Heart-brain axis: pushing the boundaries of cardiovascular molecular imaging.

Despite decades of research, the heart-brain axis continues to challenge investigators seeking to unravel its complex pathobiology. Strong epidemiologic evidence supports a link by which insult or injury to one of the organs increases the risk of pathology in the other. The putative pathways have important differences between sexes and include alterations in autonomic function, metabolism, inflammation, and neurohormonal mechanisms that participate in crosstalk between the heart and brain and contribute to vascular changes, the development of shared risk factors, and oxidative stress. Recently, given its unique ability to characterize biological processes in multiple tissues simultaneously, molecular imaging has yielded important insights into the interplay of these organ systems under conditions of stress and disease. Yet, additional research is needed to probe further into the mechanisms underlying the heart-brain axis and to evaluate the impact of targeted interventions.

Total-Body PET/CT Applications in Cardiovascular Diseases: A Perspective Document of the SNMMI Cardiovascular Council.

Digital PET/CT systems with a long axial field of view have become available and are emerging as the current state of the art. These new camera systems provide wider anatomic coverage, leading to major increases in system sensitivity. Preliminary results have demonstrated improvements in image quality and quantification, as well as substantial advantages in tracer kinetic modeling from dynamic imaging. These systems also potentially allow for low-dose examinations and major reductions in acquisition time. Thereby, they hold great promise to improve PET-based interrogation of cardiac physiology and biology. Additionally, the whole-body coverage enables simultaneous assessment of multiple organs and the large vascular structures of the body, opening new opportunities for imaging systemic mechanisms, disorders, or treatments and their interactions with the cardiovascular system as a whole. The aim of this perspective document is to debate the potential applications, challenges, opportunities, and remaining challenges of applying PET/CT with a long axial field of view to the field of cardiovascular disease.

Toward Quantitative Multisite Preclinical Imaging Studies in Acute Myocardial Infarction: Evaluation of the Immune-Fibrosis Axis.

The immune-fibrosis axis plays a critical role in cardiac remodeling after acute myocardial infarction. Imaging approaches to monitor temporal inflammation and fibroblast activation in mice have seen wide application in recent years. However, the repeatability of quantitative measurements remains challenging, particularly across multiple imaging centers. We aimed to determine reproducibility of quantitative inflammation and fibroblast activation images acquired at 2 facilities after myocardial infarction in mice. Methods: Mice underwent coronary artery ligation and sequential imaging with 68Ga-DOTA-ECL1i to assess chemokine receptor type 2 expression at 3 d after myocardial infarction and 68Ga-FAPI-46 to assess fibroblast activation protein expression at 7 d after myocardial infarction. Images were acquired at 1 center using either a local or a consensus protocol developed with the second center; the protocols differed in the duration of isoflurane anesthesia and the injected tracer dose. A second group of animals were scanned at the second site using the consensus protocol. Image analyses performed by each site and just by 1 site were also compared. Results: The uptake of 68Ga-DOTA-ECL1i in the infarct territory tended to be higher when the consensus protocol was used (P = 0.03). No difference was observed between protocol acquisitions for 68Ga-FAPI-46. Compared with the local protocol, the consensus protocol decreased variability between individual animals. When a matched consensus protocol was used, the 68Ga-DOTA-ECL1i infarct territory percentage injected dose per gram of tissue was higher on images acquired at site B than on those acquired at site A (P = 0.006). When normalized to body weight as SUV, this difference was mitigated. Both the percentage injected dose per gram of tissue and the SUV were comparable between sites for 68Ga-FAPI-46. Image analyses at the sites differed significantly, but this difference was mitigated when all images were analyzed at site A. Conclusion: The application of a standardized acquisition protocol may lower variability within datasets and facilitate comparison of molecular radiotracer distribution between preclinical imaging centers. Like clinical studies, multicenter preclinical studies should use centralized core-based image analysis to maximize reproducibility across sites.

Unique 40-year survival after heart transplantation with normal graft function and spontaneous operational tolerance.

Unique 40-year survival after heart transplantation with normal graft function and spontaneous operational tolerance.

Myocardial Fibrosis: Emerging Target for Cardiac Molecular Imaging and Opportunity for Image-Guided Therapy.

Myocardial fibrosis is a major contributor to the development and progression of heart failure. Significant progress in the understanding of its pathobiology has led to the introduction and preclinical testing of multiple highly specific antifibrotic therapies. Because the mechanisms of fibrosis are highly dynamic, and because the involved cell populations are heterogeneous and plastic, there is increasing emphasis that any therapy directed specifically against myocardial fibrosis will require personalization and guidance by equally specific diagnostic testing for successful clinical translation. Noninvasive imaging techniques have undergone significant progress and provide increasingly specific information about the quantity, quality, and activity of myocardial fibrosis. Cardiac MRI can precisely map the extracellular space of the myocardium, whereas nuclear imaging characterizes activated fibroblasts and immune cells as the cellular components contributing to fibrosis. Existing techniques may be used in complementarity to provide the imaging biomarkers needed for the success of novel targeted therapies. This review provides a road map on how progress in basic fibrosis research, antifibrotic drug development, and high-end noninvasive imaging may come together to facilitate the success of fibrosis-directed cardiovascular medicine.

Multicenter Evaluation of the Feasibility of Clinical Implementation of SPECT Myocardial Blood Flow Measurement: Intersite Variability and Imaging Time.

BACKGROUND: Single-center studies have shown that single photon emission computed tomography myocardial blood flow (MBF) measurement is accurate compared with MBF measured with microspheres in a porcine model, positron emission tomography, and angiography. Clinical implementation requires consistency across multiple sites. The study goal is to determine the intersite processing repeatability of single photon emission computed tomography MBF and the additional camera time required. METHODS: Five sites (Canada, Italy, Japan, Germany, and Singapore) each acquired 25 to 35 MBF studies at rest and with pharmacological stress using technetium-99m-tetrofosmin on a pinhole-collimated cadmium-zinc-telluride-based cardiac single photon emission computed tomography camera with standardized list-mode imaging and processing protocols. Patients had intermediate to high pretest probability of coronary artery disease. MBF was measured locally and at a core laboratory using commercially available software. The time a room was occupied for an MBF study was compared with that for a standard rest/stress myocardial perfusion study. RESULTS: With motion correction, the overall correlation in MBF between core laboratory and local site was 0.93 (range, 0.87-0.97) at rest, 0.90 (range, 0.84-0.96) at stress, and 0.84 (range, 0.70-0.92) for myocardial flow reserve. The local-to-core difference in global MBF (bias-MBF) was 5.4% (-3.8% to 14.8%; median [interquartile range]) at rest and 5.4% (-6.2% to 19.4%) at stress. Between the 5 sites, bias-MBF ranged from -1.6% to 11.0% at rest and from -1.9% to 16.3% at stress; the interquartile range in bias-MBF was between 9.3% (4.8%-14.0%) and 22.3% (-10.3% to 12.0%) at rest and between 17.0% (-11.3% to 5.6%) and 33.3% (-10.4% to 22.9%) at stress and was not significantly different between most sites. Both bias and interquartile range were like previously reported interobserver variability and less than the SD of the test-retest difference of 30%. The overall difference in myocardial flow reserve was 1.52% (-10.6% to 11.3%). There were no significant differences between with and without motion correction. The average additional acquisition time varied between sites from 44 to 79 minutes. CONCLUSIONS: The average bias-MBF and bias-MFR values were small with standard deviations substantially less than the test-retest variability. This demonstrates that MBF can be measured consistently across multiple sites and further supports that this technique can be reliably implemented. REGISTRATION: URL: https://www. CLINICALTRIALS: gov; Unique identifier: NCT03427749.

Interplay between driveline infection, vessel wall inflammation, cerebrovascular events and mortality in patients with left ventricular assist device.

In patients with left ventricular assist device (LVAD), infections and thrombotic events represent severe complications. We investigated device-specific local and systemic inflammation and its impact on cerebrovascular events (CVE) and mortality. In 118 LVAD patients referred for 18F-FDG-PET/CT, metabolic activity of LVAD components, thoracic aortic wall, lymphoid and hematopoietic organs, was quantified and correlated with clinical characteristics, laboratory findings, and outcome. Driveline infection was detected in 92/118 (78%) patients by 18F-FDG-PET/CT. Activity at the driveline entry site was associated with increased signals in aortic wall (r = 0.32, p < 0.001), spleen (r = 0.20, p = 0.03) and bone marrow (r = 0.20, p = 0.03), indicating systemic interactions. Multivariable analysis revealed independent associations of aortic wall activity with activity of spleen (ß = 0.43, 95% CI 0.18-0.68, p < 0.001) and driveline entry site (ß = 0.04, 95% CI 0.01-0.06, p = 0.001). Twenty-two (19%) patients suffered CVE after PET/CT. In a binary logistic regression analysis metabolic activity at the driveline entry site missed the level of significance as an influencing factor for CVE after adjusting for anticoagulation (OR = 1.16, 95% CI 1-1.33, p = 0.05). Metabolic activity of the subcutaneous driveline (OR = 1.13, 95% CI 1.02-1.24, p = 0.016) emerged as independent risk factor for mortality. Molecular imaging revealed systemic inflammatory interplay between thoracic aorta, hematopoietic organs, and infected device components in LVAD patients, the latter predicting CVE and mortality.

Total-Body Positron Emission Tomography: Adding New Perspectives to Cardiovascular Research.

The recent advent of positron emission tomography (PET) scanners that can image the entire human body opens up intriguing possibilities for cardiovascular research and future clinical applications. These new systems permit radiotracer kinetics to be measured in all organs simultaneously. They are particularly well suited to study cardiovascular disease and its effects on the entire body. They could also play a role in quantitatively measuring physiologic, metabolic, and immunologic responses in healthy individuals to a variety of stressors and lifestyle interventions, and may ultimately be instrumental for evaluating novel therapeutic agents and their molecular effects across different tissues. In this review, we summarize recent progress in PET technology and methodology, discuss several emerging cardiovascular applications for total-body PET, and place this in the context of multiorgan and systems medicine. Finally, we discuss opportunities that will be enabled by the technology, while also pointing to some of the challenges that still need to be addressed.

Molecular Imaging of Myocardial Fibroblast Activation in Patients with Advanced Aortic Stenosis Before Transcatheter Aortic Valve Replacement: A Pilot Study.

Using multimodal imaging, we investigated the extent and functional correlates of myocardial fibroblast activation in patients with aortic stenosis (AS) scheduled for transcatheter aortic valve replacement (TAVR). AS may cause myocardial fibrosis, which is associated with disease progression and may limit response to TAVR. Novel radiopharmaceuticals identify upregulation of fibroblast activation protein (FAP) as a cellular substrate of cardiac profibrotic activity. Methods: Twenty-three AS patients underwent 68Ga-FAP inhibitor 46 (68Ga-FAPI) PET, cardiac MRI, and echocardiography within 1-3 d before TAVR. Imaging parameters were correlated and then were integrated with clinical and blood biomarkers. Control cohorts of subjects without a history of cardiac disease and with (n = 5) and without (n = 9) arterial hypertension were compared with matched AS subgroups. Results: Myocardial FAP volume varied significantly among AS subjects (range, 1.54-138 cm3, mean ± SD, 42.2 ± 35.6 cm3) and was significantly higher than in controls with (7.42 ± 8.56 cm3, P = 0.007) and without (2.90 ± 6.67 cm3; P < 0.001) hypertension. FAP volume correlated with N-terminal prohormone of brain natriuretic peptide (r = 0.58, P = 0.005), left ventricular ejection fraction (r = -0.58, P = 0.02), mass (r = 0.47, P = 0.03), and global longitudinal strain (r = 0.55, P = 0.01) but not with cardiac MRI T1 (spin-lattice relaxation time) and extracellular volume (P = not statistically significant). In-hospital improvement in left ventricular ejection fraction after TAVR correlated with pre-TAVR FAP volume (r = 0.440, P = 0.035), N-terminal prohormone of brain natriuretic peptide, and strain but not with other imaging parameters. Conclusion: FAP-targeted PET identifies varying degrees of left ventricular fibroblast activation in TAVR candidates with advanced AS. 68Ga-FAPI signal does not match other imaging parameters, generating the hypothesis that it may become useful as a tool for personalized selection of optimal TAVR candidates.

Regression of Myocardial 99mTc-DPD Uptake After Tafamidis Treatment of Cardiac Transthyretin Amyloidosis.

Cardiac transthyretin amyloidosis is an infiltrative cardiomyopathy with high mortality. To date, there are no specific biomarkers to directly assess disease activity and response to specific treatments. Our aim was to evaluate scintigraphic changes after treatment with the transthyretin stabilizer tafamidis. Methods: We included patients who had undergone 99mTc-3,3-diphosphono-1,2-propanodicarboxylic acid (99mTc-DPD) scintigraphy before tafamidis initiation and after at least 9 mo. Tracer activity was assessed visually and quantitatively as SUVmax Results: The study included 14 patients who were on tafamidis for 44 ± 14 mo. We observed regression of Perugini grade in 5 patients, unchanged grade in 9 patients, and regression of mean heart-to-contralateral-lung ratio (P = 0.015) and SUVmax (P = 0.005). There were no changes in N-terminal pro-B-type natriuretic peptide or echocardiographic measures. Conclusion: Treatment with tafamidis results in regression of myocardial 99mTc-DPD uptake. 99mTc-DPD scintigraphy may provide useful imaging biomarkers to assess response to treatment.

[Positron emission tomography in oncology]. / Positronenemissionstomographie in der Onkologie.

Positron emission tomography (PET) is a highly sensitive imaging tool that noninvasively characterizes metabolic processes and molecular targets. PET has become an integral part of oncological diagnostics and an increasingly important tool for oncological therapy management. PET assessment, for example, directly influences treatment escalation or de-escalation in context of Hodgkin lymphomas or is, in case of lung cancer, able to reduce unnecessary surgeries. Hence, molecular PET imaging represents an indispensable tool in the development of personalized treatments. Furthermore, the development of new radiotracers for specific cell surface structures offers a promising potential for diagnostics and-combined with therapeutic nuclides-also for therapies. One recent example are radioligands targeting prostate-specific membrane antigen, which are relevant in prostate cancer.

PSMA-heterogeneity in metastatic castration-resistant prostate cancer: Circulating tumor cells, metastatic tumor burden, and response to targeted radioligand therapy.

BACKGROUND: We explored the interrelation between prostate-specific membrane antigen (PSMA) expression on circulating tumor cells (CTCs) and that of solid metastatic lesions as determined by whole-body PSMA-targeted positron emission tomography (PET) to refine the prediction of response to subsequent PSMA-targeted radioligand therapy (RLT). METHODS: A prospective study was performed in 20 patients with advanced mCRPC. Of these, 16 underwent subsequent RLT with [177 Lu]Lu-PSMA-617 at a dose of 7.4 GBq every 6-8 weeks. PSMA expression on CTCs using the CellSearch system was compared to clinical and serological results, and to marker expression in targeted imaging and available histological sections of prostatectomy specimens (19% of RLT patients). Clinical outcome was obtained after two cycles of RLT. RESULTS: Marked heterogeneity of PSMA expression was observed already at first diagnosis in available histological specimens. Targeted whole-body imaging also showed heterogeneous inter- and intra-patient PSMA expression between metastases. Heterogeneity of CTC PSMA expression was partially paralleled by heterogeneity of whole-body tumor burden PSMA expression. Twenty percent of CTC samples showed no PSMA expression, despite unequivocal PSMA expression of solid metastases at PET. A high fraction of PSMA-negative CTCs emerged as the sole predictor of poor RLT response (odds ratio [OR]: 0.9379 [95% confidence interval, CI, 0.8558-0.9902]; p = 0.0160), and was prognostic for both shorter progression-free survival (OR: 1.236 [95% CI, 1.035-2.587]; p = 0.0043) and overall survival (OR: 1.056 [95% CI, 1.008-1.141]; p = 0.0182). CONCLUSION: This proof-of-principle study suggests that liquid biopsy for CTC PSMA expression is complementary to PET for individual PSMA phenotyping of mCRPC.

Cardiac Applications of Fibroblast Activation Protein Imaging.

Several promising applications of cardiac molecular fibroblast activation protein (FAP) imaging are emerging. Myocardial fibrosis plays a key role in the complex process of cardiac remodeling and can lead to adverse clinical outcomes such as left ventricular dysfunction, propensity to arrhythmias, and reduction of perfusion. If fibrosis becomes irreversible, patients can develop heart failure. Therefore identification and early fibrosis treatment is highly warranted. FAP-targeted imaging enables new insights into pathogenesis and treatment response in various cardiac diseases such as myocardial infarction, heart failure or systemic diseases being a new selective biomarker.

Exploring Vessel Wall Biology In Vivo by Ultrasensitive Total-Body PET.

Ultrasensitive, high-resolution, extended-field-of-view total-body (TB) PET using the first-of-its-kind 194-cm axial-field-of-view uEXPLORER may facilitate the interrogation of biologic hallmarks of hitherto difficult-to-evaluate low-signal vessel wall pathology in cardiovascular disease. Methods: Healthy volunteers were imaged serially for up to 12 h after a standard dose of 18F-FDG (n = 15) or for up to 3 h after injection of a very low dose (about 5% of a standard dose; n = 15). A cohort undergoing standard 18F-FDG PET (n = 15) on a conventional scanner with a 22-cm axial field of view served as a comparison group. Arterial wall signal, crosstalk with hematopoietic and lymphoid organs, and image quality were analyzed using standardized techniques. Results: TB PET depicted the large vessel walls with excellent quality. The arterial wall could be imaged with high contrast up to 12 h after tracer injection. Ultralow-dose TB 18F-FDG images yielded a vessel wall signal and target-to-background ratio comparable to those of conventional-dose, short-axial-field-of-view PET. Crosstalk between vessel wall and lymphoid organs was identified with better accuracy in both TB PET cohorts than in conventional PET. Conclusion: TB PET enables detailed assessment of in vivo vessel wall biology and its crosstalk with other organs over an extended time window after tracer injection or at an ultralow tracer dose. These initial observations support the feasibility of serial imaging in low-risk populations and will stimulate future mechanistic studies or therapy monitoring in atherosclerosis and other vessel wall pathologies.

Individual radiosensitivity reflected by γ-H2AX and 53BP1 foci predicts outcome in PSMA-targeted radioligand therapy.

PURPOSE: γ-H2AX and 53BP1 are fundamental for cellular DNA damage response (DDR) after radiation exposure and are linked to cell repair, arrest, or apoptosis. We aimed to evaluate whether DDR-markers in peripheral blood lymphocytes (PBLs) may have predictive potential for outcome in metastatic castration-resistant prostate cancer (mCRPC) patients receiving [177Lu]Lu-prostate-specific membrane antigen (PSMA) radioligand therapy (RLT). METHODS: We prospectively enrolled 20 men with advanced mCRPC scheduled for PSMA-targeted RLT. Prior to the first cycle of [177Lu]Lu-PSMA RLT, all patients underwent [18F]F-PSMA-1007 positron emission tomography (PET)/computed tomography (CT) for assessment of tumor PSMA expression (assessing maximum standardized uptake value (SUVmax) of all tumor lesions). Blood samples were collected prior to, + 1 h after, and + 24 h after administration of [177Lu]Lu-PSMA, and DDR-markers γ-H2AX and 53BP1 were determined in PBLs through immunocytofluorescence. We then tested the predictive performance of DDR-markers relative to clinical and PET-based parameters for progressive disease (PSA-PD) after 2 cycles. In addition, the predictive value for progression-free survival (PSA-PFS, provided as median and 95% confidence interval [CI]) was explored. RESULTS: Low baseline 53BP1 and γ-H2AX foci (P = 0.17) tended to predict early PSA-PD, whereas low SUVmax was significantly associated with higher risk for PSA-PD (P = 0.04). In Kaplan-Meier analysis, there was a trend towards prolonged PSA-PFS in patients with higher baseline 53BP1 of 6 months (mo; 95%CI, 4-9 mo) compared to 3 mo in patients with low 53BP1 (95% CI, 2-3 mo; P = 0.12). Comparable results were recorded for higher γ-H2AX expression (6 mo [95% CI, 3-9 mo] relative to 3 mo [95% CI, 2-4 mo] in patients with low γ-H2AX; P = 0.12). SUVmax, however, did not demonstrate predictive value (P = 0.29). Consistently, in univariate Cox-regression analysis, baseline 53BP1 foci demonstrated borderline significance for predicting PSA-PFS under [177Lu]Lu-PSMA RLT (P = 0.05). CONCLUSION: In this prospective study investigating mCRPC patients undergoing [177Lu]Lu-PSMA RLT, low baseline DDR-markers in PBLs tended to predict poor outcome. Although the study group was small and results need further confirmation, these preliminary findings lay the foundation for exploring additive radiosensitizing or treatment intensification in future studies with high-risk individuals scheduled for RLT.

Parametric Imaging of Biologic Activity of Atherosclerosis Using Dynamic Whole-Body Positron Emission Tomography.

BACKGROUND: For molecular imaging of atherosclerotic vessel wall activity, tracer kinetic analysis may yield improved contrast versus blood, more robust quantitative parameters, and more reliable characterization of systems biology. OBJECTIVES: The authors introduce a novel dynamic whole-body positron emission tomography (PET) protocol that is enabled by rapid continuous camera table motion, followed by reconstruction of parametric data sets using voxel-based Patlak graphical analysis. METHODS: Twenty-five subjects were prospectively enrolled and underwent dynamic PET up to 90 minutes after injection of 2-[18F]fluoro-2-deoxy-D-glucose (FDG). Two sets of images were generated: 1) the established standard of static standardized uptake value (SUV) images; and 2) parametric images of the metabolic rate of FDG (MRFDG) using the Patlak plot-derived influx rate. Arterial wall signal was measured and compared using the volume-of-interest technique, and its association with hematopoietic and lymphoid organ signal and atherosclerotic risk factors was explored. RESULTS: Parametric MRFDG images provided excellent arterial wall visualization, with elimination of blood-pool activity, and enhanced focus detectability and reader confidence. Target-to-background ratio (TBR) from MRFDG images was significantly higher compared with SUV images (2.6 ± 0.8 vs 1.4 ± 0.2; P < 0.0001), confirming improved arterial wall contrast. On MRFDG images, arterial wall signal showed improved correlation with hematopoietic and lymphoid organ activity (spleen P = 0.0009; lymph nodes P = 0.0055; and bone marrow P = 0.0202) and increased with the number of atherosclerotic risk factors (r = 0.49; P = 0.0138), where signal from SUV images (SUVmaxP = 0.9754; TBRmaxP = 0.8760) did not. CONCLUSIONS: Absolute quantification of MRFDG is feasible for arterial wall using dynamic whole-body PET imaging. Parametric images provide superior arterial wall contrast, and they might be better suited to explore the relationship between arterial wall activity, systemic organ networks, and cardiovascular risk. This novel methodology may serve as a platform for future diagnostic and therapeutic clinical studies targeting the biology of arterial wall disease.

Molecular imaging of the brain-heart axis provides insights into cardiac dysfunction after cerebral ischemia.

Ischemic stroke imparts elevated risk of heart failure though the underlying mechanisms remain poorly described. We aimed to characterize the influence of cerebral ischemic injury on cardiac function using multimodality molecular imaging to investigate brain and cardiac morphology and tissue inflammation in two mouse models of variable stroke severity. Transient middle cerebral artery occlusion (MCAo) generated extensive stroke damage (56.31 ± 40.39 mm3). Positron emission tomography imaging of inflammation targeting the mitochondrial translocator protein (TSPO) revealed localized neuroinflammation at 7 days after stroke compared to sham (3.8 ± 0.8 vs 2.6 ± 0.7 %ID/g max, p < 0.001). By contrast, parenchyma topical application of vasoconstrictor endothelin-1 did not generate significant stroke damage or neuroinflammatory cell activity. MCAo evoked a modest reduction in left ventricle ejection fraction at both 1 weeks and 3 weeks after stroke (LVEF at 3 weeks: 54.3 ± 5.7 vs 66.1 ± 3.5%, p < 0.001). This contractile impairment was paralleled by elevated cardiac TSPO PET signal compared to sham (8.6 ± 2.4 vs 5.8 ± 0.7%ID/g, p = 0.022), but was independent of leukocyte infiltration defined by flow cytometry. Stroke size correlated with severity of cardiac dysfunction (r = 0.590, p = 0.008). Statistical parametric mapping identified a direct association between neuroinflammation at 7 days in a cluster of voxels including the insular cortex and reduced ejection fraction (ρ = - 0.396, p = 0.027). Suppression of microglia led to lower TSPO signal at 7 days which correlated with spared late cardiac function after MCAo (r = - 0.759, p = 0.029). Regional neuroinflammation early after cerebral ischemia influences subsequent cardiac dysfunction. Total body TSPO PET enables monitoring of neuroinflammation, providing insights into brain-heart inter-organ communication and may guide therapeutic intervention to spare cardiac function post-stroke.