Opportunistic Screening of Chronic Liver Disease with Deep Learning Enhanced Echocardiography

Importance

Chronic liver disease affects more than 1.5 billion adults worldwide, however the majority of cases are asymptomatic and undiagnosed. Echocardiography is broadly performed and visualizes the liver; but this information is not leveraged.

Objective

To develop and evaluate a deep learning algorithm on echocardiography videos to enable opportunistic screening for chronic liver disease.

Design

Retrospective observational cohorts.

Setting

Two large urban academic medical centers.

Participants

Adult patients who received echocardiography and abdominal imaging (either abdominal ultrasound or abdominal magnetic resonance imaging) with ≤30 days between tests, between July 4, 2012, to June 4, 2022.

Exposure

Deep learning model predictions from a deep-learning computer vision pipeline that identifies subcostal view echocardiogram videos and detects the presence of cirrhosis or steatotic liver disease (SLD).

Main Outcome and Measures

Clinical diagnosis by paired abdominal ultrasound or magnetic resonance imaging (MRI).

Results

A total of 1,596,640 echocardiogram videos (66,922 studies from 24,276 patients) from Cedars-Sinai Medical Center (CSMC) were used to develop EchoNet-Liver, an automated pipeline that identifies high quality subcostal images from echocardiogram studies and detects the presence of cirrhosis or SLD. In the held-out CSMC test cohort, EchoNet-Liver was able to detect the presence of cirrhosis with an AUC of 0.837 (0.789 - 0.880) and SLD with an AUC of 0.799 (0.758 - 0.837). In a separate test cohort with paired abdominal MRIs, cirrhosis was detected with an AUC of 0.704 (0.689-0.718) and SLD was detected with an AUC of 0.726 (0.659-0.790). In an external test cohort of 106 patients (n = 5,280 videos), the model detected cirrhosis with an AUC of 0.830 (0.738 - 0.909) and SLD with an AUC of 0.768 (0.652 - 0.875).

Conclusions and Relevance

Deep learning assessment of clinical echocardiography enables opportunistic screening of SLD and cirrhosis. Application of this algorithm may identify patients who may benefit from further diagnostic testing and treatment for chronic liver disease.

Primary care usage at the end of life: a retrospective cohort study of cancer patients using linked primary and hospital care data

PURPOSE

Health service use is most intensive in the final year of a person's life, with 80% of this expenditure occurring in hospital. Close involvement of primary care services has been promoted to enhance quality end-of-life care that is appropriate to the needs of patients. However, the relationship between primary care involvement and patients' use of hospital care is not well described. This study aims to examine primary care use in the last year of life for cancer patients and its relationship to hospital usage.

METHODS

Retrospective cohort study in Victoria, Australia, using linked routine care data from primary care, hospital and death certificates. Patients were included who died related to cancer between 2008 and 2017.

RESULTS

A total of 758 patients were included, of whom 88% (n = 667) visited primary care during the last 6 months (median 9.1 consultations). In the last month of life, 45% of patients were prescribed opioids, and 3% had imaging requested. Patients who received home visits (13%) or anticipatory medications (15%) had less than half the median bed days in the last 3 months (4 vs 9 days, p < 0.001, 5 vs 10 days, p = 0.001) and 1 month of life (0 vs 2 days, p = 0.002, 0 vs 3 days, p < 0.001), and reduced emergency department presentations (32% vs 46%, p = 0.006, 31% vs 47% p < 0.001) in the final month.

CONCLUSION

This study identifies two important primary care processes-home visits and anticipatory medication-associated with reduced hospital usage and intervention at the end of life.

Deep learning for transesophageal echocardiography view classification

Transesophageal echocardiography (TEE) imaging is a vital tool used in the evaluation of complex cardiac pathology and the management of cardiac surgery patients. A key limitation to the application of deep learning strategies to intraoperative and intraprocedural TEE data is the complexity and unstructured nature of these images. In the present study, we developed a deep learning-based, multi-category TEE view classification model that can be used to add structure to intraoperative and intraprocedural TEE imaging data. More specifically, we trained a convolutional neural network (CNN) to predict standardized TEE views using labeled intraoperative and intraprocedural TEE videos from Cedars-Sinai Medical Center (CSMC). We externally validated our model on intraoperative TEE videos from Stanford University Medical Center (SUMC). Accuracy of our model was high across all labeled views. The highest performance was achieved for the Trans-Gastric Left Ventricular Short Axis View (area under the receiver operating curve [AUC] = 0.971 at CSMC, 0.957 at SUMC), the Mid-Esophageal Long Axis View (AUC = 0.954 at CSMC, 0.905 at SUMC), the Mid-Esophageal Aortic Valve Short Axis View (AUC = 0.946 at CSMC, 0.898 at SUMC), and the Mid-Esophageal 4-Chamber View (AUC = 0.939 at CSMC, 0.902 at SUMC). Ultimately, we demonstrate that our deep learning model can accurately classify standardized TEE views, which will facilitate further downstream deep learning analyses for intraoperative and intraprocedural TEE imaging.

Aubin M, Nilles EJ, Musk ER, Menon AS, Saphire EO, Krammer F, Lauffenburger DA, Barouch DH, Alter G. mRNA-1273 and BNT162b2 COVID-19 vaccines elicit antibodies with differences in Fc-mediated effector functions

The successful development of several coronavirus disease 2019 (COVID-19) vaccines has substantially reduced morbidity and mortality in regions of the world where the vaccines have been deployed. However, in the wake of the emergence of viral variants that are able to evade vaccine-induced neutralizing antibodies, real-world vaccine efficacy has begun to show differences across the two approved mRNA platforms, BNT162b2 and mRNA-1273; these findings suggest that subtle variation in immune responses induced by the BNT162b2 and mRNA-1273 vaccines may confer differential protection. Given our emerging appreciation for the importance of additional antibody functions beyond neutralization, we profiled the postboost binding and functional capacity of humoral immune responses induced by the BNT162b2 and mRNA-1273 vaccines in a cohort of hospital staff. Both vaccines induced robust humoral immune responses to wild-type severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and to variants of concern. However, differences emerged across epitope-specific responses, with higher concentrations of receptor binding domain (RBD)- and N-terminal domain-specific IgA observed in recipients of mRNA-1273. Antibodies eliciting neutrophil phagocytosis and natural killer cell activation were also increased in mRNA-1273 vaccine recipients as compared to BNT162b2 recipients. RBD-specific antibody depletion highlighted the different roles of non-RBD-specific antibody effector functions induced across the mRNA vaccines. These data provide insights into potential differences in protective immunity conferred by these vaccines.

SARS-CoV-2 antibodies protect against reinfection for at least 6 months in a multicentre seroepidemiological workplace cohort

Identifying the potential for SARS-CoV-2 reinfection is crucial for understanding possible long-term epidemic dynamics. We analysed longitudinal PCR and serological testing data from a prospective cohort of 4,411 United States employees in 4 states between April 2020 and February 2021. We conducted a multivariable logistic regression investigating the association between baseline serological status and subsequent PCR test result in order to calculate an odds ratio for reinfection. We estimated an odds ratio for reinfection ranging from 0.14 (95% CI: 0.019 to 0.63) to 0.28 (95% CI: 0.05 to 1.1), implying that the presence of SARS-CoV-2 antibodies at baseline is associated with around 72% to 86% reduced odds of a subsequent PCR positive test based on our point estimates. This suggests that primary infection with SARS-CoV-2 provides protection against reinfection in the majority of individuals, at least over a 6-month time period. We also highlight 2 major sources of bias and uncertainty to be considered when estimating the relative risk of reinfection, confounders and the choice of baseline time point, and show how to account for both in reinfection analysis.

Identifying the potential for SARS-CoV-2 reinfection is crucial for understanding possible long-term epidemic dynamics. Analysis of a seroepidemiological cohort suggests that primary infection with SARS-CoV-2 protects against reinfection in the majority of individuals, at least over a six month period.

A decision support tool for the detection of pancreatic cancer in general practice: A modified Delphi consensus

BACKGROUND/OBJECTIVES

Diagnosis of pancreatic cancer is often delayed, contributing to patient and family distress and leading to worse survival. We aimed to develop a decision support tool to support primary care providers to identify patients that should undergo investigations for pancreatic cancer, and to recommend initial diagnostic pathways.

METHODS

A modified Delphi process, including a series of three surveys, was undertaken to ascertain clinical expert opinion on which combinations of signs, symptoms and risk factors should be included in a tool for the early identification of pancreatic cancer. A group of clinical specialists finalised the development of the tool during a focus group meeting.

RESULTS

The tool presents individual or combinations of signs, symptoms, and risk factors in three tiers which direct the urgency of investigation. Tier 1 includes 5 clinical presentation and risk factors clusters that indicate the need for urgent investigation of the pancreas. A further five clusters are included as Tier 2 aiming to elimate other causes and reduce the time to investigating the pancreas. Tier 3 includes a list of non-specific signs, symptoms and risk factors that indicate the need to consider pancreatic cancer as a potential diagnosis, but without specific recommendations for investigation.

CONCLUSIONS

Prospective validation studies are now required prior to implementation in the primary care setting. Implementation into primary care practice and as an educational resource may facilitate rapid diagnosis and improve outcomes such as distress and survival.

Aubin M, Zhou G, Gluck MJ, Berger S, Rhee J, Petersen E, Mormann B, Loesche M, Hu Y, Chen Z, Yu J, Gebre M, Atyeo C, Gorman MJ, Zhu AL, Burke J, Slein M, Hasdianda MA, Jambaulikar G, Boyer EW, Sabeti PC, Barouch DH, Julg B, Kucharski AJ, Musk ER, Lauffenburger DA, Alter G, Menon AS. Epidemiological and Immunological Features of Obesity and SARS-CoV-2

Obesity is a key correlate of severe SARS-CoV-2 outcomes while the role of obesity on risk of SARS-CoV-2 infection, symptom phenotype, and immune response remain poorly defined. We examined data from a prospective SARS-CoV-2 cohort study to address these questions. Serostatus, body mass index, demographics, comorbidities, and prior COVID-19 compatible symptoms were assessed at baseline and serostatus and symptoms monthly thereafter. SARS-CoV-2 immunoassays included an IgG ELISA targeting the spike RBD, multiarray Luminex targeting 20 viral antigens, pseudovirus neutralization, and T cell ELISPOT assays. Our results from a large prospective SARS-CoV-2 cohort study indicate symptom phenotype is strongly influenced by obesity among younger but not older age groups; we did not identify evidence to suggest obese individuals are at higher risk of SARS-CoV-2 infection; and remarkably homogenous immune activity across BMI categories suggests immune protection across these groups may be similar.

Aubin M, Nilles EJ, Musk ER, Menon AS, Saphire EO, Krammer F, Lauffenburger DA, Barouch DH, Alter G. Subtle immunological differences in mRNA-1273 and BNT162b2 COVID-19 vaccine induced Fc-functional profiles

The successful development of several COVID-19 vaccines has substantially reduced morbidity and mortality in regions of the world where the vaccines have been deployed. However, in the wake of the emergence of viral variants, able to evade vaccine induced neutralizing antibodies, real world vaccine efficacy has begun to show differences across the mRNA platforms, suggesting that subtle variation in immune responses induced by the BNT162b2 and mRNA1273 vaccines may provide differential protection. Given our emerging appreciation for the importance of additional antibody functions, beyond neutralization, here we profiled the postboost binding and functional capacity of the humoral response induced by the BNT162b2 and mRNA-1273 in a cohort of hospital staff. Both vaccines induced robust humoral immune responses to WT SARS-CoV-2 and VOCs. However, differences emerged across epitopespecific responses, with higher RBD- and NTD-specific IgA, as well as functional antibodies (ADNP and ADNK) in mRNA-1273 vaccine recipients. Additionally, RBD-specific antibody depletion highlighted the different roles of non-RBD-specific antibody effector function induced across the mRNA vaccines, providing novel insights into potential differences in protective immunity generated across these vaccines in the setting of newly emerging VOCs.

418 The Management of Distal Radius Fractures in Adults with A Removable Soft Cast During the Peak Of The Covid-19 Pandemic: Patient Reported Outcomes And Experiences

Introduction

Immobilisation tools for non-operative management of distal radius fractures (DRFs) in adults, include plaster casts, orthotic splints, and recently removable soft cast as per the BOAST COVID-19 guidelines. Limited evidence exists on the effectiveness of removable soft casts in adults with DRFs. We assessed patient reported functional outcomes and experiences of adults with DRFs managed non-operatively in a removable soft cast during the COVID-19 pandemic.

Method

We retrospectively reviewed adults with DRFs, treated with a removable soft cast for 6 weeks between 9^th April 2020 and 18^th May 2020. Functional scores - Disabilities of the Arms, Shoulders and Hands (DASH) and a 7-question Patient Reported Experience Questionnaire (PREQ) were conducted at 6 weeks, along with a record of complications. A DASH score of < 25 was deemed a satisfactory functional score.

Results

34 patients formed our cohort, with a 65% (22patients) completion rate, mean age of 69years. 82% had a DASH score < 25 and 18% had a DASH score > 25 at 6 weeks. 86% reported satisfactory comfort in the PREQ with 14% reporting discomfort. Two patients re-attended hospital with soft cast related problems.

Conclusions

Most adults with DRFs managed in a removable soft cast reported good early functional outcomes, satisfactory experiences, and no clinic visits at 6weeks.

Discrete SARS-CoV-2 antibody titers track with functional humoral stability

Antibodies serve as biomarkers of infection, but if sustained can confer long-term immunity. Yet, for most clinically approved vaccines, binding antibody titers only serve as a surrogate of protection. Instead, the ability of vaccine induced antibodies to neutralize or mediate Fc-effector functions is mechanistically linked to protection. While evidence has begun to point to persisting antibody responses among SARS-CoV-2 infected individuals, cases of re-infection have begun to emerge, calling the protective nature of humoral immunity against this highly infectious pathogen into question. Using a community-based surveillance study, we aimed to define the relationship between titers and functional antibody activity to SARS-CoV-2 over time. Here we report significant heterogeneity, but limited decay, across antibody titers amongst 120 identified seroconverters, most of whom had asymptomatic infection. Notably, neutralization, Fc-function, and SARS-CoV-2 specific T cell responses were only observed in subjects that elicited RBD-specific antibody titers above a threshold. The findings point to a switch-like relationship between observed antibody titer and function, where a distinct threshold of activity-defined by the level of antibodies-is required to elicit vigorous humoral and cellular response. This response activity level may be essential for durable protection, potentially explaining why re-infections occur with SARS-CoV-2 and other common coronaviruses.

Epidemiological and immunological features of obesity and SARS-CoV-2

Obesity is a key correlate of severe SARS-CoV-2 outcomes while the role of obesity on risk of SARS-CoV-2 infection, symptom phenotype, and immune response are poorly defined. We examined data from a prospective SARS-CoV-2 cohort study to address these questions. Serostatus, body mass index, demographics, comorbidities, and prior COVID-19 compatible symptoms were assessed at baseline and serostatus and symptoms monthly thereafter. SARS-CoV-2 immunoassays included an IgG ELISA targeting the spike RBD, multiarray Luminex targeting 20 viral antigens, pseudovirus neutralization, and T cell ELISPOT assays. Our results from a large prospective SARS-CoV-2 cohort study indicate symptom phenotype is strongly influenced by obesity among younger but not older age groups; we did not identify evidence to suggest obese individuals are at higher risk of SARS-CoV-2 infection; and, remarkably homogenous immune activity across BMI categories suggests natural- and vaccine-induced protection may be similar across these groups.

Association of blood mercury levels with nonmelanoma skin cancer in the U.S.A. using National Health and Nutrition Examination Survey data (2003-2016)

BACKGROUND

Some studies have reported increased incidence or mortality of lung and brain cancers associated with occupations involving potential mercury exposure. Epidemiological evidence related to skin cancer is also limited.

OBJECTIVES

To investigate the association between blood mercury (Hg) levels and nonmelanoma skin cancer (NMSC).

METHODS

We used National Health and Nutrition Examination Survey data from 2003 to 2016. The exposures were blood total (tHg), inorganic (iHg) and methylmercury (MeHg). The outcome was a self-reported diagnosis of NMSC. We included participants aged ≥ 20 years who had information on blood mercury and sociodemographic factors. We conducted a logistic regression analysis to estimate odds ratios (ORs) and 95% confidence intervals (CIs) for the risk of NMSC associated with quartiles of blood Hg, after adjusting for the sociodemographic factors and survey year.

RESULTS

The number of participants was 29 413; mean age was 49 years and 52% were female. Compared with those with a tHg ≤ 0·47 μg L^-1 (Q1), those with a tHg > 1·74 μg L^-1 (Q4) had nearly double the odds of NMSC (OR 1·79, 95% CI 1·19-2·71; P_trend = 0·004). Similarly, those in the highest quartile of MeHg (> 1·44 μg L^-1 ) had 1·7 times greater odds of NMSC (OR 1·74, 95% CI 1·13-2·70; P_trend = 0·01) than those in the lowest quartile (≤ 0·21 μg L^-1 ). iHg levels were nonsignificantly positively associated with NMSC (P_trend = 0·08).

CONCLUSIONS

We found that higher blood tHg and MeHg levels were associated with a higher prevalence of NMSC. Linked Comment: Taylor. Br J Dermatol 2020; 183:413-414.

Upregulation of Atrogin-1/FBXO32 is not necessary for cartilage destruction in mouse models of osteoarthritis

OBJECTIVE

In a preliminary study, we found that recently identified catabolic regulators of osteoarthritis (OA), including hypoxia-inducible factor (HIF)-2α and members of the zinc-ZIP8-MTF1 axis, upregulate the E3 ubiquitin ligase, Atrogin-1 (encoded by Fbxo32), in chondrocytes. As the ubiquitination/proteasomal degradation pathways are tightly regulated to modulate the expression of catabolic factors in chondrocytes, we examined the in vivo functions of Atrogin-1 in mouse models of OA.

METHODS

The mRNA and protein levels of Atrogin-1 and other regulators of OA were determined in primary cultured mouse chondrocytes, OA human cartilage, and OA cartilage from wild-type (WT) and Fbxo32-knockout (KO) mice subjected to destabilization of the medial meniscus or intra-articular (IA) injection of adenoviruses expressing HIF-2α (Ad-Epas1), ZIP8 (Ad-Zip8), or Atrogin-1 (Ad-Fbxo32). The effect of Atrogin-1 overexpression on the cartilage of WT mice was examined by IA injection of Ad-Fbxo32.

RESULTS

Atrogin-1 mRNA levels in chondrocytes were markedly increased by treatment with interleukin-1β, HIF-2α, and members of the zinc-ZIP8-MTF1 axis. Atrogin-1 protein levels were also increased in OA cartilage from humans and various mouse OA models. However, the forced overexpression of Atrogin-1 in chondrocytes did not modulate the expression of cartilage matrix molecules or matrix-degrading enzymes. Moreover, overexpression of Atrogin-1 in the mouse joint tissues failed to cause OA pathogenesis, and Fbxo32 knockout failed to affect post-traumatic OA cartilage destruction in mice.

CONCLUSIONS

Although Atrogin-1 is upregulated in OA cartilage, overexpression of Atrogin-1 in the joint tissues or knockout of Fbxo32 does not affect OA cartilage destruction in mice.