Umeå University's logo

Background: Heart failure (HF) is defined as a clinical syndrome with typical symptoms and signs determined by congenital or acquired abnormalities of the structure or the function of the heart. Several therapeutic options have improved the quality of life and the outcome of patients with HF in recent years. However, also because an increasing number of individuals survives longer, up to the 10% of HF population reaches the advanced stage of the disease. Advanced HF (AdHF) is characterized by persistent severe symptoms despite optimal HF medical and electrical therapy with very poor functional capacity and episodes of unplanned hospitalizations or visits to reduce congestion or improve cardiac output. AdHF patients periodically undergo invasive right heart catheterization (RHC) to estimate pulmonary pressure and vascular resistance. Heart transplantation (HTX) remains the gold standard treatment for AdHF, allowing patients good mid- and long-term survival rates. Main complications after HTX include rejection, more common within the first year after surgery, and cardiac allograft vasculopathy (CAV) in the long term. Echocardiography is a key first line tool for the routine assessment of cardiac function in AdHF and HTX, to monitor effectiveness of therapies and to stratify prognosis. Second level echocardiography techniques, especially Speckle Tracking-derived Myocardial work, are promising in assessing with more sensibility changes in left ventricular function, especially when associated with biomarkers as natriuretic peptides.

Objectives: This thesis is focused on the added role of echocardiography in AdHF and HTX patients and the main aims include: to study the reliability of echocardiography in the estimation of pulmonary arterial pressures (PAP) and the diagnosis of pulmonary hypertension when compared to gold standard method (i.e. RHC, paper I); to estimate prognostic value of myocardial work in AdHF (paper II); to determine normal reference value of myocardial work indices in the transplanted heart compared to general healthy population (paper III); to describe the trend of NT-viproBNP (the most used natriuretic peptide in HF) after HTX, assessing its possible predictors among pre- and post-operative echocardiographic and clinical variables (paper IV).

Materials and methods: We retrospectively screened patients with AdHF in regular follow up at our University Hospital. For paper I, we selected all individuals with available RHC data and echocardiographic images recorded on the same day, excluding those with diseases which could represent biases, as chronic obstructive pulmonary disease, and those with poor acoustic window or undetectable tricuspid regurgitation; all patients underwent accurate echocardiographic analysis for the estimation of pulmonary pressures including peak tricuspid regurgitation velocity (TRV) and mean right ventricular–right atrial (RV–RA) pressure gradient. For paper II we included all patients with good acoustic windows and brachial artery cuff systemic blood pressure measured at the same time as the echocardiographic exam, for the calculation of myocardial work indices, excluding those with more than mild heart valve disease or atrial fibrillation. Applying dedicated software, myocardial work indices, including global constructive work (GCW), global work efficiency (GWE); global work index (GWI) and global wasted work (GWW) were calculated in each patient which was then followed up for the development of major events (all-cause mortality, HTX, left ventricular assist devices implantation – primary endpoint – or acute HF hospitalization – secondary endpoint). A population of HTX patients without history of CAV or rejection were screened for paper III and included if the acoustic window was good and brachial artery cuff systemic blood pressure was measured at the same time as the echocardiographic examination. Patients were excluded also in the presence of donor-specific antibodies or atrial fibrillation, more than mild mitral or aortic regurgitation, or abnormal left ventricular function. Myocardial work indices were calculated and compared to general population with similar age and no comorbidities, derived from the European Association of Cardiovascular Imaging (EACVI) NORRE study. Lastly, for paper IV, a wider population of HTX patients with available long term follow up and pre-surgical information were screened and described in terms of NT-proBNP values at 10 different time points including 1 month, 3 months, 6 months and 1 year after theviiHTX. Continuous variables were reported as either mean with standard deviation or median with interquartile range according to normal distribution. Receiver Operating Characteristic (ROC) curves were used to evaluate the ability of echocardiographic parameters to predict outcome (PH for Paper I, adverse events for paper II). Pearson's correlation coefficient was utilized to examine the strength of the association between echocardiographic measures and RHC findings in Paper I or NT-proBNP values in paper IV. Univariate and multivariate Cox proportional hazard regression analyses were applied to assess predictors of outcomes in paper II where Kaplan–Meier analysis estimated event-free survival. Linear regression was applied to test possible association with MW indices and population characteristics in paper III.

Results: Paper I: in the 41 patients enrolled, peak TRV was superior in terms of area under the curve by ROC analysis to mean RV–RA gradient in predicting increased mean PAP at RHC, both when using 20 or 25 mmHg as pathological cut off value. In particular, a peak TRV >2.4 m/s had 65% sensitivity and 100% positive predictive value for predicting PH according to the new guidelines’ definition.

Paper II: among 138 enrolled individuals, 35 patients developed at least 1 event at follow up. While myocardial work parameters were not associated with primary endpoint occurrence, the hazard ratio for each increase in GWI by 50 mmHg% was 0.90 (p = 0.025) and for each increase in GCW by 50 mmHg% was 0.90 (p = 0.022) when estimating the risk of acute HF hospitalization. Patients with GWI ≥ 369 mmHg% had a better event-free survival at Kaplan–Meier analysis.

Paper III: 82 HTx patients, 68.3% male with a median age of 53 (46–62) years were included in a median time lapse for HTX of 5 (2–22) months. No significant differences were described in terms of gender in HTX patients. On the contrary, all the myocardial work indices significantly differed from those reported in the EACVI NORRE study (all P-value <0.001), in particular with lower GWI, GCW, and GWE and higher GWW values in the HTX population.viii

Paper IV: in a population of 71 HTX patients, major reduction of NT-proBNP was described at month 3 after surgery, with further reduction at 6 months and 1 year after which it tended to remain stable. Among predictors of NT-proBNP values, at regression analysis, 1-year NT-proBNP values was related to RHC measured pulmonary wedge pressure and ischemic etiology but also to post-HTX kidney function and tricuspid regurgitation severity; long term NT-proBNP values were instead predicted by positive Human Leucocyte Antigen (HLA) antibodies, age at HTX and mitral and tricuspid regurgitation severity.

Conclusions: Standard and advanced echocardiography is confirmed to be an essential and non-invasive tool to describe pathological conditions in AdHF, to determine the best follow up timing to avoid major events or HF hospitalizations but also to early diagnose modification of physiological deformation in case of CAV of rejection or to predict an 0increase of NT-proBNP.