Search

Background: A systemic proinflammatory state has been hypothesized to mediate the association between comorbidities and abnormal cardiac structure/function in heart failure with preserved ejection fraction (HFpEF). We conducted a proteomic analysis to investigate this paradigm., Methods: In 228 patients with HFpEF from the multicenter PROMIS-HFpEF study (Prevalence of Microvascular Dysfunction in Heart Failure With Preserved Ejection Fraction), 248 unique circulating proteins were quantified by a multiplex immunoassay (Olink) and used to recapitulate systemic inflammation. In a deductive approach, we performed principal component analysis to summarize 47 proteins known a priori to be involved in inflammation. In an inductive approach, we performed unbiased weighted coexpression network analyses of all 248 proteins to identify clusters of proteins that overrepresented inflammatory pathways. We defined comorbidity burden as the sum of 8 common HFpEF comorbidities. We used multivariable linear regression and statistical mediation analyses to determine whether and to what extent inflammation mediates the association of comorbidity burden with abnormal cardiac structure/function in HFpEF. We also externally validated our findings in an independent cohort of 117 HFpEF cases and 30 comorbidity controls without heart failure., Results: Comorbidity burden was associated with abnormal cardiac structure/function and with principal components/clusters of inflammation proteins. Systemic inflammation was also associated with increased mitral E velocity, E/e' ratio, and tricuspid regurgitation velocity; and worse right ventricular function (tricuspid annular plane systolic excursion and right ventricular free wall strain). Inflammation mediated the association between comorbidity burden and mitral E velocity (proportion mediated 19%-35%), E/e' ratio (18%-29%), tricuspid regurgitation velocity (27%-41%), and tricuspid annular plane systolic excursion (13%) ( P <0.05 for all), but not right ventricular free wall strain. TNFR1 (tumor necrosis factor receptor 1), UPAR (urokinase plasminogen activator receptor), IGFBP7 (insulin-like growth factor binding protein 7), and GDF-15 (growth differentiation factor-15) were the top individual proteins that mediated the relationship between comorbidity burden and echocardiographic parameters. In the validation cohort, inflammation was upregulated in HFpEF cases versus controls, and the most prominent inflammation protein cluster identified in PROMIS-HFpEF was also present in HFpEF cases (but not controls) in the validation cohort., Conclusions: Proteins involved in inflammation form a conserved network in HFpEF across 2 independent cohorts and may mediate the association between comorbidity burden and echocardiographic indicators of worse hemodynamics and right ventricular dysfunction. These findings support the comorbidity-inflammation paradigm in HFpEF.

Background: Automated cardiac image interpretation has the potential to transform clinical practice in multiple ways, including enabling serial assessment of cardiac function by nonexperts in primary care and rural settings. We hypothesized that advances in computer vision could enable building a fully automated, scalable analysis pipeline for echocardiogram interpretation, including (1) view identification, (2) image segmentation, (3) quantification of structure and function, and (4) disease detection., Methods: Using 14 035 echocardiograms spanning a 10-year period, we trained and evaluated convolutional neural network models for multiple tasks, including automated identification of 23 viewpoints and segmentation of cardiac chambers across 5 common views. The segmentation output was used to quantify chamber volumes and left ventricular mass, determine ejection fraction, and facilitate automated determination of longitudinal strain through speckle tracking. Results were evaluated through comparison to manual segmentation and measurements from 8666 echocardiograms obtained during the routine clinical workflow. Finally, we developed models to detect 3 diseases: hypertrophic cardiomyopathy, cardiac amyloid, and pulmonary arterial hypertension., Results: Convolutional neural networks accurately identified views (eg, 96% for parasternal long axis), including flagging partially obscured cardiac chambers, and enabled the segmentation of individual cardiac chambers. The resulting cardiac structure measurements agreed with study report values (eg, median absolute deviations of 15% to 17% of observed values for left ventricular mass, left ventricular diastolic volume, and left atrial volume). In terms of function, we computed automated ejection fraction and longitudinal strain measurements (within 2 cohorts), which agreed with commercial software-derived values (for ejection fraction, median absolute deviation=9.7% of observed, N=6407 studies; for strain, median absolute deviation=7.5%, n=419, and 9.0%, n=110) and demonstrated applicability to serial monitoring of patients with breast cancer for trastuzumab cardiotoxicity. Overall, we found automated measurements to be comparable or superior to manual measurements across 11 internal consistency metrics (eg, the correlation of left atrial and ventricular volumes). Finally, we trained convolutional neural networks to detect hypertrophic cardiomyopathy, cardiac amyloidosis, and pulmonary arterial hypertension with C statistics of 0.93, 0.87, and 0.85, respectively., Conclusions: Our pipeline lays the groundwork for using automated interpretation to support serial patient tracking and scalable analysis of millions of echocardiograms archived within healthcare systems.