| BackgroundChronic heart failure (CHF) is a clinic syndrome characterized by high morbidity and mortality. Evidence comfirm that the activation of proinflammatory cytokines exerts an important role in the process of heart failure recently. Pentraxin-3 (PTX-3), which belongs to the pentraxins superfamily including the traditional short pentraxin such as C-reactive protein (CRP), is a new biomarker of inflammation. PTX-3 is mainly secreted by the vascular endothelial cell, macrophage and damaged cadiocyte stimulated by proinflammatory factors (TNF-a and IL-I, etc.). It has been demonstrated that serum PTX-3 is significantly elevated in patients with acute coronary syndromes (ACS) and can be used as an independent predictor of recent prognosis for the patients with ACS. To date, the relation between serum pentraxin-3 and CHF is scarcely investigated and reported.ObjectivesThis present study aimed to explore the role of serum PTX-3 levels in CHF, compare PTX-3 with clinical predictors for heart failure such as N-terminal portion of pro-brain nairiuretic peptid (NT-proBNP), high sensitivity C-reactive protein (hs-CRP) and estimated glomerular filtration rate (eGFR), and determine different roles of the above markers in predicting the cardiac function in patients with CHF.Methods130 patients with CHF were enrolled in the study who were admitted in the department of cardiology at the Second Xiangya Hospital of Central South University from September 2008～December 2009. We divided them into heart failure with normal left ventricular ejection fraction (LVEF) group (HFNEF) (54 cases) and heart failure with reduced LVEF group (HFREF) (76 cases) ccording to LVEF>50% or≤50%. At the same time, cases will be divided into non left ventricular diastolic dysfunction (non-LVDD) group (64 cases) and LVDD group (66 cases) According to E/A (ratio of early to late mitral vavle peak diastolic flow velocity)>1 or≤1. Measured levels of serum PTX-3, NT-proBNP, hs-CRP, and eGFR according to Cockcroft-Gault formula will be done. Analysis of the differences between serum PTX-3 and clinical predictors for heart failure (NT-proBNP, hs-CRP and eGFR) between the groups of CHF, and evaluated the difference of correlation between these indexes and the cardiac function in patients with CHF.Results1. Comparison between HFREF group and HFNEF group:(1) Serum PTX-3 level (4.78±0.77 ng/ml vs 4.17±0.42 ng/ml), NT-proBNP level (2641.18±595.00 pg/ml vs 1796.74±531.57 pg/ml) and hs-CRP level (4.78±2.04 mg/L vs 4.12±0.51 mg/L) were significantly higher in HFREF group than HFNEF group, and the difference had statistical significance (all P<0.01). eGFR was significantly lower (60.71±12.99 ml/min vs 65.43±11.43 ml/min) in HFREF group than HFNEF group, and the difference was statistically significant (P<0.05).(2) The single factor Logistic regression analysis (PTX-3, NT-proBNP, hs-CRP, eGFR, etc.) showed that PTX-3 (OR=2.68), NT-proBNP (OR=3.22), hs-CRP (OR=1.32), eGFR (OR=0.97) were closely related to HFREF (all P<0.05). However, multivariate Logistic regression analysis (PTX-3, NT-proBNP, hs-CRP and eGFR) showed that only PTX-3 (OR=2.37) and NT-proBNP (OR=2.75) were closely associated with HFREF (both P<0.05).(3) Serum PTX-3 area under the diagnosis of HFREF ROC curve is 0.75, NT-proBNP 0.85, hs-CRP 0.56 and eGFR 0.60. Serum PTX-3 level in best cutoff value was 4.29 ng/ml and the corresponding sensitivity, specificity and Youden index were 77.6%,64.8% and 42.4% respectively; Serum NT-proBNP level in best cutoff value was 2196.00 pg/ml and the corresponding sensitivity, specificity and Youden index were 77.6%, 75.9% and 53.5% respectively; Serum hs-CRP level in best cutoff value was 4.27 mg/L and the corresponding sensitivity, specificity and Youden index were 52.6%,70.4% and 23.0% respectively; eGFR level in best cutoff value was 64.53 ml/min and the corresponding sensitivity, specificity and Youden index were 52.6%,64.8% and 27.4% respectively. 2. Comparison between LVDD group and non-LVDD group:(1) Serum PTX-3 level (4.79±0.68 ng/ml vs 4.25±0.65 ng/ml) and NT-proBNP level (2585.68±642.49 pg/ml vs 1985.92±636.46 pg/ml) were significantly higher in LVDD group than non-LVDD group, and the difference had statistical significance (both P<0.01). However, hs-CRP level (4.75±1.64 mg/L vs 4.20±1.57 mg/L) and eGFR (61.61±13.38 ml/min vs 63.76±11.61 ml/min) were not statistically significant (both P>0.05).(2) The single factor Logistic regression analysis (PTX-3, NT-proBNP, hs-CRP, eGFR, etc.) showed that PTX-3 (OR=3.51) and NT-proBNP (OR=2.01) were closely related to LVDD (both P<0.01). However, hs-CRP (OR=1.24) and eGFR (OR=0.99) were not closely related with LVDD (both P>0.05). Forthemore, multivariate Logistic regression analysis (PTX-3, NT-proBNP and hs-CRP) showed that PTX-3 (OR=2.23) and NT-proBNP (OR=1.62) were closely associated with LVDD (both P <0.05); However, hs-CRP (OR= 1.18) was not closely related to LVDD (P>0.05).(3) Serum PTX-3 area under the diagnosis of LVDD ROC curve is 0.74, NT-proBNP 0.74, hs-CRP 0.62 and eGFR 0.53, Serum PTX-3 level in best cutoff value was 4.36 ng/ml and the corresponding sensitivity, specificity and Youden index were 75.8%,68.7% and 44.5% respectively; Serum NT-proBNP level in best cutoff value was 2170.00 pg/ml and the corresponding sensitivity, specificity and Youden index were 77.3%, 57.8% and 35.1% respectively; Serum hs-CRP level in best cutoff value was 4.05 mg/L and the corresponding sensitivity, specificity and Youden index were 72.7%,51.6% and 24.3% respectively; eGFR level in best cutoff value was 67.53 ml/min and the corresponding sensitivity, specificity and Youden index were 57.6%,45.3% and 2.9% respectively.3. Linear regression analysis showed:serum PTX-3 and NT-proBNP has positive correlation (r=0.54, P<0.01). However, hs-CRP, eGFR has no correlation (r=0.05 or r=0.06, both P)>0.05).Conclusions1. Serum level of PTX-3 is determined to be a predictor for left ventricular diastolic and systolic dysfunction.2. Serum level of PTX-3 has more accurate than hs-CRP and eGFR, and has the same diagnostic vaule when compared with NT-proBNP in predicting left ventricular diastolic and systolic dysfunction. |
PDF Full Text Request