P1483How and when to monitor decongestive treatment in hospitalized Acute Heart Failure patients-serial monitoring using lung ultrasound and focused echocardiography

Abstract

Background: In-hospital Acute Heart Failure (AHF) treatment often fails to confer adequate decongestion despite lengthy and expensive hospitalizations -Optimal monitoring should be able to identify non-responsive patients early in order to intensify treatment, but also should be able to determine whether the patient is adequately decongested prior to discharge since this might affect postdischarge prognosis Purpose: To evaluate a rapid cardiothoracic ultrasound protocol (CaTUS), combining echocardiographically derived filling pressures (E/e' and IVC index (5- class, based on maximum diameter and respiratory variation)) with lung ultrasound (LUS), for monitoring hospitalized AHF patients Methods: We enrolled 101 hospitalized AHF patients who were followed with CaTUS at baseline, 24 and 48 hours and on the day of discharge, with symptom assessment and brain natriuretic peptides (BNP) obtained simultaneously -Inclusion criteria consisted of dyspnea at rest, bilateral B-lines or pleural fluid on LUS, a medial E/e' >15 and a BNP >100 ng/l -Exclusion criteria consisted of mitral stenosis or calcification, pulmonary fibrosis, chronic dialysis, mechanical ventilation of altered mental status -Patients were divided into two groups: patients who got rid of their pulmonary congestion on LUS were called ?responders?, whereas patients discharged with residual congestion on LUS were called ?non-responders? Results: • Among responders, E/e' declined significantly faster during the first 24 hours than during the rest of hospitalization (mean slope -2.3±5.2U/day during the first 24 hours vs. -0.67±1.4U/day during rest of hospitalization, p<0.001), while no such rapid decline was seen with IVC index or BNP • Among responders, resolution of B-lines on LUS occurred at e median timepoint of 24 (IQr 12-48) hours • Eventually, responders eventually had a significantly larger decrease in all of these congestion parameters during their hospitalization as compared to nonresponders (p<0.005 for all) • Of all congestion parameters, including LUS, symptoms, E/e', fluid loss and BNP, measured both as changes during treatment and as discharge values, IVC index at discharge was the only congestion parameter significantly predicting both six-month all-cause mortality (adjusted HR 1.74, p=0.046), as well as the composite endpoint of six-month all-cause mortality or re-hospitalization for AHF (adjusted HR 1.72, p=0.005). Conclusion: • E/e' seems like the most agile parameter for monitoring early treatment, foreseeing resolution of pulmonary congestion. • Achieving a non-plethoric IVC, indicating euvolemia, seems to be associated with a substantially improved post-discharge prognosis. • Thus, IVC seems like the most important parameter for guiding AHF treatment towards end of hospitalization.