Shortcut learning in medical AI hinders generalization: method for estimating AI model generalization without external data

Healthcare datasets are becoming larger and more complex, necessitating the development of accurate and generalizable AI models for medical applications. Unstructured datasets, including medical imaging, electrocardiograms, and natural language data, are gaining attention with advancements in deep convolutional neural networks and large language models. However, estimating the generalizability of these models to new healthcare settings without extensive validation on external data remains challenging. In experiments across 13 datasets including X-rays, CTs, ECGs, clinical discharge summaries, and lung auscultation data, our results demonstrate that model performance is frequently overestimated by up to 20% on average due to shortcut learning of hidden data acquisition biases (DAB). Shortcut learning refers to a phenomenon in which an AI model learns to solve a task based on spurious correlations present in the data as opposed to features directly related to the task itself. We propose an open source, bias-corrected external accuracy estimate, PEst, that better estimates external accuracy to within 4% on average by measuring and calibrating for DAB-induced shortcut learning.

HLA sensitization is associated with an increased risk of primary graft dysfunction after heart transplantation

Primary graft dysfunction (PGD) is a leading cause of early morbidity and mortality following heart transplantation (HT). We sought to determine the association between pretransplant human leukocyte antigen (HLA) sensitization, as measured using the calculated panel reactive antibody (cPRA) value, and the risk of PGD.

METHODS

Consecutive adult HT recipients (n = 596) from 1/2015 to 12/2019 at 2 US centers were included. Severity of PGD was based on the 2014 International Society for Heart and Lung Transplantation consensus statement. For each recipient, unacceptable HLA antigens were obtained and locus-specific cPRA (cPRA-LS) and pre-HT donor-specific antibodies (DSA) were assessed.

RESULTS

Univariable logistic modeling showed that peak cPRA-LS for all loci and HLA-A was associated with increased severity of PGD as an ordinal variable (all loci: OR 1.78, 95% CI: 1.01-1.14, p = 0.025, HLA-A: OR 1.14, 95% CI: 1.03-1.26, p = 0.011). Multivariable analysis showed peak cPRA-LS for HLA-A, recipient beta-blocker use, total ischemic time, donor age, prior cardiac surgery, and United Network for Organ Sharing status 1 or 2 were associated with increased severity of PGD. The presence of DSA to HLA-B was associated with trend toward increased risk of mild-to-moderate PGD (OR 2.56, 95% CI: 0.99-6.63, p = 0.053), but DSA to other HLA loci was not associated with PGD.

CONCLUSIONS

Sensitization for all HLA loci, and specifically HLA-A, is associated with an increased severity of PGD. These factors should be included in pre-HT risk stratification to minimize the risk of PGD.

Enhancing the Prediction of Cardiac Allograft Vasculopathy Using Intravascular Ultrasound and Machine Learning: A Proof of Concept

BACKGROUND

Cardiac allograft vasculopathy (CAV) is the leading cause of late graft dysfunction in heart transplantation. Building on previous unsupervised learning models, we sought to identify CAV clusters using serial maximal intimal thickness and baseline clinical risk factors to predict the development of early CAV.

METHODS

This is a single-center retrospective study including adult heart transplantation recipients. A latent class mixed-effects model was used to identify patient clusters with similar trajectories of maximal intimal thickness posttransplant and pretransplant covariates associated with each cluster.

RESULTS

Among 186 heart transplantation recipients, we identified 4 patient phenotypes: very low, low, moderate, and high risk. The 5-year risk (95% CI) of the International Society for Heart and Lung Transplantation-defined CAV in the high, moderate, low, and very low risk groups was 49.1% (35.2%-68.5%), 23.4% (13.3%-41.2%), 5.0% (1.3%-19.6%), and 0%, respectively. Only patients in the moderate to high risk cluster developed the International Society for Heart and Lung Transplantation CAV 2-3 at 5 years (P=0.02). Of the 4 groups, the low risk group had significantly younger female recipients, shorter ischemic time, and younger female donors compared with the high risk group.

CONCLUSIONS

We identified 4 clusters characterized by distinct maximal intimal thickness trajectories. These clusters were shown to discriminate against the development of angiographic CAV. This approach allows for the personalization of surveillance and CAV-directed treatment before the development of angiographically apparent disease.

Cardiac allograft vasculopathy and survival in pediatric heart transplant recipients transitioned to adult care

BACKGROUND

Cardiac allograft vasculopathy (CAV) is an important cause of mortality after pediatric heart transplantation (HT) but there is a paucity of data regarding its incidence and impact on survival in pediatric recipients transitioned to adult care.

METHODS

We conducted a retrospective review of consecutive pediatric HT patients from 1989 to 2017 at the Hospital for Sick Children who transitioned to adult care at ≥18 years at Toronto General Hospital. We evaluated the incidence of International Society of Heart and Lung Transplantation CAV grade ≥1 using competing risk models. We assessed the association between all-cause mortality and CAV using Cox proportional hazards and used Kaplan Meier methods to evaluate all-cause mortality stratified by CAV and transplant era (1989-2001, 2002-2017).

RESULTS

Ninety-six patients were transitioned to adult care by January 2022, of which 53 underwent repeat coronary angiography as adults. CAV was newly diagnosed in 49% patients after transition to adult care. The overall incidence of CAV was 3.9 cases per 100 person-years. There was no difference in the adjusted incidence of CAV according to transplant era (subdistribution hazard ratios = 1.17, 95% confidence interval (CI) 0.54-2.66). CAV was associated with a higher risk of death in the early era (hazard ratio (HR) 10.29, 95% CI 2.16-49.96), but not in the recent era (HR 1.61, 95% 0.35-7.47).

CONCLUSIONS

There is a role for continued CAV surveillance after the transition to adult care. The implications of diagnosing CAV after the transition to adult care require further study, particularly because the risk of death in pediatric HT recipients diagnosed with CAV in the more recent era may be attenuated compared to the earlier HT era.

Cardiac MRI and Clinical Outcomes in TMEM43 Arrhythmogenic Cardiomyopathy

Arrhythmogenic cardiomyopathy is an inherited cardiomyopathy that can involve both ventricles. Several genes have been identified as pathogenic in arrhythmogenic cardiomyopathy, including TMEM43. However, there are limited data on cardiac MRI findings in patients with TMEM43 variants to date. In this case series, cardiac MRI findings and clinical outcomes are described in 14 patients with TMEM43 variants, including eight (57%) with the pathogenic p.Ser358Leu variant (six female patients; mean age, 33 years ± 15 [SD]) and six (43%) with a TMEM43 variant of unknown significance (three female patients; mean age, 38 years ± 11). MRI findings demonstrated left ventricular systolic dysfunction in eight (57%) patients and right ventricular dysfunction in four (29%) patients. Among the nine patients with late gadolinium enhancement imaging, left ventricular late gadolinium enhancement was present in seven (78%; all subepicardial) patients. In summary, TMEM43 variants are associated with high prevalence of subepicardial late gadolinium enhancement and left ventricular dysfunction. Keywords: Arrhythmogenic Cardiomyopathy, Arrhythmogenic Right Ventricular Cardiomyopathy, TMEM43, Cardiac MRI, Genetic Variants Supplemental material is available for this article.

The International Consortium on Primary Graft Dysfunction: Redefining Clinical Risk Factors in the Contemporary Era of Heart Transplantation

BACKGROUND

Primary Graft Dysfunction (PGD) is the leading cause of morbidity and mortality early after heart transplant (HT). The International Consortium on PGD is a multicenter collaboration dedicated to identifying the clinical risk factors for PGD in the contemporary era of HT. The objectives of the current report were to 1) assess the incidence of severe PGD in an international cohort, 2) evaluate the performance of the most validated PGD risk tool, the RADIAL score, in a contemporary cohort, and 3) redefine clinical risk factors for severe PGD in the current era of HT.

METHODS

This is a retrospective, observational study of consecutive adult HT recipients between 2010 and 2020 in 10 centers in the United States, Canada, and Europe. Patients with severe PGD were compared to those without severe PGD (comprising those with no, mild and moderate PGD). The RADIAL score was calculated for each transplant recipient. The discriminatory power of the RADIAL score was evaluated using receiver operating characteristic (ROC) analysis and its calibration was assessed by plotting the percentage of PGD predicted versus observed. To identify clinical risk factors associated with severe PGD, we performed multivariable mixed-effects logistic regression modeling to account for among-center variability.

RESULTS

A total of 2,746 patients have been enrolled in the registry to date, including 2,015 (73.4%) from North America, and 731 (26.6%) from Europe. 215 participants (7.8%) met the criteria for severe PGD. There was an increase in the incidence of severe PGD over the study period (p-value for trend by difference sign test = 0.004). The Kaplan Meier estimate for 1-year survival was 75.7% [95%CI 69.4-80.9%] in patients with severe PGD as compared to 94.4% [95% CI 93.5-95.2%] in those without severe PGD (log-rank p-value <0.001). The RADIAL score performed poorly in our contemporary cohort and was not associated with severe PGD with an AUC of 0.53 (95%CI 0.48-0.58). In the multivariable regression model, acute preoperative dialysis (OR 2.41, 95% CI 1.31 - 4.43), durable LVAD support (OR 1.77, 95% CI 1.13 - 2.77), and total ischemic time (OR 1.20 for each additional hour, 95% CI 1.02 - 1.41) were associated with an increased risk of severe PGD.

CONCLUSIONS

Our consortium has identified an increasing incidence of PGD in the modern transplant era. We identified contemporary risk factors for this early post-transplant complication, which confers a high mortality risk. These results may enable the identification of patients at high risk for developing severe PGD in order to inform peri-transplant donor and recipient management practices.

Tricuspid regurgitation, right ventricular function, and renal congestion: a cardiorenal triangle

There is a growing interest in the evaluation of tricuspid regurgitation due to its increasing prevalence and detrimental impact on clinical outcomes. Historically, it has been coined the "forgotten" defect in the field of valvular heart disease due to the lack of effective treatments to improve prognosis. However, the development of percutaneous treatment techniques has led to a new era in its management, with promising results and diminished complication risk. In spite of these advances, a comprehensive exploration of the pathophysiological mechanisms is essential to establish clear indications and optimal timing for medical and percutaneous intervention. This review will address the most important aspects related to the diagnosis, pathophysiology and treatment of tricuspid regurgitation from a cardiorenal perspective, with a special emphasis on the interaction between right ventricular dysfunction and the development of hepatorenal congestion.

Predicted Heart Mass: A Tale of 2 Ventricles

BACKGROUND

Total predicted heart mass (PHM) is the recommended metric to assess donor-recipient size matching in patients undergoing heart transplantation. Separately measuring right ventricular (RV) and left ventricular (LV) PHM may improve risk prediction of 1-year graft failure.

METHODS

Adult heart transplant recipients from the UNOS database from 2000 to 2018 were included in the study. LV and RV PHM were modeled as restricted cubic splines. The association with 1-year graft failure was determined using adjusted Cox regression. The risk reclassification of using both LV and RV PHM versus total PHM was assessed using the net reclassification index.

RESULTS

A total of 34 976 recipients were included. We observed a U-shaped association between total PHM and 1-year graft failure, such that risk increased for hearts undersized by >15% and those oversized by more than 27%. Graft failure incrementally increased when LV PHM was undersized by more than 5% and when RV was oversized by >20%. There was 1.5-fold greater risk of graft failure for an LV undersized by >26% or an RV oversized by more than 40%. Using LV and RV PHM risk-assessment separately led to a net reclassification index=8.5% ([95% CI, 5.3%-11.7%], nonevent net reclassification index=9.1%, event net reclassification index=-0.6%).

CONCLUSIONS

The association between donor-recipient PHM match and the risk of graft failure after heart transplantation can be further understood as risk attributable to LV undersizing and RV oversizing. Assessing LV and RV PHM separately instead of total PHM could further refine the methods used to match donors and recipients for heart transplantation, minimize the risk of 1-year graft failure, and increase the use of donor organs.

Primary Graft Dysfunction Is Associated With Development of Early Cardiac Allograft Vasculopathy, but Not Other Immune-mediated Complications, After Heart Transplantation

BACKGROUND

We investigated associations between primary graft dysfunction (PGD) and development of acute cellular rejection (ACR), de novo donor-specific antibodies (DSAs), and cardiac allograft vasculopathy (CAV) after heart transplantation (HT).

METHODS

A total of 381 consecutive adult HT patients from January 2015 to July 2020 at a single center were retrospectively analyzed. The primary outcome was incidence of treated ACR (International Society for Heart and Lung Transplantation grade 2R or 3R) and de novo DSA (mean fluorescence intensity >500) within 1 y post-HT. Secondary outcomes included median gene expression profiling score and donor-derived cell-free DNA level within 1 y and incidence of cardiac allograft vasculopathy (CAV) within 3 y post-HT.

RESULTS

When adjusted for death as a competing risk, the estimated cumulative incidence of ACR (PGD 0.13 versus no PGD 0.21; P  = 0.28), median gene expression profiling score (30 [interquartile range, 25-32] versus 30 [interquartile range, 25-33]; P  = 0.34), and median donor-derived cell-free DNA levels was similar in patients with and without PGD. After adjusting for death as a competing risk, estimated cumulative incidence of de novo DSA within 1 y post-HT in patients with PGD was similar to those without PGD (0.29 versus 0.26; P  = 0.10) with a similar DSA profile based on HLA loci. There was increased incidence of CAV in patients with PGD compared with patients without PGD (52.6% versus 24.8%; P  = 0.01) within the first 3 y post-HT.

CONCLUSIONS

During the first year after HT, patients with PGD had a similar incidence of ACR and development of de novo DSA, but a higher incidence of CAV when compared with patients without PGD.

Rethinking Donor and Recipient Risk Matching in Europe and North America: Using Heart Transplant Predictors of Donor and Recipient Risk

BACKGROUND

In Europe, there is greater acceptance of hearts from higher-risk donors for transplantation, whereas in North America, the donor heart discard rate is significantly higher. A Donor Utilization Score (DUS) was used to compare European and North American donor characteristics for recipients included in the International Society for Heart and Lung Transplantation registry from 2000 to 2018. DUS was further evaluated as an independent predictor for 1-year freedom from graft failure, after adjusting for recipient risk. Lastly, we assessed donor-recipient risk matching with the outcome of 1-year graft failure.

METHODS

DUS was applied to the International Society for Heart and Lung Transplantation cohort using meta-modeling. Posttransplant freedom from graft failure was summarized by Kaplan-Meier survival. Multivariable Cox proportional hazard regression was applied to quantify the effects of DUS and Index for Mortality Prediction After Cardiac Transplantation score on the 1-year risk of graft failure. We present 4 donor/recipient risk groups using the Kaplan-Meier method.

RESULTS

European centers accept significantly higher-risk donor hearts compared to North America. DUS 0.45 versus 0.54, P<0.005). DUS was an independent predictor for graft failure with an inverse linear relationship when adjusted for covariates (P<0.001). The Index for Mortality Prediction After Cardiac Transplantation score, a validated tool to assess recipient risk, was also independently associated with 1-year graft failure (P<0.001). In North America, 1-year graft failure was significantly associated with donor-recipient risk matching (log-rank P<0.001). One-year graft failure was highest with pairing of high-risk recipients and donors (13.1% [95% CI, 10.7%-13.9%]) and lowest among low-risk recipients and donors (7.4% [95% CI, 6.8%-8.0%]). Matching of low-risk recipients with high-risk donors was associated with significantly less graft failure (9.0% [95% CI, 8.3%-9.7%]) than high-risk recipients with low-risk donors (11.4% [95% CI, 10.7%-12.2%]) Conclusions: European heart transplantation centers are more likely to accept higher-risk donor hearts than North American centers. Acceptance of borderline-quality donor hearts for lower-risk recipients could improve donor heart utilization without compromising recipient survival.

Extracorporeal Membrane Oxygenation for Graft Dysfunction Early After Heart Transplantation: A Systematic Review and Meta-analysis

INTRODUCTION

Venoarterial extracorporeal membrane oxygenation (VA-ECMO) is a prevailing option for the management of severe early graft dysfunction. This systematic review and individual patient data (IPD) meta-analysis aims to evaluate (1) mortality, (2) rates of major complications, (3) prognostic factors, and (4) the effect of different VA-ECMO strategies on outcomes in adult heart transplant (HT) recipients supported with VA-ECMO.

METHODS AND RESULTS

We conducted a systematic search and included studies of adults (≥18 years) who received VA-ECMO during their index hospitalization after HT and reported on mortality at any timepoint. We pooled data using random effects models. To identify prognostic factors, we analysed IPD using mixed effects logistic regression. We assessed the certainty in the evidence using the GRADE framework. We included 49 observational studies of 1477 patients who received VA-ECMO after HT, of which 15 studies provided IPD for 448 patients. There were no differences in mortality estimates between IPD and non-IPD studies. The short-term (30-day/in-hospital) mortality estimate was 33% (moderate certainty, 95% confidence interval [CI] 28%-39%) and 1-year mortality estimate 50% (moderate certainty, 95% CI 43%-57%). Recipient age (odds ratio 1.02, 95% CI 1.01-1.04) and prior sternotomy (OR 1.57, 95% CI 0.99-2.49) are associated with increased short-term mortality. There is low certainty evidence that early intraoperative cannulation and peripheral cannulation reduce the risk of short-term death.

CONCLUSIONS

One-third of patients who receive VA-ECMO for early graft dysfunction do not survive 30 days or to hospital discharge, and one-half do not survive to 1 year after HT. Improving outcomes will require ongoing research focused on optimizing VA-ECMO strategies and care in the first year after HT.

Combining donor derived cell free DNA and gene expression profiling for non-invasive surveillance after heart transplantation

BACKGROUND

Donor-derived cell free DNA (dd-cfDNA) and gene expression profiling (GEP) offer noninvasive alternatives to rejection surveillance after heart transplantation; however, there is little evidence on the paired use of GEP and dd-cfDNA for rejection surveillance.

METHODS

A single center, retrospective analysis of adult heart transplant recipients. A GEP cohort, transplanted from January 1, 2015 through December 31, 2017 and eligible for rejection surveillance with GEP was compared to a paired testing cohort, transplanted July 1, 2018 through June 30, 2020, with surveillance from both dd-cfDNA and GEP. The primary outcomes were survival and rejection-free survival at 1 year post-transplant.

RESULTS

In total 159 patients were included, 95 in the GEP and 64 in the paired testing group. There were no differences in baseline characteristics, except for less use of induction in the paired testing group (65.6%) compared to the GEP group (98.9%), P < .01. At 1-year, there were no differences between the paired testing and GEP groups in survival (98.4% vs. 94.7%, P = .23) or rejection-free survival (81.3% vs. 73.7% P = .28).

CONCLUSIONS

Compared to post-transplant rejection surveillance with GEP alone, pairing dd-cfDNA and GEP testing was associated with similar survival and rejection-free survival at 1 year while requiring significantly fewer biopsies.

Prognostic Value of Cardiac MRI and FDG PET in Cardiac Sarcoidosis: A Systematic Review and Meta-Analysis

Background There is no consensus regarding the relative prognostic value of cardiac MRI and fluorodeoxyglucose (FDG) PET in cardiac sarcoidosis. Purpose To perform a systematic review and meta-analysis of the prognostic value of cardiac MRI and FDG PET for major adverse cardiac events (MACE) in cardiac sarcoidosis. Materials and Methods In this systematic review, MEDLINE, Ovid Epub, CENTRAL, Embase, Emcare, and Scopus were searched from inception until January 2022. Studies that evaluated the prognostic value of cardiac MRI or FDG PET in adults with cardiac sarcoidosis were included. The primary outcome of MACE was assessed as a composite including death, ventricular arrhythmia, and heart failure hospitalization. Summary metrics were obtained using random-effects meta-analysis. Meta-regression was used to assess covariates. Risk of bias was assessed using the Quality in Prognostic Studies, or QUIPS, tool. Results Thirty-seven studies were included (3489 patients with mean follow-up of 3.1 years ± 1.5 [SD]); 29 studies evaluated MRI (2931 patients) and 17 evaluated FDG PET (1243 patients). Five studies directly compared MRI and PET in the same patients (276 patients). Left ventricular late gadolinium enhancement (LGE) at MRI and FDG uptake at PET were both predictive of MACE (odds ratio [OR], 8.0 [95% CI: 4.3, 15.0] [P < .001] and 2.1 [95% CI: 1.4, 3.2] [P < .001], respectively). At meta-regression, results varied by modality (P = .006). LGE (OR, 10.4 [95% CI: 3.5, 30.5]; P < .001) was also predictive of MACE when restricted to studies with direct comparison, whereas FDG uptake (OR, 1.9 [95% CI: 0.82, 4.4]; P = .13) was not. Right ventricular LGE and FDG uptake were also associated with MACE (OR, 13.1 [95% CI: 5.2, 33] [P < .001] and 4.1 [95% CI: 1.9, 8.9] [P < .001], respectively). Thirty-two studies were at risk for bias. Conclusion Left and right ventricular late gadolinium enhancement at cardiac MRI and fluorodeoxyglucose uptake at PET were predictive of major adverse cardiac events in cardiac sarcoidosis. Limitations include few studies with direct comparison and risk of bias. Systematic review registration no. CRD42021214776 (PROSPERO) © RSNA, 2023 Supplemental material is available for this article.

Getting ahead of the game: in-hospital initiation of HFrEF therapies

Hospitalizations for heart failure (HF) have become a global problem worldwide. Each episode of HF decompensation may lead to deleterious short- and long- term consequences, but on the other hand is an unique opportunity to adjust the heart failure pharmacotherapy. Thus, in-hospital and an early post-discharge period comprise an optimal timing for initiation and optimization of the comprehensive management of HF. This timeframe affords clinicians an opportunity to up titrate and adjust guideline-directed medical therapies (GDMT) to potentially mitigate poor outcomes associated post-discharge and longer-term. This review will cover this timely concept, present the data of utilization of GDMT in HF populations, discuss recent evidence for in-hospital initiation and up-titration of GDMT with a need for post-discharge follow-up and implementation this into clinical practice in patients with heart failure and reduced ejection fraction.

Predictors of early renal dysfunction after heart transplantation: a report from the International Consortium on Primary Graft Dysfunction

Background

Renal dysfunction is a common complication after heart transplantation (HT). Renal replacement therapy (RRT) after HT has been associated with increased risk of death. Long-term renal dysfunction is associated mainly to immunosuppressive therapy but is also strongly related to post-transplant renal failure. Predictors of early renal dysfunction after HT have not been clearly identified.

Purpose

We aimed to define predictors of early renal dysfunction after HT.

Methods

Our consortium includes 10 centers in the US, Canada and Europe. We collected data on all consecutive single-organ HT recipients from 2010 to 2020. The primary outcome was early renal dysfunction (ERD), defined as a composite of need for RRT or creatinine ≥2.5 mg/dL 24 hours after HT. We assessed the incidence of early renal dysfunction and performed univariate and multivariate analyses to identify the recipient and transplant characteristics associated with its development.

Results

We included 2,764 HT recipients: 282 (10.2%) presented early renal dysfunction and 2482 (89.8%) did not. Recipients who presented postoperative renal dysfunction were more frequently male, Caucasian, with previous sternotomy, higher baseline creatinine, longer ischemic time and worse donor LVEF. They were also more likely to be under RRT, intravenous inotropes or ECMO support and there was more incidence of severe primary graft dysfunction (PGD) (Table 1). Multi-variable logistic regression demonstrated that the strongest predictors for post-transplant renal dysfunction were development of severe PGD (OR 5.26, 2.88–9.62, p&lt;0,001) and RRT prior to HT (OR 5.80, 2.93–11.5, p&lt;0.001). Other predictors were male sex, previous sternotomy, long ischemic time and need for inotropes prior to HT.

Conclusions

Early renal dysfunction is a common complication after HT with an incidence around 10% in a large and contemporary cohort. The presence of PGD and need for RRT pre-transplant were the strongest predictors for its development. Interestingly, emergent transplantation or need for MCS were not independently associated with ERD. Further studies are needed to identify patients at high risk of early and late kidney dysfunction that may benefit from combined transplantation.

Funding Acknowledgement

Type of funding sources: None.

Diagnostic Accuracy of Cardiac MRI versus FDG PET for Cardiac Sarcoidosis: A Systematic Review and Meta-Analysis

Background There is limited consensus regarding the relative diagnostic performance of cardiac MRI and fluorodeoxyglucose (FDG) PET for cardiac sarcoidosis. Purpose To perform a systematic review and meta-analysis to compare the diagnostic accuracy of cardiac MRI and FDG PET for cardiac sarcoidosis. Materials and Methods Medline, Ovid Epub, Cochrane Central Register of Controlled Trials, Embase, Emcare, and Scopus were searched from inception until January 2022. Inclusion criteria included studies that evaluated the diagnostic accuracy of cardiac MRI or FDG PET for cardiac sarcoidosis in adults. Data were independently extracted by two investigators. Summary accuracy metrics were obtained by using bivariate random-effects meta-analysis. Meta-regression was used to assess the effect of different covariates. Risk of bias was assessed using the Quality Assessment Tool for Diagnostic Accuracy Studies-2 tool. The study protocol was registered a priori in the International Prospective Register of Systematic Reviews (Prospero protocol CRD42021214776). Results Thirty-three studies were included (1997 patients, 687 with cardiac sarcoidosis); 17 studies evaluated cardiac MRI (1031 patients) and 26 evaluated FDG PET (1363 patients). Six studies directly compared cardiac MRI and PET in the same patients (303 patients). Cardiac MRI had higher sensitivity than FDG PET (95% vs 84%; P = .002), with no difference in specificity (85% vs 82%; P = .85). In a sensitivity analysis restricted to studies with direct comparison, point estimates were similar to those from the overall analysis: cardiac MRI and FDG PET had sensitivities of 92% and 81% and specificities of 72% and 82%, respectively. Covariate analysis demonstrated that sensitivity for FDG PET was highest with quantitative versus qualitative evaluation (93% vs 76%; P = .01), whereas sensitivity for MRI was highest with inclusion of T2 imaging (99% vs 88%; P = .001). Thirty studies were at risk of bias. Conclusion Cardiac MRI had higher sensitivity than fluorodeoxyglucose PET for diagnosis of cardiac sarcoidosis but similar specificity. Limitations, including risk of bias and few studies with direct comparison, necessitate additional study. © RSNA, 2022 Online supplemental material is available for this article.

Sparing the Prod: Providing an Alternative to Endomyocardial Biopsies With Noninvasive Surveillance After Heart Transplantation During COVID-19

Background

The COVID-19 pandemic has reduced access to endomyocardial biopsy (EMB) rejection surveillance in heart transplant (HT) recipients. This study is the first in Canada to assess the role for noninvasive rejection surveillance in personalizing titration of immunosuppression and patient satisfaction post-HT.

Methods

In this mixed-methods prospective cohort study, adult HT recipients more than 6 months from HT had their routine EMBs replaced by noninvasive rejection surveillance with gene expression profiling (GEP) and donor-derived cell-free DNA (dd-cfDNA) testing. Demographics, outcomes of noninvasive surveillance score, hospital admissions, patient satisfaction, and health status on the medical outcomes study 12-item short-form health survey (SF-12) were collected and analyzed, using t tests and χ2 tests. Thematic qualitative analysis was performed for open-ended responses.

Results

Among 90 patients, 31 (33%) were enrolled. A total of 36 combined GEP/dd-cfDNA tests were performed; 22 (61%) had negative results for both, 10 (27%) had positive GEP/negative dd-cfDNA results, 4 (11%) had negative GEP/positive dd-cfDNA results, and 0 were positive on both. All patients with a positive dd-cfDNA result (range: 0.19%-0.81%) underwent EMB with no significant cellular or antibody-mediated rejection. A total of 15 cases (42%) had immunosuppression reduction, and this increased to 55% in patients with negative concordant testing. Overall, patients' reported satisfaction was 90%, and on thematic analysis they were more satisfied, with less anxiety, during the noninvasive testing experience.

Conclusions

Noninvasive rejection surveillance was associated with the ability to lower immunosuppression, increase satisfaction, and reduce anxiety in HT recipients, minimizing exposure for patients and providers during a global pandemic.

Poke Not Prod: First Canadian Experience Using Donor-Derived Cell Free DNA to Replace Endomyocardial Biopsy During COVID-19

Purpose

After a heart transplant (HT), non-invasive methods for rejection surveillance minimize the need for endomyocardial biopsies (EMBx). We describe the first experience with combined use of genetic expression profiling (GEP) and donor-derived cell-free DNA (dd-cfDNA) testing in Canada as part of a quality improvement project to minimize patient risk during the COVID pandemic.

Methods

Adult outpatients at least 6 months after HT were screened from May 2021 to July 2021 to have their routine EMBx replaced by a combination of GEP and dd-cfDNA. Demographics, modification of immunosuppression (IS) and outcomes (hospital admission, rejection, and need for EMBx) were collected.

Results

Among 90 patients, 31 (33%) were enrolled, and 37 non-invasive tests were performed. The median time after HT was 2 years and patients were predominantly Caucasian (52%) and male (68%). 53% had a history of acute cellular rejection during the first year and 32% had cardiac allograft vasculopathy. Of the tests performed, 23 (60%) were - GEP / - dd-cfDNA, 10 (27%) were + GEP / - dd-cfDNA, 4 (11%) were - GEP / + dd-cfDNA and none were + GEP / + dd-cfDNA. Being bridged with a VAD (OR = 5.5, p=0.034) and a history of a previously treated CMV (OR = 16.0, p=0.003) were associated with a positive GEP and a negative dd-cfDNA result. Having received a COVID vaccine in the last 3 months did not affect GEP results (GEP was positive in 23.8% after vaccination vs 33.3% in non-vaccinated patients, p=0.690; average GEP score 29.8 vs 30.7, p=0.673). The 4 patients with a + dd-cfDNA (range 0.19 - 0.81%) underwent an EMBx with no significant cellular or antibody mediated rejection, thus avoiding 89% of the EMBx. No unscheduled clinic visits, emergency department or hospital admissions were recorded. After non-invasive testing, the IS was reduced in 16 cases (43.2%). IS was reduced in in 59% of patients with negative concordant tests (- GEP / - dd-cfDNA), 30% in patients with + GEP / - dd-cfDNA and no reduction in IS occurred in those with + dd-cfDNA.

Conclusion

The combination of GEP and dd-cfDNA for rejection surveillance allowed for a marked reduction in EMBx (89%) and for a personalized downtitration of IS without adverse events in the short term. The use of non-invasive rejection surveillance testing was an effective strategy to avoid hospital contact for HT recipients during the COVID-19 pandemic.

Management of Frail and Older Homebound Patients With Heart Failure: A Contemporary Virtual Ambulatory Model

Background

Heart failure (HF) affects many patients who are older and frail, presenting multiple physical barriers to accessing specialty care in a traditional ambulatory clinic model. Here, we present an assisted virtual care model in which a home visiting nurse facilitated video visits with a HF cardiologist to follow homebound, frail, and older patients with HF.

Methods

This is a pragmatic, quasi-experimental, pre–post, single-centre study. It included homebound, frail, and older patients with HF from 2015 to 2019 who were followed for 1 year; in-person clinic visits were completely replaced by nurse-facilitated virtual video visits. Outcomes evaluated included annualized hospitalization rate, number of hospitalization days, and number of emergency department visits.

Results

A total of 49 patients were included, with a median age of 86 (83-93) years, and were followed for 1 year after enrollment. Among patients enrolled, HF with preserved ejection fraction was the most common subtype (57%). Compared to the year prior to enrollment, patients had a lower mortality-adjusted all-cause annualized hospitalization rate in the year following enrollment (2.57 vs 1.78, P < 0.0001). Compared to the year prior, the number of mortality-adjusted all-cause hospitalization days was significantly lower in the year following enrollment (27.2 vs 21.4, P < 0.0001). There was a reduction in the number of all-cause annualized emergency department visits (3.10 vs 2.27, P = 0.003).

Conclusions

Nurse-assisted virtual visits may be a preferable strategy for homebound, frail, and older patients with HF to receive longitudinal care. This approach may represent a plausible strategy to care for other patients with significant barriers to accessing specialized cardiac care.

Combined simultaneous FDG-PET/MRI with T1 and T2 mapping as an imaging biomarker for the diagnosis and prognosis of suspected cardiac sarcoidosis

PURPOSE

To evaluate the diagnostic and prognostic significance of combined cardiac 18F-fluorodeoxyglucose (FDG) PET/MRI with T1/T2 mapping in the evaluation of suspected cardiac sarcoidosis.

METHODS

Patients with suspected cardiac sarcoidosis were prospectively enrolled for cardiac 18F-FDG PET/MRI, including late gadolinium enhancement (LGE) and T1/T2 mapping with calculation of extracellular volume (ECV). The final diagnosis of cardiac sarcoidosis was established using modified JMHW guidelines. Major adverse cardiac events (MACE) were assessed as a composite of cardiovascular death, ventricular tachyarrhythmia, bradyarrhythmia, cardiac transplantation or heart failure. Statistical analysis included Cox proportional hazard models.

RESULTS

Forty-two patients (53 ± 13 years, 67% male) were evaluated, 13 (31%) with a final diagnosis of cardiac sarcoidosis. Among patients with cardiac sarcoidosis, 100% of patients had at least one abnormality on PET/MRI: FDG uptake in 69%, LGE in 100%, elevated T1 and ECV in 100%, and elevated T2 in 46%. FDG uptake co-localized with LGE in 69% of patients with cardiac sarcoidosis compared to 24% of those without, p = 0.014. Diagnostic specificity for cardiac sarcoidosis was highest for FDG uptake (69%), elevated T2 (79%), and FDG uptake co-localizing with LGE (76%). Diagnostic sensitivity was highest for LGE, elevated T1 and ECV (100%). After median follow-up duration of 634 days, 13 patients experienced MACE. All patients who experienced MACE had LGE, elevated T1 and elevated ECV. FDG uptake (HR 14.7, p = 0.002), elevated T2 (HR 9.0, p = 0.002) and native T1 (HR 1.1 per 10 ms increase, p = 0.044) were significant predictors of MACE even after adjusting for left ventricular ejection fraction and immune suppression treatment. The presence of FDG uptake co-localizing with LGE had the highest diagnostic performance overall (AUC 0.73) and was the best predictor of MACE based on model goodness of fit (HR 14.9, p = 0.001).

CONCLUSIONS

Combined cardiac FDG-PET/MRI with T1/T2 mapping provides complementary diagnostic information and predicts MACE in patients with suspected cardiac sarcoidosis.

Digital Technology Application for Improved Responses to Health Care Challenges: Lessons Learned From COVID-19

While COVID-19 is still ongoing and associated with more than 5 million deaths, the scope and speed of advances over the past year in terms of scientific discovery, data dissemination, and technology have been staggering. It is not a matter of “if” but “when” we will face the next pandemic, and how we leverage technology and data management effectively to create flexible ecosystems that facilitate collaboration, equitable care, and innovation will determine its severity and scale. The aim of this review is to address emerging challenges that came to light during the pandemic in health care and innovations that enabled us to adapt and continue to care for patients. The pandemic highlighted the need for seismic shifts in care paradigms and technology with considerations related to the digital divide and health literacy for digital health interventions to reach full potential and improve health outcomes. We discuss advances in telemedicine, remote patient monitoring, and emerging wearable technologies. Despite the promise of digital health, we emphasise the importance of addressing its limitations, including interpretation challenges, accuracy of findings, and artificial intelligence–driven algorithms. We summarise the most recent recommendation of the Virtual Care Task Force to scaling virtual medical services in Canada. Finally, we propose a model for optimal implementation of health digital innovations with 5 tenets including data management, data security, digital biomarkers, useful artificial intelligence, and clinical integration.

Impact of using higher-risk donor hearts for candidates with pre-transplant mechanical circulatory support

BACKGROUND

We evaluated post-heart transplant (HTx) outcomes after use of higher-risk donor hearts for candidates supported with pre-HTx mechanical circulatory support (MCS).

METHODS

In this retrospective analysis of the national United Network for Organ Sharing registry, a total of 9,915 adult candidates on MCS underwent HTx from January 1, 2010 to March 31, 2019. Multi-organ, re-transplant, and congenital heart disease patients were excluded. Higher-risk donor organs met at least one of the following criteria: left ventricular ejection fraction <50%, donor to recipient predicted heart mass ratio <0.86, donor age >55 years, or ischemic time >4 hours. Primary outcome was 1 year post-transplant survival.

RESULTS

Among HTx recipients, 3688 (37.2%) received higher-risk donor hearts. Candidates supported with pre-HTx extracorporeal membrane oxygenation or biventricular assist device (n = 374, 3.8%) who received higher-risk donor hearts had comparable 1 year survival (HR: 1.14, 95% CI: [0.67-1.93], p = 0.64) to recipients of standard-risk donor hearts, when adjusted for recipient age and sex. In candidates supported with intra-aortic balloon pump (n = 1391, 14.6%), transplantation of higher-risk donor hearts did not adversely affect 1 year survival (HR: 0.80, 95% CI: [0.52-1.22], p = 0.30). Patients on durable left ventricular assist devices (LVAD) who received higher-risk donor hearts had comparable 1 year survival to continued LVAD support on the waitlist, but mortality was increased compared to those who received standard-risk donor hearts (HR: 1.37, 95% CI: [1.11-1.70], p = 0.004).

CONCLUSIONS

Patients requiring pre-HTx temporary MCS who received higher-risk donor hearts had comparable 1 year post-transplant survival to those who received standard-risk donor hearts. Stable patients on durable LVADs may benefit from waiting for standard-risk donor hearts.

Remote Mobile Outpatient Monitoring in Transplant (Reboot) 2.0: Protocol for a Randomized Controlled Trial

Background

The number of solid organ transplants in Canada has increased 33% over the past decade. Hospital readmissions are common within the first year after transplant and are linked to increased morbidity and mortality. Nearly half of these admissions to the hospital appear to be preventable. Mobile health (mHealth) technologies hold promise to reduce admission to the hospital and improve patient outcomes, as they allow real-time monitoring and timely clinical intervention.

Objective

This study aims to determine whether an innovative mHealth intervention can reduce hospital readmission and unscheduled visits to the emergency department or transplant clinic. Our second objective is to assess the use of clinical and continuous ambulatory physiologic data to develop machine learning algorithms to predict the risk of infection, organ rejection, and early mortality in adult heart, kidney, and liver transplant recipients.

Methods

Remote Mobile Outpatient Monitoring in Transplant (Reboot) 2.0 is a two-phased single-center study to be conducted at the University Health Network in Toronto, Canada. Phase one will consist of a 1-year concealed randomized controlled trial of 400 adult heart, kidney, and liver transplant recipients. Participants will be randomized to receive either personalized communication using an mHealth app in addition to standard of care phone communication (intervention group) or standard of care communication only (control group). In phase two, the prior collected data set will be used to develop machine learning algorithms to identify early markers of rejection, infection, and graft dysfunction posttransplantation. The primary outcome will be a composite of any unscheduled hospital admission, visits to the emergency department or transplant clinic, following discharge from the index admission. Secondary outcomes will include patient-reported outcomes using validated self-administered questionnaires, 1-year graft survival rate, 1-year patient survival rate, and the number of standard of care phone voice messages.

Results

At the time of this paper’s completion, no results are available.

Conclusions

Building from previous work, this project will aim to leverage an innovative mHealth app to improve outcomes and reduce hospital readmission in adult solid organ transplant recipients. Additionally, the development of machine learning algorithms to better predict adverse health outcomes will allow for personalized medicine to tailor clinician-patient interactions and mitigate the health care burden of a growing patient population.

Trial Registration

ClinicalTrials.gov NCT04721288; https://www.clinicaltrials.gov/ct2/show/NCT04721288

International Registered Report Identifier (IRRID)

PRR1-10.2196/26816

Cardiopulmonary Exercise Testing With Echocardiography to Assess Recovery in Patients With Ventricular Assist Devices

The left ventricular assist device (LVAD) is an established treatment for select patients with end-stage heart failure. Some patients recovered and are considered for explantation. Assessing recovery involves exercise testing and echo ramping on full and minimal LVAD support. Combined cardiopulmonary exercise testing with simultaneous echo ramping (CPET-R) has not been well studied. Patients were included if they had CPET within the previous 6 months, were clinically stable, and had an INR >2.0 on the day of examination. Patients had CPET-R on two occasions within 14 days: (a) with LVAD at therapeutic speed and (b) with LVAD at the lowest speed possible. Six patients were between 29 and 75 years (two female). One patient did not complete a turn-down test due to evidence of ischemia on initial CPET-R subsequently confirmed as a significant coronary artery stenosis on angiography. There were no significant differences in CPET or echo metrics between LVAD speeds. Two patients were explanted due to presumed LV recovery and remained event free for 30 and 47 months, respectively. Serial CPET-R seems safe and feasible for the evaluation of LV and global function and may result in improved clinical decision making for LVAD explantation.

Impact of diabetes mellitus on clinical outcomes after heart transplantation

PURPOSE

Diabetes mellitus (DM) is common among recipients of heart transplantation (HTx) but its impact on clinical outcomes is unclear. We evaluated the associations between pretransplant DM and posttransplant DM (PTDM) and outcomes among adults receiving HTx at a single center.

METHODS

We performed a retrospective study (range 01/2008 - 07/2018), n = 244. The primary outcome was survival; secondary outcomes included acute rejection, cardiac allograft vasculopathy, infection requiring hospitalization, macrovascular events, and dialysis initiation post-transplant. Comparisons were performed using Kaplan-Meier and multivariable Cox regression analyses.

RESULTS

Pretransplant DM was present in 75 (30.7%) patients and was associated with a higher risk for infection requiring hospitalization (p<0.05), but not with survival or other outcomes. Among the 144 patients without pretransplant DM surviving to one year, 29 (20.1%) were diagnosed with PTDM at the 1-year follow-up. After multivariable adjustment, PTDM diagnosis at 1-year remained associated with worse subsequent survival (hazard ratio 2.72, 95% confidence interval 1.03-7.16). Predictors of PTDM at 1-year included cytomegalovirus seropositivity and higher prednisone dose (>5mg/day) at 1-year follow-up.

CONCLUSIONS

Compared to HTx recipients without baseline DM, those with baseline DM have a higher risk for infections requiring hospitalization, and those who develop DM after HTx have worse survival. This article is protected by copyright. All rights reserved.