Prediction of congestive state and prognosis in acute and chronic heart failure: the association between NT-proBNP and left atrial strain

Background In acute and chronic heart failure (HF), the relief of congestion is one of the pivotal elements to improve patient's quality of life and prognosis. N-terminal pro-brain natriuretic peptide (NT-proBNP) is a well-known marker of cardiovascular congestion in HF, although with limited specificity. Peak atrial longitudinal strain (PALS) by speckle tracking echocardiography is emerging as an index of left ventricular (LV) filling pressure and prognosis in HF, however, its role as a marker of congestion should be further elucidated. Purpose the aim of our study was to determine the association between NT-proBNP and PALS and their prognostic implications in patients acute and chronic HF. Methods patients hospitalized for de-novo or recurrent acute HF and patients with chronic HF referred to our echo-labs for follow-up evaluation were included in this retrospective study. Patients with missing data, previous cardiac surgery, heart transplant and/or left ventricular assist device implantation, non-feasible speckle tracking analysis were excluded. Clinical characteristics, laboratory examinations, transthoracic echocardiography data were collected. Speckle tracking analysis was performed offline on the echocardiographic records. Follow up data were obtained via electronical records or phone-calls. The primary clinical endpoint was a combination of all-cause death and HF hospitalization. Results the overall study cohort included 388 patients, of which 172 with acute HF and 216 with chronic HF. Mean age was 65±12, 37% were female. The majority of patients had reduced LV systolic function (mean LV ejection fraction = 30±10%; mean LV global longitudinal strain = −8.3±3.9%). Patients with acute HF presented higher values of NT-proBNP than those with chronic HF (median [interquartile range] = 6039 [2989; 13535]pg/ml vs. 544[200; 1533] pg/ml) and lower global PALS (10.4 [6.3; 16.45] vs. 15.6 [10.6; 21]%). Global PALS showed a significant inverse correlation with NT-proBNP both in acute and chronic HF (Fig. 1; all p<0.001) and to be a significant predictor of NT-proBNP with linear regression analysis (R2=0.2; p<0.001). During a median follow-up of 1 year, 98 patients reached the combined endpoint (49 all-cause deaths, 16 cardiovascular deaths, 62 HF hospitalizations). With ROC curves, both NT-proBNP and global PALS showed to be good predictors of the combined endpoint (AUC=0.87 and 0.82 respectively, Fig. 2). Kaplan-Meier analysis showed a good risk stratification for the composite endpoint for ROC-derived cut-off of PALS ≤15% and NTproBNP ≥874.5%, and above all for their combination (Fig. 1B) Conclusions global PALS is associated with NT-proBNP in acute and chronic HF and may be used as additional index of congestion to optimize therapeutic management in these patients. The combination of global PALS and NT-proBNP could enhance the prognostic stratification of HF.