Carter, Rickey E

Background: Identification of systolic heart failure among patients presenting to the emergency department (ED) with acute dyspnea is challenging. The reasons for dyspnea are often multifactorial. A focused physical evaluation and diagnostic testing can lack sensitivity and specificity. The objective of this study was to assess the accuracy of an artificial intelligence-enabled ECG to identify patients presenting with dyspnea who have left ventricular systolic dysfunction (LVSD). Methods: We retrospectively applied a validated artificial intelligence-enabled ECG algorithm for the identification of LVSD (defined as LV ejection fraction ≤35%) to a cohort of patients aged ≥18 years who were evaluated in the ED at a Mayo Clinic site with dyspnea. Patients were included if they had at least one standard 12-lead ECG acquired on the date of the ED visit and an echocardiogram performed within 30 days of presentation. Patients with prior LVSD were excluded. We assessed the model performance using area under the receiver operating characteristic curve, accuracy, sensitivity, and specificity. Results: A total of 1606 patients were included. Median time from ECG to echocardiogram was 1 day (Q1: 1, Q3: 2). The artificial intelligence-enabled ECG algorithm identified LVSD with an area under the receiver operating characteristic curve of 0.89 (95% CI, 0.86–0.91) and accuracy of 85.9%. Sensitivity, specificity, negative predictive value, and positive predictive value were 74%, 87%, 97%, and 40%, respectively. To identify an ejection fraction <50%, the area under the receiver operating characteristic curve, accuracy, sensitivity, and specificity were 0.85 (95% CI, 0.83–0.88), 86%, 63%, and 91%, respectively. NT-proBNP (N-terminal pro-B-type natriuretic peptide) alone at a cutoff of >800 identified LVSD with an area under the receiver operating characteristic curve of 0.80 (95% CI, 0.76–0.84). Conclusions: The ECG is an inexpensive, ubiquitous, painless test which can be quickly obtained in the ED. It effectively identifies LVSD in selected patients presenting to the ED with dyspnea when analyzed with artificial intelligence and outperforms NT-proBNP. Graphic Abstract: A graphic abstract is available for this article.

Background: Comorbidity-driven microvascular inflammation is posited as a unifying pathophysiologic mechanism for heart failure with preserved ejection fraction (HFpEF). Obesity is proinflammatory and common in HFpEF. We hypothesized that unique obesity-inflammation HFpEF phenotypes exist and are associated with differences in clinical features, fibrosis biomarkers, and functional performance. Methods: Patients (n=301) from 3 HFpEF clinical trials were studied. Unsupervised machine learning (hierarchical clustering) with obese status and 13 inflammatory biomarkers as input variables was performed. Associations of clusters with HFpEF severity and fibrosis biomarkers (PIIINP [procollagen III N-terminal peptide], CITP [C-telopeptide for type I collagen], IGFBP7 [insulin-like growth factor-binding protein-7], and GAL-3 [galectin-3]) were assessed. Results: Hierarchical clustering revealed 3 phenotypes: pan-inflammatory (n=129; 64% obese), noninflammatory (n=83; 55% obese), and obese high CRP (C-reactive protein; n=89; 98% obese). The pan-inflammatory phenotype had more comorbidities and heart failure hospitalizations; higher left atrial volume, NT-proBNP (N-terminal pro-B-type natriuretic peptide), and fibrosis biomarkers; and lower glomerular filtration rate, peak oxygen consumption, 6-minute walk distance, and active hours/day ( P <0.05 for all). The noninflammatory phenotype had the most favorable values for all measures. The obese high CRP phenotype resembled the noninflammatory phenotype except for isolated elevation of CRP and lower functional performance. Hierarchical cluster assignment was independent of CRP genotype combinations that alter CRP levels and more biologically plausible than other clustering approaches. Multiple traditional analytic techniques confirmed and extended the hierarchical clustering findings. Conclusions: Unique obesity-inflammation phenotypes exist in HFpEF and are associated with differences in comorbidity burden, HFpEF severity, and fibrosis. These data support comorbidity-driven microvascular inflammation as a pathophysiologic mechanism for many but not all HFpEF patients.

OBJECTIVE To distinguish the effects of hyperglycemia and hyperinsulinemia on exercise-induced increases in Rd and endogenous glucose production (EGP) in type 1 diabetes. RESEARCH DESIGN AND METHODS We studied six participants without diabetes and six participants with type 1 diabetes on three visits in random order for the following: euglycemia, low insulin (EuLoI); euglycemia, high insulin (EuHiI); and hyperglycemia, low insulin (HyLoI). Glucose fluxes were measured using [6,6-2H2] glucose before, during, and after 60 min of exercise. RESULTS Rd increased (P &lt; 0.01) with exercise within groups, while peak Rd during exercise was lower (P &lt; 0.01) in participants with type 1 diabetes than participants without diabetes during all visits. In type 1 diabetes participants, EGP increased (P &lt; 0.001) with exercise during EuLoI and HyLoI but not during EuHiI. This demonstrates that hyperinsulinemia, but not hyperglycemia, blunts the compensatory exercise-induced increase in EGP in type 1 diabetes. CONCLUSIONS The data from this pilot study indicate that 1) exercise-induced compensatory increase in EGP was inhibited in participants with type 1 diabetes with hyperinsulinemia but not with hyperglycemia; 2) in contrast, in participants without diabetes, exercise-induced increase in EGP was inhibited only during combined hyperinsulinemia and hyperglycemia. Taken together, these results suggest that low insulin coupled with euglycemia or modest hyperglycemia appear to be the most favorable milieu for type 1 diabetes during exercise.

Abstract Context The effect of physiological changes in night-time cortisol and glucagon on endogenous glucose production (EGP) and nocturnal glycemia are unknown. Objective To determine the effects of changes in cortisol and glucagon on EGP during the night. Design Two overnight protocols were conducted. In Protocol 1, endogenous cortisol was blocked with metyrapone and hydrocortisone infused either at constant (constant) or increasing (variable) rates to mimic basal or physiological nocturnal cortisol concentrations. In Protocol 2, endogenous glucagon was blocked with somatostatin and exogenous glucagon was infused at either basal or elevated rates to mimic nocturnal glucagon concentrations observed in nondiabetic (ND) and type 2 diabetes (T2D) individuals. EGP was measured using [3-3H] glucose and gluconeogenesis estimated with 2H2O in all studies. Setting Mayo Clinic Clinical Research Trials Unit, Rochester, MN, US. Participants In Protocol 1, 34 subjects (17 ND and 17 T2D) and in Protocol 2, 39 subjects (21 ND and 18 T2D) were studied. Main Outcome Measures Endogenous glucose production Results EGP, gluconeogenesis, and glycogenolysis were higher with variable than with constant cortisol at 7 am in T2D subjects. In contrast, nocturnal EGP did not differ in ND subjects between variable and constant cortisol. While elevated glucagon increased EGP, glycogenolysis, and gluconeogenesis in ND, the data in T2D subjects indicated that EGP and gluconeogenesis but not glycogenolysis were higher during the early part of the night. Conclusion Nocturnal hyperglucagonemia, but not physiological rise in cortisol, contributes to nocturnal hyperglycemia in T2D due to increased gluconeogenesis.

Background Artificial Intelligent Virtual Assistants (AIVA) is a segment of artificial intelligence that is rapidly developing. However, its utilization to address patients' frequently asked questions remains unexplored. Methods We developed an AIVA to answer questions related to 10 frequent topics asked by plastic surgery patients in our institution. Between July 27, 2018, and August 10 of 2018, we recruited subjects with administrative positions at our health care institution to chat with the virtual assistant. They asked, with their own words, 1 question for each topic and filled out a satisfaction questionnaire. Postsurvey analysis of questions and answers allowed assessment of the virtual assistant's accuracy. Results Thirty participants completed the survey. The majority was female (70%), and the mean age was 27.76 years (SD, 8.68 [19–51] years). The overall accuracy of the plastic surgery AIVA was 92.3% (277/294 questions), and participants considered the answer correct in 83.3% of the time (250/294 answers). Most of the participants considered the AIVA easy to use, answered adequately, and could be helpful for patients. However, when asked if this technology could replace a human assistant, they stayed neutral.

New Findings What is the central question of this study? Are sex difference in the central airways present in healthy paediatric patients? What is the main finding and its importance? In patients ≤12 years we found no sex differences in central airway luminal area. After 14 years, the males had significantly larger central airway luminal areas than the females. The sex differences were minimized, but preserved when correcting for height. Luminal area is the main determinant of airway resistance and our finding could help explain sex differences in pulmonary system limitations to exercise in paediatric patients. Abstract Cross‐sectional airway area is the main determinant of resistance to airflow in the respiratory system. In paediatric patients (<18 years), previous evidence for sex differences in cross‐sectional airway area was limited to patients with history of pulmonary disease or cadaveric studies with small numbers of subjects. These studies either only report tracheal data and do not include a range of ages or correct for height. Therefore, we sought to assess sex differences in airway luminal area utilizing paediatric patients of varying ages and no history of respiratory disease. Using three‐dimensional reconstructions from high‐resolution computed tomography scans, we retrospectively assessed the cross‐sectional airway area in healthy paediatric females ( n = 97) and males ( n = 128) over a range of ages (1–17 years). The areas of the trachea, left main bronchus, left upper lobe, left lower lobe, right main bronchus, intermediate bronchus and right upper lobe were measured at three discrete points by a blinded investigator. No differences between the sexes were noted in the cross‐sectional areas of the youngest (ages 1–12 years) patients ( P > 0.05). However, in patients ≥14 years the cross‐sectional areas were larger in the males compared to females in most airway sites. For instance, the cross‐sectional size of the trachea was 25% (218 ± 44 vs . 163 ± 24 mm 2 , P < 0.01) larger in males vs . females among ages 13–17 years. When accounting for height, these sex differences in airway areas were attenuated, but persisted. Our results indicate that sex differences in paediatric airway cross‐sectional area manifest after age ≥14 years and are independent of height.

Background. Currently, no surveillance guidelines for hepatocellular carcinoma (HCC) recurrence after liver transplantation (LT) exist. In this retrospective, multicenter study, we have investigated the role of surveillance imaging on postrecurrence outcomes. Methods. Patients with recurrent HCC after LT from 2002 to 2016 were reviewed from 3 transplant centers (University of California San Francisco, Mayo Clinic Florida, and University of Toronto). For this study, we proposed the term cumulative exposure to surveillance (CETS) as a way to define the cumulative sum of all the protected intervals that each surveillance test provides. In our analysis, CETS has been treated as a continuous variable in months. Results. Two hundred twenty-three patients from 3 centers had recurrent HCC post-LT. The median follow-up was 31.3 months, and median time to recurrence was 13.3 months. Increasing CETS was associated with improved postrecurrence survival (hazard ratio, 0.94; P < 0.01) as was treatment of recurrence with resection or ablation (hazard ratio, 0.31; P < 0.001). An receiver operating characteristic curve (area under the curve, 0.64) for CETS covariate showed that 252 days of coverage (or 3 surveillance scans) within the first 24 months provided the highest probability for aggressive postrecurrence treatment. Conclusions. In this review of 223 patients with post-LT HCC recurrence, we found that increasing CETS does lead to improved postrecurrence survival as well as a higher probability for aggressive recurrence treatment. We found that 252 days of monitoring (ie, 3 surveillance scans) in the first 24 months was associated with the ability to offer potentially curative treatment.

AbstractBackgroundThe accurate separation of undifferentiated wide complex tachycardias (WCTs) into ventricular tachycardia (VT) or supraventricular wide complex tachycardia (SWCT) using conventional, manually‐applied 12‐lead electrocardiogram (ECG) interpretation methods is difficult.PurposeWe sought to devise a new WCT differentiation method that operates solely on automated measurements routinely provided by computerized ECG interpretation software.MethodsIn a two‐part analysis, we developed and validated a logistic regression model (ie, VT Prediction Model) that utilizes routinely available computerized measurements derived from patients' paired WCT and baseline ECGs.ResultsThe derivation cohort consisted of 601 paired WCT (273 VT, 328 SWCT) and baseline ECGs from 421 patients. The VT Prediction Model, composed of WCT QRS duration (ms) (P < .0001), QRS duration change (ms) (P < .0001), QRS axis change (°) (P < .0001) and T axis change (°) (P < .0001), yielded effective VT and SWCT differentiation (area under the curve [AUC]: 0.924; confidence interval [CI]: 0.903‐0.944) for the derivation cohort. The validation cohort comprised 241 paired WCT (97 VT, 144 SWCT) and baseline ECGs from 177 patients. The VT Prediction Model's implementation on the validation cohort yielded effective WCT differentiation (AUC: 0.900; CI: 0.862‐0.939) with overall accuracy, sensitivity, and specificity of 85.0%, 80.4%, and 88.2%, respectively.ConclusionThe VT Prediction Model is an example of how readily available ECG measurements may be used to distinguish VT and SWCT effectively. Further study is needed to develop and refine newer WCT differentiation approaches that utilize computerized measurements provided by ECG interpretation software.