Virtual optimization of guideline‐directed medical therapy in hospitalized patients with heart failure with reduced ejection fraction: the IMPLEMENT‐HF pilot study
Ankeet S. Bhatt, Anubodh S. Varshney, Mahan Nekoui, Alea Moscone, Jonathan W. Cunningham, Karola Jering, Parth Patel, Lauren Sinnenberg, Thomas D. Bernier, Leo F. Buckley, Bryan M. Cook, Jillian Dempsey, Julie Kelly, Danielle Knowles, Kenneth Lupi, Rhynn Malloy, Lina Matta, Megan N. Rhoten, Krishan K. Sharma, Caroline A. Snyder, Clara Ting, Erin E. McElrath, Mary G. Amato, Maryam Alobaidly, Catherine Ulbricht, Niteesh K. Choudhry, Dale Adler, Muthiah Vaduganathan
Abstract
AIMS: Implementation of guideline-directed medical therapy (GDMT) for heart failure with reduced ejection fraction (HFrEF) remains incomplete. Non-cardiovascular hospitalization may present opportunities for GDMT optimization. We assessed the efficacy and durability of a virtual, multidisciplinary 'GDMT Team' on medical therapy prescription for HFrEF. METHODS AND RESULTS: Consecutive hospitalizations in patients with HFrEF (ejection fraction ≤40%) were prospectively identified from 3 February to 1 March 2020 (usual care group) and 2 March to 28 August 2020 (intervention group). Patients with critical illness, de novo heart failure, and systolic blood pressure <90 mmHg in the preceeding 24 hs prior to enrollment were excluded. In the intervention group, a pharmacist-physician GDMT Team provided optimization suggestions to treating teams based on an evidence-based algorithm. The primary outcome was a GDMT optimization score, the sum of positive (+1 for new initiations or up-titrations) and negative therapeutic changes (-1 for discontinuations or down-titrations) at hospital discharge. Serious in-hospital safety events were assessed. Among 278 consecutive encounters with HFrEF, 118 met eligibility criteria; 29 (25%) received usual care and 89 (75%) received the GDMT Team intervention. Among usual care encounters, there were no changes in GDMT prescription during hospitalization. In the intervention group, β-blocker (72% to 88%; P = 0.01), angiotensin receptor-neprilysin inhibitor (6% to 17%; P = 0.03), mineralocorticoid receptor antagonist (16% to 29%; P = 0.05), and triple therapy (9% to 26%; P < 0.01) prescriptions increased during hospitalization. After adjustment for clinically relevant covariates, the GDMT Team was associated with an increase in GDMT optimization score (+0.58; 95% confidence interval +0.09 to +1.07; P = 0.02). There were no serious in-hospital adverse events. CONCLUSIONS: Non-cardiovascular hospitalizations are a potentially safe and effective setting for GDMT optimization. A virtual GDMT Team was associated with improved heart failure therapeutic optimization. This implementation strategy warrants testing in a prospective randomized controlled trial.