with risk of adverse safety (HR= 0.87; 95% CI, 0.42-1.80, p=0.710) and efficacy outcomes (HR=0.96; 95% CI, 0.40-2.32, p=0.930). CONCLUSIONS: Entering the NOAC era, the prescription of TAT increased alongside the decrease in DAT. As the prescription rate of DAPT without anticoagulation remained high, future efforts are mandatory to improve the implementation of guidelines and clinical practice.

Factors Associated With Rates of HBsAg Seroclearance in Adults With Chronic HBV Infection: A Systematic Review and Meta-analysis.

BACKGROUND & AIMS: Seroclearance of hepatitis B surface antigen (HBsAg) is a marker for clearance of chronic hepatitis B virus (HBV) infection, but reported annual incidence rates of HBsAg seroclearance vary. We performed a systematic review and meta-analysis to provide more precise estimates of HBsAg seroclearance rates among subgroups and populations. METHODS: We searched PubMed, Embase, and the Cochrane library for cohort studies that reported HBsAg seroclearance in adults with chronic HBV infection with more than 1 year of follow-up and at least 1 repeat test for HBsAg. Annual and 5-, 10-, and 15-year cumulative incidence rates were pooled using a random effects model. RESULTS: We analyzed 34 published studies (with 42,588 patients, 303,754 person-years of follow-up, and 3194 HBsAg seroclearance events), including additional and updated aggregated data from 19 studies. The pooled annual rate of HBsAg seroclearance was 1.02% (95% CI, 0.79-1.27). Cumulative incidence rates were 4.03% at 5 years (95% CI, 2.49-5.93), 8.16% at 10 years (95% CI, 5.24-11.72), and 17.99% at 15 years (95% CI, 6.18-23.24). There were no significant differences between the sexes. A higher proportion of patients who tested negative for HBeAg at baseline had seroclearance (1.33%; 95% CI, 0.76-2.05) than those who tested positive for HBeAg (0.40%; 95% CI, 0.25-0.59) (P < .01). Having HBsAg seroclearance was also associated with a lower baseline HBV DNA level (6.61 log10 IU/mL; 95% CI, 5.94-7.27) vs not having HBsAg seroclearance (7.71 log10 IU/mL; 95% CI, 7.41-8.02) (P < .01) and with a lower level of HBsAg at baseline (2.74 log10 IU/mL; 95% CI, 1.88-3.60) vs not having HBsAg seroclearance (3.90 log10 IU/mL, 95% CI, 3.73-4.06) (P < .01). HBsAg seroclearance was not associated with HBV genotype or treatment history. Heterogeneity was substantial across the studies (I2 = 97.49%). CONCLUSION: In a systematic review and meta-analysis, we found a low rate of HBsAg seroclearance in untreated and treated patients (pooled annual rate, approximately 1%). Seroclearance occurred mainly in patients with less active disease. Patients with chronic HBV infection should therefore be counseled on the need for lifelong treatment, and curative therapies are needed.

Systematic review and network meta-analysis: Comparative effectiveness of therapies for second-line Helicobacter pylori eradication.

BACKGROUND AND AIM: The eradication rate of Helicobacter pylori (H. pylori) has been declining over the past decades. A rescue plan is needed for increasing populations with treatment failure. However, the optimum second-line eradication regimen remains inconclusive. We conducted a network meta-analysis to assess the comparative effectiveness of second-line H. pylori eradication therapies and determine the optimum regimen. METHODS: We searched electronic databases from January 2005 to February 2018 for randomized controlled trials assessing the effectiveness of second-line regimens in patients with persistent H. pylori infection after first-line treatment. Bayesian network meta-analysis was performed to combine the direct and indirect evidence and to investigate the rank order of second-line therapies. We also appraised the quality of evidence using Grading of Recommendations Assessment, Development, and Evaluation guidance. RESULTS: Twenty-six trials with 3628 participants who received second-line eradication therapy were identified. All regimens showed pooled eradication rates < 90%. Compared with 7-day triple therapy, quinolone-based (odds ratio [OR] 4.29, 95% credible interval [CrI] 1.67-12.12, surface under the cumulative ranking [SUCRA] 0.95), non-quinolone-based bismuth-containing quadruple therapies for 10 days or more (OR 2.25, 95% CrI 1.10-4.62, SUCRA 0.78), and sequential therapy (OR 2.91, 95% CrI 1.16-7.65, SUCRA 0.66) showed significantly higher effectiveness. Overall, regimens with longer duration demonstrated higher eradication rates but higher rates of adverse events. More adverse events were reported in those patients treated with concomitant therapy. CONCLUSIONS: Quinolone-based bismuth-containing quadruple therapies for 10 days or more are the optimum second-line regimens for H. pylori eradication.

The net clinical benefits of febuxostat versus allopurinol in patients with gout or asymptomatic hyperuricemia - A systematic review and meta-analysis.

BACKGROUND AND AIMS: Systemic reviews and meta-analyses suggest hyperuricemia is a cardiovascular risk factor. The effects of xanthine oxidase inhibitors on cardiac outcomes remain unclear. We assessed the effects of febuxostat and allopurinol on mortality and adverse reactions in adult patients with hyperuricemia. METHODS AND RESULTS: PubMed and EMBASE were searched to retrieve randomized controlled trials of febuxostat and allopurinol from January 2005 to July 2018. The meta-analysis consisted of 13 randomized controlled trials with a combined sample size of 13,539 patients. Febuxostat vs. allopurinol was not associated with an increased risk of cardiac-related mortality in the overall population (OR: 0.72, 95% CI: 0.24-2.13, P = 0.55). Regarding adverse skin reactions, the patients receiving febuxostat had significantly fewer adverse skin reactions than those receiving allopurinol treatment (OR: 0.50, 95% CI: 0.30-085, P = 0.01). Compared with allopurinol, febuxostat was associated with an improved safety outcome of cardiac-related mortality and adverse skin reactions (OR: 0.72, 95% CI: 0.55-0.96, P = 0.02). The net clinical outcome, composite of incident gout and the safety outcome, was not different significantly in the patients receiving febuxostat or allopurinol (OR: 1.04, 95% CI: 0.76-0.1.42, P = 0.79). In sensitivity analyses, a borderline significance was found in the patients randomized to febuxostat vs. allopurinol regarding cardiac-related mortality (OR: 1.29, 95% CI: 1.00-1.67, P = 0.05) after the CARES study was included. CONCLUSION: Febuxostat vs. allopurinol was associated with the improved safety outcome and have comparable mortality and net clinical outcome in patients with hyperuricemia. REGISTRATION NUMBER: PROSPERO(CRD42018091657).

AI-enabled electrocardiography alert intervention and all-cause mortality: a pragmatic randomized clinical trial.

The early identification of vulnerable patients has the potential to improve outcomes but poses a substantial challenge in clinical practice. This study evaluated the ability of an artificial intelligence (AI)-enabled electrocardiogram (ECG) to identify hospitalized patients with a high risk of mortality in a multisite randomized controlled trial involving 39 physicians and 15,965 patients. The AI-ECG alert intervention included an AI report and warning messages delivered to the physicians, flagging patients predicted to be at high risk of mortality. The trial met its primary outcome, finding that implementation of the AI-ECG alert was associated with a significant reduction in all-cause mortality within 90 days: 3.6% patients in the intervention group died within 90 days, compared to 4.3% in the control group (4.3%) (hazard ratio (HR) = 0.83, 95% confidence interval (CI) = 0.70-0.99). A prespecified analysis showed that reduction in all-cause mortality associated with the AI-ECG alert was observed primarily in patients with high-risk ECGs (HR = 0.69, 95% CI = 0.53-0.90). In analyses of secondary outcomes, patients in the intervention group with high-risk ECGs received increased levels of intensive care compared to the control group; for the high-risk ECG group of patients, implementation of the AI-ECG alert was associated with a significant reduction in the risk of cardiac death (0.2% in the intervention arm versus 2.4% in the control arm, HR = 0.07, 95% CI = 0.01-0.56). While the precise means by which implementation of the AI-ECG alert led to decreased mortality are to be fully elucidated, these results indicate that such implementation assists in the detection of high-risk patients, prompting timely clinical care and reducing mortality. ClinicalTrials.gov registration: NCT05118035 .

A deep learning-based system capable of detecting pneumothorax via electrocardiogram.

PURPOSE: To determine if an electrocardiogram-based artificial intelligence system can identify pneumothorax prior to radiological examination. METHODS: This is a single-center, retrospective, electrocardiogram-based artificial intelligence (AI) system study that included 107 ECGs from 98 pneumothorax patients. Seven patients received needle decompression due to tension pneumothorax, and the others received thoracostomy due to instability (respiratory rate ≥ 24 breaths/min; heart rate, < 60 beats/min or > 120 beats/min; hypotension; room air O2 saturation, < 90%; and patient could not speak in whole sentences between breaths). Traumatic pneumothorax and bilateral pneumothorax were excluded. The ECGs of 132,127 patients presenting to the emergency department without pneumothorax were used as the control group. The development cohort included approximately 80% of the ECGs for training the deep learning model (DLM), and the other 20% of ECGs were used to validate the performance. A human-machine competition involving three physicians was conducted to assess the model performance. RESULTS: The areas under the receiver operating characteristic (ROC) curves (AUCs) of the DLM in the validation cohort and competition set were 0.947 and 0.957, respectively. The sensitivity and specificity of our DLM were 94.7% and 88.1% in the validation cohort, respectively, which were significantly higher than those of all physicians. Our DLM could also recognize the location of pneumothorax with 100% accuracy. Lead-specific analysis showed that lead I ECG made a major contribution, achieving an AUC of 0.930 (94.7% sensitivity, 86.0% specificity). The inclusion of the patient characteristics allowed our AI system to achieve an AUC of 0.994. CONCLUSION: The present AI system may assist the medical system in the early identification of pneumothorax through 12-lead ECG, and it performs as well with lead I ECG alone as with 12-lead ECG.

Hyperuricemia Is Associated With a Higher Prevalence of Metabolic Syndrome in Military Individuals.

Introduction: Hyperuricemia (HUA) is associated with metabolic syndrome (MetS) in the general population. Military individuals who perform high-intensity physical training might have lower rates of MetS. The present study aimed to investigate whether HUA might be associated with the prevalence of MetS in military individuals. Material and Methods: We retrospectively collected data from the annual military exam and randomly selected a single unit to represent the overall study population. The study population consisted of 460 military individuals between January 2016 and December 2016. We divided this cohort into the HUA group and the normouricemic group. Hyperuricemia is defined as a serum uric acid level of 7 mg/dL or more in men or 6 mg/dL or more in women. Results: The cohort consisted of 460 individuals with a mean age of 35.9 yr old; 80% were male and 15% were diagnosed with MetS between January 1, 2016 and December 31, 2016. The prevalence of MetS was greater in the HUA group than in the normouricemic group (32.5% vs. 8.8%, p < 0.001). HUA was independently associated with the prevalent MetS after adjusting for age, gender, creatinine, alanine transaminase, and hemoglobin (adjusted OR: 4.305, 95% CI: 2.370-7.818, p < 0.001). Given that the cohort was predominantly male, we divided the cohort into men and women for a subgroup analysis. A significant association was found in men but not in women (adjusted OR: 3.59 95% CI: 1.905-6.765, p < 0.001 for men and adjusted OR: 16.7 95% CI: 0.295-946, p = 0.172 for women, respectively). Conclusion: Hyperuricemia was independently associated with the prevalence of metabolic syndrome in a military cohort from Taiwan. Future studies should look at whether hyperuricemia in individuals without metabolic syndrome can predict the future onset of metabolic syndrome.