Sophia Antipolis, France – 27 Aug 2021: Haemodynamic guided management may reduce heart failure hospitalisations in patients with earlier stage heart failure, according to late breaking research presented in a Hot Line session today at ESC Congress 20211 and published in The Lancet.2 The observed reduction did not reach statistical significance in the overall analysis, but was statistically significant when limited to follow-up preceding the COVID-19 pandemic.
In patients with heart failure, pulmonary congestion is a major cause of poor functional capacity, reduced quality of life, heart failure hospitalisations and increased mortality. Elevated or increasing pulmonary artery pressure predicts congestion and can be tracked longitudinally with implantable devices. For most patients with increasing and/or elevated pulmonary artery pressures there is ample time to initiate therapies such as diuretics or vasodilators to reduce pulmonary artery pressure and prevent heart failure hospitalisations.
CardioMEMS is a wireless, battery-free sensor that is implanted into the pulmonary artery via a right heart catheterisation procedure to transmit pulmonary artery pressures. Pressure data are used by clinicians to titrate heart failure medications in order to control congestion, optimise pulmonary artery pressures, and prevent heart failure events. The device is approved in Europe and the US to reduce heart failure hospitalisations in patients with New York Heart Association (NYHA) class III heart failure and an admission for heart failure in the previous year.
The GUIDE-HF Trial Randomised Arm examined whether haemodynamic guided heart failure management with the CardioMEMS device reduces heart failure events and mortality in heart failure patients with NYHA class II, III and IV and either a previous heart failure hospitalisation or elevated natriuretic peptides. Principal investigator Professor JoAnn Lindenfeld of Vanderbilt University Medical Center, Nashville, US said: “The trial was designed to evaluate whether interventions with haemodynamic guided heart failure management provide benefit in earlier (NYHA class II) and later (class IV) stages of heart failure and before a heart failure hospitalisation occurs but with biomarkers indicating a potential future risk of heart failure events (i.e. elevated natriuretic peptides).”
The trial enrolled 1,000 patients with NYHA class II–IV heart failure and either a hospitalisation for heart failure within the preceding 12 months or elevated natriuretic peptide levels within 30 days prior to consent to participate in the study. All participants underwent implantation of a CardioMEMS device. Patients were then randomised 1:1 to haemodynamic guided management using transmitted pulmonary artery pressure or to a control group that did not have pulmonary artery pressures made available. Patients were blinded to the treatment allocation but the investigators were not blinded.
The primary endpoint was a composite of cumulative heart failure hospitalisations, urgent heart failure visits, and mortality. Heart failure hospitalisations were defined as an inpatient admission for more than 24 hours and an increase in heart failure therapies, primarily intravenous treatments. Urgent heart failure visits were defined as unplanned outpatient visits to a hospital or emergency department for heart failure symptoms requiring intravenous therapy for heart failure.
During a median follow-up of 11.7 months, 253 primary endpoint events occurred in the treatment group and 289 in the control group. In the overall analysis, the primary endpoint was reduced by 12% in the treatment group but did not meet statistical significance (hazard ratio [HR] 0.88; 95% confidence interval [CI] 0.74–1.05; p=0.16). However, a prespecified COVID-19 impact analysis showed a significant interaction warranting a pre-COVID-19 analysis. The pre-COVID-19 analysis assessed the primary endpoint up to 13 March 2020, the date of the national COVID emergency declaration in the US. The pre-COVID-19 analysis demonstrated a significant 19% reduction in primary endpoint events in the treatment group (HR 0.81; CI 0.66–1.00; p=0.0489).
Regarding heart failure hospitalisations, these were reduced by 17% in the treatment group in the overall analysis but just missed statistical significance (p=0.064). However, the pre-COVID analysis demonstrated a highly significant 28% reduction in heart failure hospitalisations (p=0.0072) – the same reduction achieved in the CHAMPION trial which enrolled NYHA class III heart failure patients with a previous heart failure hospitalisation.3
Neither urgent heart failure visits nor mortality were reduced independently with treatment in the overall or pre-COVID-19 analyses.
Professor Lindenfeld said: “The findings indicate that the benefits of haemodynamic guided management in reducing heart failure hospitalisations extend to patients with less severe heart failure (NYHA class II) and to those with NYHA II and III symptoms and elevated natriuretic peptides but no previous hospitalisation. The NYHA class IV heart failure patients did not show consistent results but were small in number. The COVID-19 pandemic clearly affected the outcomes of GUIDE-HF, as it did the AFFIRM-AHF trial.4 Clinical trials conducted during pandemics will require statistical analysis plans to account for their effects.”
ENDS