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.

Association of blood pressure variability during acute care hospitalization and incident dementia

Background and objectives

Recognized as a potential risk factor for Alzheimer's disease and related dementias (ADRD), blood pressure variability (BPV) could be leveraged to facilitate identification of at-risk individuals at a population level. Granular BPV data are available during acute care hospitalization periods for potentially high-risk patients, but the incident ADRD risk association with BPV measured in this setting is unknown. Our objective was to evaluate the relation of BPV, measured during acute care hospitalization, and incidence of ADRD.

Methods

We retrospectively studied adults, without a prior ADRD diagnosis, who were admitted to a large quaternary care medical center in Southern California between January 1, 2013 and December 31, 2019. For all patients, determined BPV, calculated as variability independent of the mean (VIM), using blood pressure readings obtained as part of routine clinical care. We used multivariable Cox proportional hazards regression to examine the association between BP VIM during hospitalization and the development of incident dementia, determined by new ICD-9/10 coding or the new prescription of dementia medication, occurring at least 2 years after the index hospitalization.

Results

Of 81,892 adults hospitalized without a prior ADRD diagnosis, 2,442 (2.98%) went on to develop ADRD (2.6 to 5.2 years after hospitalization). In multivariable-adjusted Cox models, both systolic (HR 1.05, 95% CI 1.00-1.09) and diastolic (1.06, 1.02-1.10) VIM were associated with incident ADRD. In pre-specified stratified analyses, the VIM associations with incident ADRD were most pronounced in individuals over age 60 years and among those with renal disease or hypertension. Results were similar when repeated to include incident ADRD diagnoses made at least 1 or 3 years after index hospitalization.

Discussion

We found that measurements of BPV from acute care hospitalizations can be used to identify individuals at risk for developing a diagnosis of ADRD within approximately 5 years. Use of the readily accessible BPV measure may allow healthcare systems to risk stratify patients during periods of intense patient-provider interaction and, in turn, facilitate engagement in ADRD screening programs.

Association of COVID-19 Vaccination With Risk for Incident Diabetes After COVID-19 Infection

This cohort study compares the risk of new-onset hypertension, hyperlipidemia, and diabetes before and after COVID-19 infection among patients who were vaccinated vs unvaccinated before infection.

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.

Prediction of Coronary Artery Calcium Using Deep Learning of Echocardiograms

BACKGROUND

Coronary artery calcification (CAC), often assessed by computed tomography (CT), is a powerful marker of coronary artery disease that can guide preventive therapies. Computed tomographies, however, are not always accessible or serially obtainable. It remains unclear whether other widespread tests such as transthoracic echocardiograms (TTEs) can be used to predict CAC.

METHODS

Using a data set of 2,881 TTE videos paired with coronary calcium CTs, we trained a video-based artificial intelligence convolutional neural network to predict CAC scores from parasternal long-axis views. We evaluated the model's ability to classify patients from a held-out sample as well as an external site sample into zero CAC and high CAC (CAC ≥ 400 Agatston units) groups by receiver operating characteristic and precision-recall curves. We also investigated whether such classifications prognosticated significant differences in 1-year mortality rates by the log-rank test of Kaplan-Meier curves.

RESULTS

Transthoracic echocardiogram artificial intelligence models had high discriminatory abilities in predicting zero CAC (receiver operating characteristic area under the curve [AUC] = 0.81 [95% CI, 0.74-0.88], F1 score = 0.95) and high CAC (AUC = 0.74 [0.68-0.8], F1 score = 0.74). This performance was confirmed in an external test data set of 92 TTEs (AUC = 0.75 [0.65-0.85], F1 score = 0.77; and AUC = 0.85 [0.76-0.93], F1 score = 0.59, respectively). Risk stratification by TTE-predicted CAC performed similarly to CT CAC scores in prognosticating significant differences in 1-year survival in high-CAC patients (CT CAC ≥ 400 vs CT CAC < 400, P = .03; TTE-predicted CAC ≥ 400 vs TTE-predicted CAC < 400, P = .02).

CONCLUSIONS

A video-based deep learning model successfully used TTE videos to predict zero CAC and high CAC with high accuracy. Transthoracic echocardiography-predicted CAC prognosticated differences in 1-year survival similar to CT CAC. Deep learning of TTEs holds promise for future adjunctive coronary artery disease risk stratification to guide preventive therapies.

Cardiovascular and hepatic disease associations by magnetic resonance imaging: A retrospective cohort study

Background

Hepatic disease is linked to cardiovascular events but the independent association between hepatic and cardiovascular disease remains unclear, given shared risk factors.

Methods

This was a retrospective study of consecutive patients with a clinical cardiac MRI (CMR) and a serological marker of hepatic fibrosis, the FIB-4 score, within one year of clinical imaging. We assessed the relations between FIB-4 scores grouped based on prior literature: low (< 1.3), moderate (1.3–3.25), and high (>3.25), and abnormalities detected by comprehensive CMR grouped into 4 domains: cardiac structure (end diastolic volumes, atrial dimensions, wall thickness); cardiac function (ejection fractions, wall motion abnormalities, cardiac output); vascular structure (ascending aortic and pulmonary arterial sizes); and cardiac composition (late gadolinium enhancement, T1 and T2 times). We used Poisson regression to examine the association between the conventionally defined FIB-4 category (low <1.3, moderate 1.3–3.25, and high >3.25) and any CMR abnormality while adjusting for demographics and traditional cardiovascular risk factors.

Results

Of the 1668 patients studied (mean age: 55.971 ± 7.28, 901 [54%] male), 85.9% had ≥1 cardiac abnormality with increasing prevalence seen within the low (82.0%) to moderate (88.8%) to high (92.3%) FIB-4 categories. Multivariable analyses demonstrated the presence of any cardiac abnormality was significantly associated with having a high-range FIB-4 (prevalence ratio 1.07, 95% CI: 1.01–1.13); notably, the presence of functional cardiac abnormalities were associated with being in the high FIB-4 range (1.41, 1.21–1.65) and any vascular abnormalities with being in the moderate FIB-4 range (1.22, 1.01–1.47).

Conclusions

Elevated FIB-4 was associated with cardiac functional and vascular abnormalities even after adjustment for shared risk factors in a cohort of patients with clinically referred CMR. These CMR findings indicate that cardiovascular abnormalities exist in the presence of subclinical hepatic fibrosis, irrespective of shared risk factors, underscoring the need for further studies of the heart-liver axis.

Variability independent of mean blood pressure as a real-world measure of cardiovascular risk

BACKGROUND

Individual-level blood pressure (BP) variability, independent of mean BP levels, has been associated with increased risk for cardiovascular events in cohort studies and clinical trials using standardized BP measurements. The extent to which BP variability relates to cardiovascular risk in the real-world clinical practice setting is unclear. We sought to determine if BP variability in clinical practice is associated with adverse cardiovascular outcomes using clinically generated data from the electronic health record (EHR).

METHODS

We identified 42,482 patients followed continuously at a single academic medical center in Southern California between 2013 and 2019 and calculated their systolic and diastolic BP variability independent of the mean (VIM) over the first 3 years of the study period. We then performed multivariable Cox proportional hazards regression to examine the association between VIM and both composite and individual outcomes of interest (incident myocardial infarction, heart failure, stroke, and death).

FINDINGS

Both systolic (HR, 95% CI 1.22, 1.17-1.28) and diastolic VIM (1.24, 1.19-1.30) were positively associated with the composite outcome, as well as all individual outcome measures. These findings were robust to stratification by age, sex and clinical comorbidities. In sensitivity analyses using a time-shifted follow-up period, VIM remained significantly associated with the composite outcome for both systolic (1.15, 1.11-1.20) and diastolic (1.18, 1.13-1.22) values.

INTERPRETATION

VIM derived from clinically generated data remains associated with adverse cardiovascular outcomes and represents a risk marker beyond mean BP, including in important demographic and clinical subgroups. The demonstrated prognostic ability of VIM derived from non-standardized BP readings indicates the utility of this measure for risk stratification in a real-world practice setting, although residual confounding from unmeasured variables cannot be excluded.

FUNDING

This study was funded in part by National Institutes of Health grants R01-HL134168, R01-HL131532, R01-HL143227, R01-HL142983, U54-AG065141; R01-HL153382, K23-HL136853, K23-HL153888, and K99-HL157421; China Scholarship Council grant 201806260086; Academy of Finland (Grant no: 321351); Emil Aaltonen Foundation; Finnish Foundation for Cardiovascular Research.

Seasonal COVID-19 surge related hospital volumes and case fatality rates

BACKGROUND

Seasonal and regional surges in COVID-19 have imposed substantial strain on healthcare systems. Whereas sharp inclines in hospital volume were accompanied by overt increases in case fatality rates during the very early phases of the pandemic, the relative impact during later phases of the pandemic are less clear. We sought to characterize how the 2020 winter surge in COVID-19 volumes impacted case fatality in an adequately-resourced health system.

METHODS

We performed a retrospective cohort study of all adult diagnosed with COVID-19 in a large academic healthcare system between August 25, 2020 to May 8, 2021, using multivariable logistic regression to examine case fatality rates across 3 sequential time periods around the 2020 winter surge: pre-surge, surge, and post-surge. Subgroup analyses of patients admitted to the hospital and those receiving ICU-level care were also performed. Additionally, we used multivariable logistic regression to examine risk factors for mortality during the surge period.

RESULTS

We studied 7388 patients (aged 52.8 ± 19.6 years, 48% male) who received outpatient or inpatient care for COVID-19 during the study period. Patients treated during surge (N = 6372) compared to the pre-surge (N = 536) period had 2.64 greater odds (95% CI 1.46-5.27) of mortality after adjusting for sociodemographic and clinical factors. Adjusted mortality risk returned to pre-surge levels during the post-surge period. Notably, first-encounter patient-level measures of illness severity appeared higher during surge compared to non-surge periods.

CONCLUSIONS

We observed excess mortality risk during a recent winter COVID-19 surge that was not explained by conventional risk factors or easily measurable variables, although recovered rapidly in the setting of targeted facility resources. These findings point to how complex interrelations of population- and patient-level pandemic factors can profoundly augment health system strain and drive dynamic, if short-lived, changes in outcomes.

Symptomology following mRNA vaccination against SARS-CoV-2

Despite demonstrated efficacy of vaccines against severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), the causative agent of coronavirus disease-2019 (COVID-19), widespread hesitancy to vaccination persists. Improved knowledge regarding frequency, severity, and duration of vaccine-associated symptoms may help reduce hesitancy. In this prospective observational study, we studied 1032 healthcare workers who received both doses of the Pfizer-BioNTech SARS-CoV-2 mRNA vaccine and completed post-vaccine symptom surveys both after dose 1 and after dose 2. We defined appreciable post-vaccine symptoms as those of at least moderate severity and lasting at least 2 days. We found that symptoms were more frequent following the second vaccine dose than the first (74% vs. 60%, P < 0.001), with >80% of all symptoms resolving within 2 days. The most common symptom was injection site pain, followed by fatigue and malaise. Overall, 20% of participants experienced appreciable symptoms after dose 1 and 30% after dose 2. In multivariable analyses, female sex was associated with greater odds of appreciable symptoms after both dose 1 (OR, 95% CI 1.73, 1.19-2.51) and dose 2 (1.76, 1.28-2.42). Prior COVID-19 was also associated with appreciable symptoms following dose 1, while younger age and history of hypertension were associated with appreciable symptoms after dose 2. We conclude that most post-vaccine symptoms are reportedly mild and last <2 days. Appreciable post-vaccine symptoms are associated with female sex, prior COVID-19, younger age, and hypertension. This information can aid clinicians in advising patients on the safety and expected symptomatology associated with vaccination.

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.

Temporal variations in the severity of COVID-19 illness by race and ethnicity

Introduction

Early reports highlighted racial/ethnic disparities in the severity of COVID-19 seen across the USA; the extent to which these disparities have persisted over time remains unclear. Our research objective was to understand temporal trends in racial/ethnic variation in severity of COVID-19 illness presenting over time.

Methods

We conducted a retrospective cohort analysis using longitudinal data from Cedars-Sinai Medical Center, a high-volume health system in Southern California. We studied patients admitted to the hospital with COVID-19 illness from 4 March 2020 through 5 December 2020. Our primary outcome was COVID-19 severity of illness among hospitalised patients, assessed by racial/ethnic group status. We defined overall illness severity as an ordinal outcome: hospitalisation but no intensive care unit (ICU) admission; admission to the ICU but no intubation; and intubation or death.

Results

A total of 1584 patients with COVID-19 with available demographic and clinical data were included. Hispanic/Latinx compared with non-Hispanic white patients had higher odds of experiencing more severe illness among hospitalised patients (OR 2.28, 95% CI 1.62 to 3.22) and this disparity persisted over time. During the initial 2 months of the pandemic, non-Hispanic blacks were more likely to suffer severe illness than non-Hispanic whites (OR 2.02, 95% CI 1.07 to 3.78); this disparity improved by May, only to return later in the pandemic.

Conclusion

In our patient sample, the severity of observed COVID-19 illness declined steadily over time, but these clinical improvements were not seen evenly across racial/ethnic groups; greater illness severity continues to be experienced among Hispanic/Latinx patients.

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.

Seroprevalence of antibodies to SARS-CoV-2 in healthcare workers: a cross-sectional study

OBJECTIVE

We sought to determine the extent of SARS-CoV-2 seroprevalence and the factors associated with seroprevalence across a diverse cohort of healthcare workers.

DESIGN

Observational cohort study of healthcare workers, including SARS-CoV-2 serology testing and participant questionnaires.

SETTINGS

A multisite healthcare delivery system located in Los Angeles County.

PARTICIPANTS

A diverse and unselected population of adults (n=6062) employed in a multisite healthcare delivery system located in Los Angeles County, including individuals with direct patient contact and others with non-patient-oriented work functions.

MAIN OUTCOMES

Using Bayesian and multivariate analyses, we estimated seroprevalence and factors associated with seropositivity and antibody levels, including pre-existing demographic and clinical characteristics; potential COVID-19 illness-related exposures; and symptoms consistent with COVID-19 infection.

RESULTS

We observed a seroprevalence rate of 4.1%, with anosmia as the most prominently associated self-reported symptom (OR 11.04, p<0.001) in addition to fever (OR 2.02, p=0.002) and myalgias (OR 1.65, p=0.035). After adjusting for potential confounders, seroprevalence was also associated with Hispanic ethnicity (OR 1.98, p=0.001) and African-American race (OR 2.02, p=0.027) as well as contact with a COVID-19-diagnosed individual in the household (OR 5.73, p<0.001) or clinical work setting (OR 1.76, p=0.002). Importantly, African-American race and Hispanic ethnicity were associated with antibody positivity even after adjusting for personal COVID-19 diagnosis status, suggesting the contribution of unmeasured structural or societal factors.

CONCLUSION AND RELEVANCE

The demographic factors associated with SARS-CoV-2 seroprevalence among our healthcare workers underscore the importance of exposure sources beyond the workplace. The size and diversity of our study population, combined with robust survey and modelling techniques, provide a vibrant picture of the demographic factors, exposures and symptoms that can identify individuals with susceptibility as well as potential to mount an immune response to COVID-19.

Left ventricular global longitudinal strain in identifying subclinical myocardial dysfunction among patients hospitalized with COVID-19

Background

The incidence of acute cardiac injury in COVID-19 patients is very often subclinical and can be detected by cardiac magnetic resonance imaging. The aim of this study was to assess if subclinical myocardial dysfunction could be identified using left ventricular global longitudinal strain (LV-GLS) in patients hospitalized with COVID-19.

Methods

We performed a search of COVID-19 patients admitted to our institution from January 1st, 2020 to June 8th, 2020, which revealed 589 patients (mean age = 66 ± 18, male = 56%). All available 60 transthoracic echocardiograms (TTE) were reviewed and off-line assessment of LV-GLS was performed in 40 studies that had sufficient quality images and the views required to calculate LV-GLS. We also analyzed electrocardiograms and laboratory findings including inflammatory markers, Troponin-I, and B-type natriuretic peptide (BNP).

Results

Of 589 patients admitted with COVID-19 during our study period, 60 (10.1%) underwent TTE during hospitalization. Findings consistent with overt myocardial involvement included reduced ejection fraction (23%), wall motion abnormalities (22%), low stroke volume (82%) and increased LV wall thickness (45%). LV-GLS analysis was available for 40 patients and was abnormal in 32 patients (80%). All patients with LV dysfunction had elevated cardiac enzymes and positive inflammatory biomarkers.

Conclusions

Subclinical myocardial dysfunction as measured via reduced LV-GLS is frequent, occurring in 80% of patients hospitalized with COVID-19, while prevalent LV function parameters such as reduced EF and wall motion abnormalities were less frequent findings. The mechanism of cardiac injury in COVID-19 infection is the subject of ongoing research.

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.

Pre-existing traits associated with Covid-19 illness severity

Importance

Certain individuals, when infected by SARS-CoV-2, tend to develop the more severe forms of Covid-19 illness for reasons that remain unclear.

Objective

To determine the demographic and clinical characteristics associated with increased severity of Covid-19 infection.

Design

Retrospective observational study. We curated data from the electronic health record, and used multivariable logistic regression to examine the association of pre-existing traits with a Covid-19 illness severity defined by level of required care: need for hospital admission, need for intensive care, and need for intubation.

Setting

A large, multihospital healthcare system in Southern California.

Participants

All patients with confirmed Covid-19 infection (N = 442).

Results

Of all patients studied, 48% required hospitalization, 17% required intensive care, and 12% required intubation. In multivariable-adjusted analyses, patients requiring a higher levels of care were more likely to be older (OR 1.5 per 10 years, P<0.001), male (OR 2.0, P = 0.001), African American (OR 2.1, P = 0.011), obese (OR 2.0, P = 0.021), with diabetes mellitus (OR 1.8, P = 0.037), and with a higher comorbidity index (OR 1.8 per SD, P<0.001). Several clinical associations were more pronounced in younger compared to older patients (P_interaction<0.05). Of all hospitalized patients, males required higher levels of care (OR 2.5, P = 0.003) irrespective of age, race, or morbidity profile.

Conclusions and relevance

In our healthcare system, greater Covid-19 illness severity is seen in patients who are older, male, African American, obese, with diabetes, and with greater overall comorbidity burden. Certain comorbidities paradoxically augment risk to a greater extent in younger patients. In hospitalized patients, male sex is the main determinant of needing more intensive care. Further investigation is needed to understand the mechanisms underlying these findings.