Polymorphic ventricular tachycardia and cardiac arrest from abiraterone-induced hypokalemia: a case report

BACKGROUND

Polymorphic ventricular tachycardia (PMVT) is an unstable and often fatal cardiac tachyarrhythmia. While there are many causes of this rhythm, including electrolyte imbalances, ischemia, and genetic disorders, iatrogenic etiologies are important to recognize. Abiraterone is an androgen synthesis antagonist effective in treating prostate cancer, but here we describe a case of severe hypokalemia secondary to abiraterone resulting in polymorphic ventricular tachycardia and cardiac arrest. While this is a potential adverse effect of the medication, severe hypokalemia causing polymorphic ventricular tachycardia and cardiac arrest, as seen in our patient's case, has not been described.

CASE PRESENTATION

A 78-year-old African-American man with history of prostate cancer presents with polymorphic ventricular tachycardia and cardiac arrest. After resuscitation, he was found to be severely hypokalemic and refractory to large doses of repletion. Evaluation of secondary causes of hypokalemia identified the likely culprit to be adverse effects from prostate cancer treatment.

CONCLUSION

A broad differential diagnosis for polymorphic ventricular tachycardia is essential in identifying and treating patients presenting in this rhythm. Here we present a case of iatrogenic polymorphic ventricular tachycardia secondary to oncologic treatment.

Postvaccination Symptoms After SARS-CoV-2 mRNA Vaccination Among Patients With Inflammatory Bowel Disease: A Prospective, Comparative Study

BACKGROUND

Vaccine hesitancy is prevalent among people with IBD, in part due to insufficient evidence regarding comparative safety of vaccines in this population.

METHODS

We conducted a nationwide comparative study of postvaccination symptoms among those with IBD and health care workers (HCWs) without IBD. Symptom frequency, severity, and duration were measured. Continuous and categorical data were analyzed using Wilcoxon rank-sum and Fisher's exact test. Regression analysis was used to adjust for confounding variables.

RESULTS

We had 2910 and 2746 subjects who completed a survey after dose 1 (D1) and dose 2 (D2) respectively (D1: HCW = 933, IBD = 1977; D2: HCW = 884, IBD = 1862). Mean age was 43 years, 67% were female, and 23% were nonwhite; 73% received BNT162b2 (Pfizer) including almost all HCWs and 60% of IBD patients. Most postvaccine symptoms were mild and lasted ≤2 days after both doses in both groups. Health care workers experienced more postvaccination symptoms overall than IBD patients after each dose (D1: 57% vs 35%, P < .001; D2: 73% vs 50%, P < .001). Gastrointestinal symptoms were noted in IBD more frequently after D1 (5.5% vs 3%, P = .003) but not after D2 (10% vs 13%, P = .07). Inflammatory bowel disease subjects who received mRNA-1273 (Moderna) reported more overall symptoms compared with BNT162b2 (57% vs 46%, P < .001) including gastrointestinal symptoms (12% vs 8%, P = .002) after D2.

CONCLUSIONS

People with IBD had fewer postvaccination symptoms following the first 2 doses of SARS-CoV-2 mRNA vaccines than HCWs. Among those with symptoms, most symptoms were mild and of short duration.

Electrocardiographic deep learning for predicting post-procedural mortality: a model development and validation study

BACKGROUND

Preoperative risk assessments used in clinical practice are insufficient in their ability to identify risk for postoperative mortality. Deep-learning analysis of electrocardiography can identify hidden risk markers that can help to prognosticate postoperative mortality. We aimed to develop a prognostic model that accurately predicts postoperative mortality in patients undergoing medical procedures and who had received preoperative electrocardiographic diagnostic testing.

METHODS

In a derivation cohort of preoperative patients with available electrocardiograms (ECGs) from Cedars-Sinai Medical Center (Los Angeles, CA, USA) between Jan 1, 2015 and Dec 31, 2019, a deep-learning algorithm was developed to leverage waveform signals to discriminate postoperative mortality. We randomly split patients (8:1:1) into subsets for training, internal validation, and final algorithm test analyses. Model performance was assessed using area under the receiver operating characteristic curve (AUC) values in the hold-out test dataset and in two external hospital cohorts and compared with the established Revised Cardiac Risk Index (RCRI) score. The primary outcome was post-procedural mortality across three health-care systems.

FINDINGS

45 969 patients had a complete ECG waveform image available for at least one 12-lead ECG performed within the 30 days before the procedure date (59 975 inpatient procedures and 112 794 ECGs): 36 839 patients in the training dataset, 4549 in the internal validation dataset, and 4581 in the internal test dataset. In the held-out internal test cohort, the algorithm discriminates mortality with an AUC value of 0·83 (95% CI 0·79-0·87), surpassing the discrimination of the RCRI score with an AUC of 0·67 (0·61-0·72). The algorithm similarly discriminated risk for mortality in two independent US health-care systems, with AUCs of 0·79 (0·75-0·83) and 0·75 (0·74-0·76), respectively. Patients determined to be high risk by the deep-learning model had an unadjusted odds ratio (OR) of 8·83 (5·57-13·20) for postoperative mortality compared with an unadjusted OR of 2·08 (0·77-3·50) for postoperative mortality for RCRI scores of more than 2. The deep-learning algorithm performed similarly for patients undergoing cardiac surgery (AUC 0·85 [0·77-0·92]), non-cardiac surgery (AUC 0·83 [0·79-0·88]), and catheterisation or endoscopy suite procedures (AUC 0·76 [0·72-0·81]).

INTERPRETATION

A deep-learning algorithm interpreting preoperative ECGs can improve discrimination of postoperative mortality. The deep-learning algorithm worked equally well for risk stratification of cardiac surgeries, non-cardiac surgeries, and catheterisation laboratory procedures, and was validated in three independent health-care systems. This algorithm can provide additional information to clinicians making the decision to perform medical procedures and stratify the risk of future complications.

FUNDING

National Heart, Lung, and Blood Institute.

Outcomes of Extremely Prolonged (> 50 d) Venovenous Extracorporeal Membrane Oxygenation Support

OBJECTIVES

There has been a sustained increase in the utilization of venovenous extracorporeal membrane oxygenation (ECMO) over the last decade, further exacerbated by the COVID-19 pandemic. We set out to describe our institutional experience with extremely prolonged (> 50 d) venovenous ECMO support for recovery or bridge to lung transplant candidacy in patients with acute respiratory failure.

DESIGN

Retrospective cohort study.

SETTING

A large tertiary urban care center.

PATIENTS

Patients 18 years or older receiving venovenous ECMO support for greater than 50 days, with initial cannulation between January 2018 and January 2022.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

One hundred thirty patients were placed on venovenous ECMO during the study period. Of these, 12 received prolonged (> 50 d) venovenous ECMO support. Eleven patients (92%) suffered from adult respiratory distress syndrome (ARDS) secondary to COVID-19, while one patient with prior bilateral lung transplant suffered from ARDS secondary to bacterial pneumonia. The median age of patients was 39 years (interquartile range [IQR], 35-51 yr). The median duration of venovenous ECMO support was 94 days (IQR, 70-128 d), with a maximum of 180 days. Median time from intubation to cannulation was 5 days (IQR, 2-14 d). Nine patients (75%) were successfully mobilized while on venovenous ECMO support. Successful weaning of venovenous ECMO support occurred in eight patients (67%); 6 (50%) were bridged to lung transplantation and 2 (17%) were bridged to recovery. Of those successfully weaned, seven patients (88%) were discharged from the hospital. All seven patients discharged from the hospital were alive 6 months post-decannulation; 83% (5/6) with sufficient follow-up time were alive 1-year after decannulation.

CONCLUSIONS

Our experience suggests that extremely prolonged venovenous ECMO support to allow native lung recovery or optimization for lung transplantation may be a feasible strategy in select critically ill patients, further supporting the expanded utilization of venovenous ECMO for refractory respiratory failure.

Deep learning-based electrocardiographic screening for chronic kidney disease

BACKGROUND

Undiagnosed chronic kidney disease (CKD) is a common and usually asymptomatic disorder that causes a high burden of morbidity and early mortality worldwide. We developed a deep learning model for CKD screening from routinely acquired ECGs.

METHODS

We collected data from a primary cohort with 111,370 patients which had 247,655 ECGs between 2005 and 2019. Using this data, we developed, trained, validated, and tested a deep learning model to predict whether an ECG was taken within one year of the patient receiving a CKD diagnosis. The model was additionally validated using an external cohort from another healthcare system which had 312,145 patients with 896,620 ECGs between 2005 and 2018.

RESULTS

Using 12-lead ECG waveforms, our deep learning algorithm achieves discrimination for CKD of any stage with an AUC of 0.767 (95% CI 0.760-0.773) in a held-out test set and an AUC of 0.709 (0.708-0.710) in the external cohort. Our 12-lead ECG-based model performance is consistent across the severity of CKD, with an AUC of 0.753 (0.735-0.770) for mild CKD, AUC of 0.759 (0.750-0.767) for moderate-severe CKD, and an AUC of 0.783 (0.773-0.793) for ESRD. In patients under 60 years old, our model achieves high performance in detecting any stage CKD with both 12-lead (AUC 0.843 [0.836-0.852]) and 1-lead ECG waveform (0.824 [0.815-0.832]).

CONCLUSIONS

Our deep learning algorithm is able to detect CKD using ECG waveforms, with stronger performance in younger patients and more severe CKD stages. This ECG algorithm has the potential to augment screening for CKD.

Hypertrophic Cardiomyopathy After Heart Transplantation: A Single-Center Case Series

We present 3 heart transplant recipients who developed hypertrophic cardiomyopathy years after transplantation. In all 3 cases, the diagnosis was initially made based on echocardiography and confirmed using cardiac magnetic resonance imaging. (Level of Difficulty: Advanced.).

Mechanically Ventilated Patients With Coronavirus Disease 2019 Had a Higher Chance of In-Hospital Death If Treated With High-Flow Nasal Cannula Oxygen Before Intubation

BACKGROUND

The impact of high-flow nasal cannula (HFNC) on outcomes of patients with respiratory failure from coronavirus disease 2019 (COVID-19) is unknown. We sought to assess whether exposure to HFNC before intubation was associated with successful extubation and in-hospital mortality compared to patients receiving intubation only.

METHODS

This single-center retrospective study examined patients with COVID-19-related respiratory failure from March 2020 to March 2021 who required HFNC, intubation, or both. Data were abstracted from the electronic health record. Use and duration of HFNC and intubation were examined' as well as demographics and clinical characteristics. We assessed the association between HFNC before intubation (versus without) and chance of successful extubation and in-hospital death using Cox proportional hazards models adjusting for age, sex, race/ethnicity, obesity, hypertension, diabetes, prior chronic obstructive pulmonary disease or asthma, HCO 3 , CO 2 , oxygen-saturation-to-inspired-oxygen (S:F) ratio, pulse, respiratory rate, temperature, and length of stay before intervention.

RESULTS

A total of n = 440 patients were identified, of whom 311 (70.7%) received HFNC before intubation, and 129 (29.3%) were intubated without prior use of HFNC. Patients who received HFNC before intubation had a higher chance of in-hospital death (hazard ratio [HR], 2.08; 95% confidence interval [CI], 1.06-4.05). No difference was found in the chance of successful extubation between the 2 groups (0.70, 0.41-1.20).

CONCLUSIONS

Among patients with respiratory failure from COVID-19 requiring mechanical ventilation, patients receiving HFNC before intubation had a higher chance of in-hospital death. Decisions on initial respiratory support modality should weigh the risks of intubation with potential increased mortality associated with HFNC.

Blinded, randomized trial of sonographer versus AI cardiac function assessment

Artificial intelligence (AI) has been developed for echocardiography1-3, although it has not yet been tested with blinding and randomization. Here we designed a blinded, randomized non-inferiority clinical trial (ClinicalTrials.gov ID: NCT05140642; no outside funding) of AI versus sonographer initial assessment of left ventricular ejection fraction (LVEF) to evaluate the impact of AI in the interpretation workflow. The primary end point was the change in the LVEF between initial AI or sonographer assessment and final cardiologist assessment, evaluated by the proportion of studies with substantial change (more than 5% change). From 3,769 echocardiographic studies screened, 274 studies were excluded owing to poor image quality. The proportion of studies substantially changed was 16.8% in the AI group and 27.2% in the sonographer group (difference of -10.4%, 95% confidence interval: -13.2% to -7.7%, P < 0.001 for non-inferiority, P < 0.001 for superiority). The mean absolute difference between final cardiologist assessment and independent previous cardiologist assessment was 6.29% in the AI group and 7.23% in the sonographer group (difference of -0.96%, 95% confidence interval: -1.34% to -0.54%, P < 0.001 for superiority). The AI-guided workflow saved time for both sonographers and cardiologists, and cardiologists were not able to distinguish between the initial assessments by AI versus the sonographer (blinding index of 0.088). For patients undergoing echocardiographic quantification of cardiac function, initial assessment of LVEF by AI was non-inferior to assessment by sonographers.

High-Throughput Assessment of Real-World Medication Effects on QT Interval Prolongation: Observational Study

BACKGROUND

Drug-induced prolongation of the corrected QT interval (QTc) increases the risk for Torsades de Pointes (TdP) and sudden cardiac death. Medication effects on the QTc have been studied in controlled settings but may not be well evaluated in real-world settings where medication effects may be modulated by patient demographics and comorbidities as well as the usage of other concomitant medications.

OBJECTIVE

We demonstrate a new, high-throughput method leveraging electronic health records (EHRs) and the Surescripts pharmacy database to monitor real-world QTc-prolonging medication and potential interacting effects from demographics and comorbidities.

METHODS

We included all outpatient electrocardiograms (ECGs) from September 2008 to December 2019 at a large academic medical system, which were in sinus rhythm with a heart rate of 40-100 beats per minute, QRS duration of <120 milliseconds, and QTc of 300-700 milliseconds, determined using the Bazett formula. We used prescription information from the Surescripts pharmacy database and EHR medication lists to classify whether a patient was on a medication during an ECG. Negative control ECGs were obtained from patients not currently on the medication but who had been or would be on that medication within 1 year. We calculated the difference in mean QTc between ECGs of patients who are on and those who are off a medication and made comparisons to known medication TdP risks per the CredibleMeds.org database. Using linear regression analysis, we studied the interaction of patient-level demographics or comorbidities on medication-related QTc prolongation.

RESULTS

We analyzed the effects of 272 medications on 310,335 ECGs from 159,397 individuals. Medications associated with the greatest QTc prolongation were dofetilide (mean QTc difference 21.52, 95% CI 10.58-32.70 milliseconds), mexiletine (mean QTc difference 18.56, 95% CI 7.70-29.27 milliseconds), amiodarone (mean QTc difference 14.96, 95% CI 13.52-16.33 milliseconds), rifaximin (mean QTc difference 14.50, 95% CI 12.12-17.13 milliseconds), and sotalol (mean QTc difference 10.73, 95% CI 7.09-14.37 milliseconds). Several top QT prolonging medications such as rifaximin, lactulose, cinacalcet, and lenalidomide were not previously known but have plausible mechanistic explanations. Significant interactions were observed between demographics or comorbidities and QTc prolongation with many medications, such as coronary disease and amiodarone.

CONCLUSIONS

We demonstrate a new, high-throughput technique for monitoring real-world effects of QTc-prolonging medications from readily accessible clinical data. Using this approach, we confirmed known medications for QTc prolongation and identified potential new associations and demographic or comorbidity interactions that could supplement findings in curated databases. Our single-center results would benefit from additional verification in future multisite studies that incorporate larger numbers of patients and ECGs along with more precise medication adherence and comorbidity data.

Racial and Ethnic Disparities in Hypertension: Barriers and Opportunities to Improve Blood Pressure Control

PURPOSE OF REVIEW

To characterize the barriers and opportunities associated with racial and ethnic disparities in blood pressure (BP) control.

RECENT FINDINGS

Blood pressure (BP) control rates in the USA have worsened over the last decade, with significantly lower rates of control among people from racial and ethnic minority groups, with non-Hispanic (NH) Black persons having 10% lower control rates compared to NH White counterparts. Many factors contribute to BP control including key social determinants of health (SDoH) such as health literacy, socioeconomic status, and access to healthcare as well as low awareness rates and dietary habits. Numerous pharmacologic and non-pharmacologic interventions have been developed to reduce racial and ethnic disparities in BP control. Among these, dietary programs designed to help reduce salt intake, faith-based interventions, and community-based programs have found success in achieving better BP control among people from racial and ethnic minority groups. Disparities in the prevalence and management of hypertension persist and remain high, particularly among racial and ethnic minority populations. Ongoing efforts are needed to address SDoH along with the unique genetic, social, economic, and cultural diversity within these groups that contribute to ongoing BP management inequalities.

Implementation of an Electronic Health Records-Based Safe Contrast Limit for Preventing Contrast-Associated Acute Kidney Injury After Percutaneous Coronary Intervention

BACKGROUND

Contrast-associated acute kidney injury (CA-AKI) after percutaneous coronary intervention is associated with increased mortality. We assessed the effectiveness of an electronic health records safe contrast limit tool in predicting CA-AKI risk and reducing contrast use and CA-AKI.

METHODS

We created an alert displaying the safe contrast limit to cardiac catheterization laboratory staff prior to percutaneous coronary intervention. The alert used risk factors automatically extracted from the electronic health records. We included procedures from June 1, 2020 to October 1, 2021; the intervention went live February 10, 2021. Using difference-in-differences analysis, we evaluated changes in contrast volume and CA-AKI rates after contrast limit tool implementation compared to control hospitals. Cardiologists were surveyed prior to and 9 months after alert implementation on beliefs, practice patterns, and safe contrast estimates for example patients.

RESULTS

At the one intervention site, there were 508 percutaneous coronary interventions before and 531 after tool deployment. At 15 control sites, there were 3550 and 3979 percutaneous coronary interventions, respectively. The contrast limit predicted CA-AKI with an accuracy of 64.1%, negative predictive value of 93.3%, and positive predictive value of 18.7%. After implementation, in high/modifiable risk patients (defined as having a calculated contrast limit <500ml) there was a small but significant -4.60 mL/month (95% CI, -8.24 to -1.00) change in average contrast use but no change in CA-AKI rates (odds ratio, 0.96 [95% CI, 0.84-1.10]). Low-risk patients had no change in contrast use (-0.50 mL/month [95% CI, -7.49 to 6.49]) or CA-AKI (odds ratio, 1.24 [95% CI, 0.79-1.93]). In assessing CA-AKI risk, clinicians heavily weighted age and diabetes but often did not consider anemia, cardiogenic shock, and heart failure.

CONCLUSIONS

Clinicians often used a simplified assessment of CA-AKI risk that did not include important risk factors, leading to risk estimations inconsistent with established models. Despite clinician skepticism, an electronic health records-based contrast limit tool more accurately predicted CA-AKI risk and was associated with a small decrease in contrast use during percutaneous coronary intervention but no change in CA-AKI rates.

Abstract 58: Implementation Of An Electronic Health Records-based Safe Contrast Limit For Preventing Contrast-associated Acute Kidney Injury After Percutaneous Coronary Intervention

Background: Contrast-associated acute kidney injury (CA-AKI) after percutaneous coronary intervention (PCI) is associated with cardiovascular complications and mortality. We previously published a multivariable model for calculating safe contrast volume limits derived from National Cardiovascular Data Registry (NCDR) data. We assessed the performance of this safe contrast limit when implemented in an electronic health records (EHR) system.

Methods: We created an advisory alert in our Epic EHR system, which displayed the safe contrast limit to catheterization lab staff and operators prior to PCI along with CA-AKI risk reduction strategies. The alert required no clinician data entry and used risk factors automatically pulled from the EHR: age, sex, body mass index, creatinine clearance, hemoglobin, and use of intra-aortic balloon pump pre-procedure. We included procedures from 11/1/2020 to 10/1/2021 that had pre- and post-procedure creatinine data; the intervention went live 2/10/2021. We used NCDR-defined CA-AKI events to assess the performance of the safe contrast limit in predicting CA-AKI. We used interrupted time series analysis to determine if the intervention reduced contrast use and CA-AKI.

Results: There were 201 PCIs; 10.4% (21 out of 201) had CA-AKI. The contrast limit predicted CA-AKI well using real-time EHR data (sensitivity 76.2%, specificity 50.0%, negative predictive value 94.7%). Among 178 patients with modifiable CA-AKI risk (defined as safe contrast limit ≤ 500mL), the intervention was associated with an immediate significant decrease in contrast use (-44.34mL (95% CI -81.71, -6.97)) and a decrease with time (-0.77mL/day (-1.31, -0.22)). For 23 patients at low CA-AKI risk (contrast limit > 500mL), there was a non-significant increase in contrast use (28.55 (-92.79, 149.88) and with time (1.15 (-0.73, 3.03)). There was no significant decrease in CA-AKI (OR 0.21 (0.03, 1.37)).

Conclusion: An EHR-based safe contrast limit using automatically derived patient data performed well in predicting CA-AKI. The intervention significantly decreased contrast use for patients with modifiable CA-AKI risk and non-significantly increased contrast in patients with low risk. Longer follow-up will help clarify effects on CA-AKI rates.

Abstract 33: Interventional Cardiologist Beliefs And Practices Before And After Implementation Of An Electronic Health Records-based Safe Contrast Limit Tool For Percutaneous Coronary Interventions

Background: Best Practice Advisories (BPAs) within electronic health record (EHR) systems can inform clinician decision-making and improve patient care. Their success depends on BPA content, clinical context, and staff buy-in. We recently implemented a safe contrast limit tool to reduce contrast-associated AKI (CA-AKI) after percutaneous coronary intervention (PCI). We evaluated the impact of this Contrast Limit BPA on interventional cardiologist attitudes and practices, and examined possible factors that influenced the BPA’s success.

Methods: Using a published model, we implemented a BPA in our Epic EHR that displayed calculated individualized safe contrast limits prior to PCI. Cardiologists were surveyed prior to and 9 months after BPA implementation. Pre-implementation questions covered beliefs about CA-AKI, practice patterns, knowledge of CA-AKI risk factors, and asked for safe contrast estimates for example patients. Post-implementation questions assessed practice patterns and the BPA’s perceived accuracy, efficacy, and utility. Survey data was compared to clinician PCI contrast use using logistic regression.

Results: We surveyed 8 clinicians pre-implementation and 10 post-implementation. Pre-implementation, 25% (2/8) reported using a contrast limit to make decisions about PCI and 12.5% (1/8) believed that they could improve their CA-AKI rates. In both their assessment of CA-AKI risk factors and contrast limit estimations, respondents often overestimated the contribution of age and diabetes while underestimating the influence of anemia and cardiogenic shock. Post-implementation, 30% (3/10) stated they were often surprised by the contrast limit and 80% (8/10) reported using the Contrast Limit when making PCI decisions. Clinicians who found the BPA to be clear and understandable or had catheterization lab staff discuss the limit prior to PCI had significantly lower odds of exceeding the safe contrast limit (OR=4.63, 95% CI [1.17, 18.35]; OR=6.76, 95% CI [1.81, 25.27]).

Conclusion: Clinicians often misestimated safe contrast limits and believed that there was little room for improvement in their CA-AKI rates. Despite initial skepticism, an EHR-based safe contrast limit BPA was frequently used and influenced contrast use.

High-Throughput Precision Phenotyping of Left Ventricular Hypertrophy With Cardiovascular Deep Learning

IMPORTANCE

Early detection and characterization of increased left ventricular (LV) wall thickness can markedly impact patient care but is limited by under-recognition of hypertrophy, measurement error and variability, and difficulty differentiating causes of increased wall thickness, such as hypertrophy, cardiomyopathy, and cardiac amyloidosis.

OBJECTIVE

To assess the accuracy of a deep learning workflow in quantifying ventricular hypertrophy and predicting the cause of increased LV wall thickness.

DESIGN, SETTINGS, AND PARTICIPANTS

This cohort study included physician-curated cohorts from the Stanford Amyloid Center and Cedars-Sinai Medical Center (CSMC) Advanced Heart Disease Clinic for cardiac amyloidosis and the Stanford Center for Inherited Cardiovascular Disease and the CSMC Hypertrophic Cardiomyopathy Clinic for hypertrophic cardiomyopathy from January 1, 2008, to December 31, 2020. The deep learning algorithm was trained and tested on retrospectively obtained independent echocardiogram videos from Stanford Healthcare, CSMC, and the Unity Imaging Collaborative.

MAIN OUTCOMES AND MEASURES

The main outcome was the accuracy of the deep learning algorithm in measuring left ventricular dimensions and identifying patients with increased LV wall thickness diagnosed with hypertrophic cardiomyopathy and cardiac amyloidosis.

RESULTS

The study included 23 745 patients: 12 001 from Stanford Health Care (6509 [54.2%] female; mean [SD] age, 61.6 [17.4] years) and 1309 from CSMC (808 [61.7%] female; mean [SD] age, 62.8 [17.2] years) with parasternal long-axis videos and 8084 from Stanford Health Care (4201 [54.0%] female; mean [SD] age, 69.1 [16.8] years) and 2351 from CSMS (6509 [54.2%] female; mean [SD] age, 69.6 [14.7] years) with apical 4-chamber videos. The deep learning algorithm accurately measured intraventricular wall thickness (mean absolute error [MAE], 1.2 mm; 95% CI, 1.1-1.3 mm), LV diameter (MAE, 2.4 mm; 95% CI, 2.2-2.6 mm), and posterior wall thickness (MAE, 1.4 mm; 95% CI, 1.2-1.5 mm) and classified cardiac amyloidosis (area under the curve [AUC], 0.83) and hypertrophic cardiomyopathy (AUC, 0.98) separately from other causes of LV hypertrophy. In external data sets from independent domestic and international health care systems, the deep learning algorithm accurately quantified ventricular parameters (domestic: R2, 0.96; international: R2, 0.90). For the domestic data set, the MAE was 1.7 mm (95% CI, 1.6-1.8 mm) for intraventricular septum thickness, 3.8 mm (95% CI, 3.5-4.0 mm) for LV internal dimension, and 1.8 mm (95% CI, 1.7-2.0 mm) for LV posterior wall thickness. For the international data set, the MAE was 1.7 mm (95% CI, 1.5-2.0 mm) for intraventricular septum thickness, 2.9 mm (95% CI, 2.4-3.3 mm) for LV internal dimension, and 2.3 mm (95% CI, 1.9-2.7 mm) for LV posterior wall thickness. The deep learning algorithm accurately detected cardiac amyloidosis (AUC, 0.79) and hypertrophic cardiomyopathy (AUC, 0.89) in the domestic external validation site.

CONCLUSIONS AND RELEVANCE

In this cohort study, the deep learning model accurately identified subtle changes in LV wall geometric measurements and the causes of hypertrophy. Unlike with human experts, the deep learning workflow is fully automated, allowing for reproducible, precise measurements, and may provide a foundation for precision diagnosis of cardiac hypertrophy.

Deep learning evaluation of biomarkers from echocardiogram videos

BACKGROUND

Laboratory testing is routinely used to assay blood biomarkers to provide information on physiologic state beyond what clinicians can evaluate from interpreting medical imaging. We hypothesized that deep learning interpretation of echocardiogram videos can provide additional value in understanding disease states and can evaluate common biomarkers results.

METHODS

We developed EchoNet-Labs, a video-based deep learning algorithm to detect evidence of anemia, elevated B-type natriuretic peptide (BNP), troponin I, and blood urea nitrogen (BUN), as well as values of ten additional lab tests directly from echocardiograms. We included patients (n = 39,460) aged 18 years or older with one or more apical-4-chamber echocardiogram videos (n = 70,066) from Stanford Healthcare for training and internal testing of EchoNet-Lab's performance in estimating the most proximal biomarker result. Without fine-tuning, the performance of EchoNet-Labs was further evaluated on an additional external test dataset (n = 1,301) from Cedars-Sinai Medical Center. We calculated the area under the curve (AUC) of the receiver operating characteristic curve for the internal and external test datasets.

FINDINGS

On the held-out test set of Stanford patients not previously seen during model training, EchoNet-Labs achieved an AUC of 0.80 (0.79-0.81) in detecting anemia (low hemoglobin), 0.86 (0.85-0.88) in detecting elevated BNP, 0.75 (0.73-0.78) in detecting elevated troponin I, and 0.74 (0.72-0.76) in detecting elevated BUN. On the external test dataset from Cedars-Sinai, EchoNet-Labs achieved an AUC of 0.80 (0.77-0.82) in detecting anemia, of 0.82 (0.79-0.84) in detecting elevated BNP, of 0.75 (0.72-0.78) in detecting elevated troponin I, and of 0.69 (0.66-0.71) in detecting elevated BUN. We further demonstrate the utility of the model in detecting abnormalities in 10 additional lab tests. We investigate the features necessary for EchoNet-Labs to make successful detection and identify potential mechanisms for each biomarker using well-known and novel explainability techniques.

INTERPRETATION

These results show that deep learning applied to diagnostic imaging can provide additional clinical value and identify phenotypic information beyond current imaging interpretation methods.

FUNDING

J.W.H. and B.H. are supported by the NSF Graduate Research Fellowship. D.O. is supported by NIH K99 HL157421-01. J.Y.Z. is supported by NSF CAREER 1942926, NIH R21 MD012867-01, NIH P30AG059307 and by a Chan-Zuckerberg Biohub Fellowship.

Improving Efficiency of the Barbershop Model of Hypertension Care for Black Men With Virtual Visits

Background

The LABBPS (Los Angeles Barbershop Blood Pressure Study) developed a new model of hypertension care for non‐Hispanic Black men that links health promotion by barbers to medication management by pharmacists. Barriers to scaling the model include inefficiencies that contribute to the cost of the intervention, most notably, pharmacist travel time. To address this, we tested whether virtual visits could be substituted for in‐person visits after blood pressure (BP) control was achieved.

Methods and Results

We enrolled 10 Black male patrons with systolic BP ≥140 mm Hg into a proof‐of‐concept study in which barbers promoted follow‐up with pharmacists who initially met each patron in the barbershop, where they prescribed BP medication under a collaborative practice agreement with the patrons' physician. Medications were titrated during bimonthly in‐person visits to achieve a BP goal of ≤130/80 mm Hg. Once BP goal was reached, visits were done by videoconference. Final BP and safety outcomes were assessed at 12 months. Nine patients completed the intervention. Baseline BP of 155±14/83.9±11 mm Hg decreased by −28.7±13/−8.9±15 mm Hg (P<0.0001). These data are statistically indistinguishable from prior LABBPS data (P=0.8 for change in systolic BP and diastolic BP). Hypertension control (≤130/80 mm Hg) was 67% (6 of 9), numerically greater than the 63% observed in LABBPS (P=not significant). As intended, the mean number of in‐person visits decreased from 11 in LABBPS to 6.6 visits over 12 months. No treatment‐related serious adverse events occurred.

Conclusions

Virtual visits represent a viable substitute for in‐person visits, both improving pharmacist efficiency and reducing cost while preserving intervention potency.

Registration

URL: https://www.clinicaltrials.gov; Unique identifier: NCT 03726710.

A Machine Learning Algorithm Predicts Duration of hospitalization in COVID-19 patients

The COVID-19 pandemic has placed unprecedented strain on the healthcare system, particularly hospital bed capacity in the setting of large variations in patient length of stay (LOS). Using electronic health record data from 966 COVID-19 patients at a large academic medical center, we developed three machine learning algorithms to predict the likelihood of prolonged LOS, defined as >8 days. The models included 353 variables and were trained on 80% of the cohort, with 20% used for model validation. The three models were created on hospital days 1, 2 and 3, each including information available at or before that point in time. The models’ predictive capabilities improved sequentially over time, reaching an accuracy of 0.765, with an AUC of 0.819 by day 3. These models, developed using readily available data, may help hospital systems prepare for bed capacity needs, and help clinicians counsel patients on their likelihood of prolonged hospitalization.

Pseudo-safety in a cohort of patients with COVID-19 discharged home from the emergency department

INTRODUCTION

EDs are often the first line of contact with individuals infected with COVID-19 and play a key role in triage. However, there is currently little specific guidance for deciding when patients with COVID-19 require hospitalisation and when they may be safely observed as an outpatient.

METHODS

In this retrospective study, we characterised all patients with COVID-19 discharged home from EDs in our US multisite healthcare system from March 2020 to August 2020, focusing on individuals who returned within 2 weeks and required hospital admission. We restricted analyses to first-encounter data that do not depend on laboratory or imaging diagnostics in order to inform point-of-care assessments in resource-limited environments. Vitals and comorbidities were extracted from the electronic health record. We performed ordinal logistic regression analyses to identify predictors of inpatient admission, intensive care and intubation.

RESULTS

Of n=923 patients who were COVID-19 positive discharged from the ED, n=107 (11.6%) returned within 2 weeks and were admitted. In a multivariable-adjusted model including n=788 patients with complete risk factor information, history of hypertension increased odds of hospitalisation and severe illness by 1.92-fold (95% CI 1.07 to 3.41), diabetes by 2.20-fold (1.18 to 4.02), chronic lung disease by 2.21-fold (1.22 to 3.92) and fever by 2.89-fold (1.71 to 4.82). Having at least two of these risk factors increased the odds of future hospitalisation by 6.68-fold (3.54 to 12.70). Patients with hypertension, diabetes, chronic lung disease or fever had significantly longer hospital stays (median 5.92 days, 3.08-10.95 vs 3.21, 1.10-5.75, p<0.01) with numerically higher but not significantly different rates of intensive care unit admission (27.02% vs 14.30%, p=0.27) and intubation (12.16% vs 7.14%, p=0.71).

DISCUSSION

Patients infected with COVID-19 may appear clinically safe for home convalescence. However, those with hypertension, diabetes, chronic lung disease and fever may in fact be only 'pseudo-safe' and are most at risk for subsequent hospitalisation with more severe illness and longer hospital stays.

Percutaneous mitral valve repair with MitraClip XTR for acute mitral regurgitation due to papillary muscle rupture

Papillary muscle rupture is an infrequent and highly morbid mechanical complication of acute myocardial infarction. Surgical repair or replacement is traditionally considered first-line therapy. However, many of these patients present in extremis with prohibitively high surgical risk. Repair of mitral regurgitation with the MitraClip device (Abbot Vascular, Menlo Park, CA, USA) is an established therapy to treat degenerative and functional mitral regurgitation. We present a case of successful repair of severe mitral regurgitation due to papillary muscle rupture in the setting of acute myocardial infarction. A two-clip strategy resulted in mild residual mitral regurgitation with resolution of cardiogenic shock and refractory hypoxemia requiring veno-venous extracorporeal membrane oxygenation. Six-month follow-up echocardiogram identified durable results with mild mitral regurgitation and left ventricular ejection fraction of 63 %. Our case demonstrates that percutaneous mitral valve repair with MitraClip is a well-tolerated procedure that can provide acute and long-term benefit for patients with acute mitral regurgitation due to papillary muscle rupture who are at prohibitively high surgical risk. .

A machine learning algorithm to increase COVID-19 inpatient diagnostic capacity

Worldwide, testing capacity for SARS-CoV-2 is limited and bottlenecks in the scale up of polymerase chain reaction (PCR-based testing exist. Our aim was to develop and evaluate a machine learning algorithm to diagnose COVID-19 in the inpatient setting. The algorithm was based on basic demographic and laboratory features to serve as a screening tool at hospitals where testing is scarce or unavailable. We used retrospectively collected data from the UCLA Health System in Los Angeles, California. We included all emergency room or inpatient cases receiving SARS-CoV-2 PCR testing who also had a set of ancillary laboratory features (n = 1,455) between 1 March 2020 and 24 May 2020. We tested seven machine learning models and used a combination of those models for the final diagnostic classification. In the test set (n = 392), our combined model had an area under the receiver operator curve of 0.91 (95% confidence interval 0.87-0.96). The model achieved a sensitivity of 0.93 (95% CI 0.85-0.98), specificity of 0.64 (95% CI 0.58-0.69). We found that our machine learning algorithm had excellent diagnostic metrics compared to SARS-CoV-2 PCR. This ensemble machine learning algorithm to diagnose COVID-19 has the potential to be used as a screening tool in hospital settings where PCR testing is scarce or unavailable.

Experience With Hydroxychloroquine and Azithromycin in the Coronavirus Disease 2019 Pandemic: Implications for QT Interval Monitoring

Background Despite a lack of clinical evidence, hydroxychloroquine and azithromycin are being administered widely to patients with verified or suspected coronavirus disease 2019 (COVID-19). Both drugs may increase risk of lethal arrhythmias associated with QT interval prolongation. Methods and Results We analyzed a case series of COVID-19-positive/suspected patients admitted between February 1, 2020, and April 4, 2020, who were treated with azithromycin, hydroxychloroquine, or a combination of both drugs. We evaluated baseline and postmedication QT interval (corrected QT interval [QTc]; Bazett) using 12-lead ECGs. Critical QTc prolongation was defined as follows: (1) maximum QTc ≥500 ms (if QRS <120 ms) or QTc ≥550 ms (if QRS ≥120 ms) and (2) QTc increase of ≥60 ms. Tisdale score and Elixhauser comorbidity index were calculated. Of 490 COVID-19-positive/suspected patients, 314 (64%) received either/both drugs and 98 (73 COVID-19 positive and 25 suspected) met study criteria (age, 62±17 years; 61% men). Azithromycin was prescribed in 28%, hydroxychloroquine in 10%, and both in 62%. Baseline mean QTc was 448±29 ms and increased to 459±36 ms (P=0.005) with medications. Significant prolongation was observed only in men (18±43 ms versus -0.2±28 ms in women; P=0.02). A total of 12% of patients reached critical QTc prolongation. Changes in QTc were highest with the combination compared with either drug, with much greater prolongation with combination versus azithromycin (17±39 ms versus 0.5±40 ms; P=0.07). No patients manifested torsades de pointes. Conclusions Overall, 12% of patients manifested critical QTc prolongation, and the combination caused greater prolongation than either drug alone. The balance between uncertain benefit and potential risk when treating COVID-19 patients should be carefully assessed.

Development and Evaluation of Novel Electronic Medical Record Tools For Avoiding Bleeding After Percutaneous Coronary Intervention

Background

Bleeding remains the most common complication of percutaneous coronary intervention. Guidelines recommend assessing bleeding risk before percutaneous coronary intervention to target use of bleeding avoidance strategies and mitigate bleeding events. Cedars‐Sinai Medical Center undertook an initiative to integrate these recommendations into the electronic medical record.

Methods and Results

The intervention included a voluntary clinical decision alert to assess bleeding risk before percutaneous coronary intervention, a bleeding risk calculator tool based on the NCDR (National Cardiovascular Data Registry) risk prediction model and, when indicated, a second alert to consider 4 bleeding avoidance strategies. We tested for changes in the use of bleeding avoidance strategies and bleeding event rates by comparing procedures performed before versus after implementation of the electronic medical record–based intervention and with versus without use of the bleeding risk calculator tool. Use of radial access increased (47.6% versus 64.8%; P<0.001) and glycoprotein IIb/IIIa inhibitors decreased (12.8% versus 3.17%; P<0.001) from before to after implementation, though risk‐adjusted bleeding event rates were stable (odds ratio, 0.82; P=0.164), even for high‐risk procedures. Use versus nonuse of the bleeding risk calculator tool was associated with increased radial access and reductions in glycoprotein IIb/IIIa inhibitors, but no change in bleeding events.

Conclusions

Integrating guideline recommendations into the electronic medical record to promote assessments of bleeding risk and use of bleeding avoidance strategies was feasible and associated with changes in clinical practice. Future work is needed to ensure that bleeding avoidance strategies are not overused among lower‐risk patients, and that, for high‐risk patients, the potential benefits of elective percutaneous coronary intervention are carefully weighed against the risk of bleeding.

Abstract P156: Long-Term Blood Pressure Outcomes and Costs Associated With a Barber-Pharmacist Hypertension Intervention: 10-Year Simulation of the Los Angeles Barber Trial

Background: Barber-pharmacist blood pressure (BP) management reduced systolic BP >20 mm Hg compared to education alone over one year in Non-Hispanic (NH) black men with uncontrolled hypertension (HTN) in the Los Angeles BARBER (LA-BARBER) Trial. Long-term BP outcomes and costs of this intervention are unknown.

Objective: Simulate 10-year BP outcomes and intervention and medication costs of a one-year barber-pharmacist HTN intervention followed by usual care compared to usual care alone.

Methods: We simulated 1000 LA BARBER-eligible NH black men sampled from the National Health and Nutrition Examination Survey. We used a discrete event simulation version of the validated BP Control Model to predict BP outcomes over 10 years. Model inputs were derived from published literature, national sources, LA-BARBER individual participant data, and interviews with LA-BARBER intervention pharmacists. Medication and intervention (clinical care, travel, and administrative time) costs were calculated from a 2018 US payer perspective and discounted 3% annually. Primary outcomes were percent with BP <130/80 mm Hg, medication and intervention costs, and the cost per patient with BP controlled. We used 100 probabilistic model iterations to examine parameter uncertainty.

Results: Our calibrated model accurately reproduced baseline characteristics and predicted 1-year BP outcomes of the LA-BARBER study (mean age 54 years, baseline BP 153/91 mm Hg, 69% predicted 1-year BP control with the barber-pharmacist HTN intervention). Over 10 years, 64% of patients (95% uncertainty interval [UI] 61%-67%) were predicted to achieve BP control with the barber-pharmacist HTN intervention compared to 38% (95% UI 26%-48%) with usual care. Projected intervention and medication costs were $8084 (95% UI $7664-$8598) with the barber-pharmacist intervention compared to $4183 ($3514-$4768) with usual care costing $15,000 per controlled patient gained.

Conclusions: Medication and intervention costs for a barber-pharmacist HTN intervention were about double usual care, but it achieved substantially higher BP control rates. Ongoing research will examine if this strategy meets traditional cost-effectiveness thresholds (e.g.,<$100,000 per quality-adjusted life year).

From D2B to B2D: Value based STEMI care!

Use of the CADILLIAC Risk Score can accurately identify patients for safe early discharge after PCI for STEMI. Early discharge has the potential to both improve the quality and decrease the cost of care for STEMI patients. Prospective validation of this risk score and formal cost analysis would facilitate widespread utilization.