Estimation of left ventricular filling pressures in transplanted hearts by echocardiography – do current guidelines work?

Abstract

Abstract Introduction End-stage heart failure (HF) can be successfully treated with heart transplantation (HTx), which reduces mortality and improves quality of life (QoL). Diastolic dysfunction and subsequently increased left ventricular (LV) filling pressures can be an early sign of rejection, while later after HTx, they can be suggestive for other insults to the myocardium, for example related to transplant vasculopathy. Therefore, right heart catheterization (RHC) remains a standard follow-up examination in patients after HTx. It has been questioned, whether echocardiography can accurately predict the LV filling pressures in HTx patients, and whether echocardiographic diastolic parameters correlate with hemodynamics in this population. Therefore, our study aimed to investigate the correlation between invasive measurements and echocardiographic estimates of LV filling pressures in HTx patients. Methods We analyzed 461 consecutive HTx patients who underwent RHC and transthoracic echocardiography on the same day. Patients were classified as having elevated LV filling pressures when the pulmonary capillary wedge (PCWP) was ≥15 mmHg. Standard echocardiographic parameters of systolic and diastolic function were measured and the decision tree of the recommendations on diastolic dysfunction assessment of the American and European imaging societies (ASE/EACVI) was applied to detect elevated LV filling pressures and determine the diastolic dysfunction grade. Results The invasive measurements showed elevated LV filling pressures in 303 (66%) of HTx patients. Based on the echocardiographic parameters, HTx patients were classified into the following groups: normal diastolic function (n=151, 33%), grade I (n=87, 19%), grade II (n=21,5%), and grade III (n=151, 33%) diastolic dysfunction, and indetermined group (n=51, 11%). The PCWP values differed between the groups: 14.4±0.3 mmHg vs. 13.3±0.4 vs. 20.5±0.9 mmHg vs. 22.2±0.3 mmHg vs. 17.7±0.6 mmHg, p=0.0001, accordingly (Figure A). The PCWP showed moderate correlation with E/A (r=0.49, p=0.0001) and E/e' (r=0.40, p=0.0001) (Figures B and C, resp.). The decision tree of the ASE/EACVI recommendations predicted elevated LV filling pressures with a sensitivity of 64%, specificity of 98%, negative predictive value of 59% and positive predictive value (PPV) of 98%. Conclusions Our study is the first to demonstrate a correlation between the LV filling pressures estimated from echocardiography and invasively measured during RHC in large HTx population. Echocardiography alone can reliably describe elevated LV filling pressures with a high PPV. The sensitivity of the algorithm, however, is limited. Additional parameters need to be identified to increase the sensitivity of the current recommendations for detection of elevated filling pressures this patients population. Figure 1 Funding Acknowledgement Type of funding source: None