Prospective associations of circulating thrombospondin-2 level with heart failure hospitalization, left ventricular remodeling and diastolic function in type 2 diabetes

• Published in: Cardiovascular diabetology 2022-11, Vol.21 (1), p.1-231, Article 231
• Main Authors: Lee, CH, Wu, MZ, Lui, DTW, Fong, CHY, Ren, QW, Yu, SY, Yuen, MMA, Chow, WS, Huang, JY, Xu, A, Yiu, KH, Lam, KSL
• Format: Article
• Language: English
• Subjects: Angiotensin; Angiotensin-converting enzyme inhibitors; Aspirin; Biomarkers; C-reactive protein; Cardiovascular diseases; Cohort analysis; Congestive heart failure; Diabetes; Diabetes mellitus; Diabetes mellitus (non-insulin dependent); Diastolic dysfunction; Diuretics; Drug dosages; Echocardiography; Ejection fraction; Glomerular filtration rate; Glucose transporter; Heart failure; Hospitalization; Hospitalization for heart failure; Hospitals; Hyperglycemia; Insulin; Metformin; Peptidyl-dipeptidase A; Risk factors; Thrombospondin; Thrombospondin-2; Transplants & implants; Ventricle
• ISSN: 1475-2840; 1475-2840
• DOI: 10.1186/s12933-022-01646-x

Background Circulating thrombospondin-2 (TSP2) levels were associated with the development of heart failure (HF) in recent studies. However, these studies included only a minority of patients with type 2 diabetes, which is associated with an increased HF risk. As hyperglycemia induces TSP2 expression and its tissue expression increases in type 2 diabetes, we investigated the prospective association of circulating TSP2 with incident HF hospitalization (HHF), and its associations with longitudinal changes of echocardiographic parameters in type 2 diabetes. Methods Baseline serum TSP2 levels were measured in 4949 patients with type 2 diabetes to determine its association with incident HHF using multivariable Cox regression analysis. In the echocardiographic study, baseline serum TSP2 levels were measured in another 146 patients with type 2 diabetes but without cardiovascular diseases who underwent detailed transthoracic echocardiography at baseline and after 1 year. Results Over a median follow-up of 7.8 years, 330 of 4949 patients (6.7%) developed incident HHF. Baseline serum TSP2 levels were independently associated with the development of HHF (HR 1.31, 95%CI 1.06–1.62, p = 0.014) after adjustments for baseline conventional cardiovascular risk factors, atrial fibrillation, estimated glomerular filtration rate, albuminuria and high-sensitivity C-reactive protein level, use of angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, loop-diuretics, aspirin, insulin, metformin and sodium-glucose co-transporter 2 inhibitors. Moreover, baseline serum TSP2 levels were independently associated with increase in average E/e’ and left atrial volume index (p = 0.04 and