High-sensitivity C-reactive protein is associated with cardiac dysfunction and morbidity in a community-based cohort

Abstract Background Inflammation is increasingly recognized as a risk factor for cardiac morbidity in the general population. Biomarkers like high-sensitivity C-reactive protein (hs-CRP) can aid in the detection of adverse systemic inflammation. The relationship between hs-CRP and cardiac structure and function is therefore a target of interest. Purpose The purpose of this study was to investigate the association of inflammation with cardiac structure and function in individuals free of overt cardiac disease. In addition, we examined if hs-CRP provided prognostic information regarding incident heart failure and myocardial infarction independent of echocardiographic measures. Methods In a community-based cohort study, participants were randomly selected for invitation from the Capital Region of Denmark. Echocardiography was performed on all the participants and measures of cardiac structure and function were obtained by blinded investigators. Participants with a history of cardiac disease were excluded from the analysis. Associations between log-transformed hs-CRP and echocardiographic measures were analyzed through uni- and multivariable linear regressions. Results A total of 3,379 individuals were included. Mean age was 55 years, 42% were female and the median hs-CRP was 1.13 (interquartile range: 0.65 - 1.47) mg/L. In the univariable analyses, increasing levels of hs-CRP were associated with worse left ventricular (LV) function including LV ejection fraction, global longitudinal strain (GLS), worse diastolic function (E/e’), worse right heart function (tricuspid regurgitation (TR) velocity and tricuspid annular plane systolic excursion (TAPSE)) and higher left ventricular (LV) mass index (p < 0.001 for all measures) (figure). When adjusting for sex, age, physical activity, smoking status, body mass index, blood pressure, hypercholesterolemia, diabetes, hypertension, creatinine, and heart rate, only LVEF (p < 0.001), GLS, (p < 0.001), E/e’ (p = 0.027), TAPSE (p = 0.008) and peak TR velocity (p = 0.015) remained significantly associated with increasing hs-CRP levels. In multivariable Cox regression adjusting for clinical characteristics and echocardiographic measures including LV ejection fraction, LV mass index, E/e’, TAPSE and GLS, higher hs-CRP was significantly associated with both heart failure and myocardial infarction (table) (HR for heart failure: 1.30, 95% CI: 1.03–1.66 p = 0.030; HR for acute myocardial infarction: 1.40, 95% CI: 1.04 – 1.89, p = 0.028, per doubling of hs-CRP). We did not find any interactions between hs-CRP and outcome among different types of echocardiographic phenotypes. Conclusion In the general population, higher levels of hs-CRP were associated with declining systolic, diastolic and right heart function. The associations persisted when adjusting for clinical characteristics. In addition, hs-CRP was associated with cardiac events independent of cardiac function as assessed by echocardiography.Linear regression splinesCox regressions