Prognostic importance of guideline-directed medical therapy after left ventricular assist devices: a 10-year single-centre experience

Introduction

Guideline-directed medical therapy (GDMT) is a cornerstone of heart failure with reduced ejection fraction (HFrEF) management, yet its long-term implementation and prognostic relevance in patients supported with left ventricular assist devices (LVAD) remain insufficiently explored.

Purpose

We aimed to evaluate longitudinal GDMT use, cumulative GDMT exposure, and their prognostic significance in LVAD recipients over a 10-year period.

Methods

We retrospectively analysed all 157 patients who underwent LVAD implantation between January 2015 and December 2024. HF therapies, including beta-blockers (BB), renin–angiotensin-system inhibitors (RASi), mineralocorticoid receptor antagonists (MRA), and sodium-glucose cotransporter-2 inhibitors (SGLT2i) were assessed at discharge and during follow-up at 6, 12, 24, 36, 48, and 60 months. The GDMT score (range 0–8) was calculated for each patient for every medication, and the average GDMT score was derived from cumulative exposure to all HF medications during follow-up. Drug doses were expressed as percentages of the guideline-recommended target doses [%MAX=(prescribed daily dose [mg])/(target daily dose [mg])*100%]. For RASi, BB, and MRA, cumulative exposure was calculated using time-weighted %MAX values [semi-annual and annual: %MAX*number of months to the next follow-up)/(12 months)], while SGLT2i exposure was expressed as the percentage of total follow-up duration.

Results

Of 157 patients (96.2% male), 28 (17.8%) died during index hospitalization and were excluded. Among the remaining 129 patients, 59 (45.7%) died during a mean follow-up of 30 ± 25 months. Overall, GDMT was underutilised at discharge, with gradual intensification and broader adoption of HF drug classes during follow-up. Survivors had higher cumulative doses of most HF therapies and achived significantly higher GDMT scores both at discharge and over the entire follow-up (Table 1). The average GDMT score was a strong predictor of survival (HR 0.769, 95%CI 0.665–0.890), along with cumulative SGLT2i (HR 0.988, p=0.03) and MRA doses (HR 0.991, p=0.04), even after adjustment for bilirubin, NT-proBNP, left ventricle ejection fraction, LVAD type, HF etiology, and atrial fibrillation. Patients achieving a GDMT score ≥5 had nearly threefold improvement in survival (HR 0.359, 95% CI 0.202–0.609).

Conclusion

In patients supported with LVADs, GDMT remains underutilised, particularly early after implantation, but its use and intensity increase progressively over time. Higher cumulative exposure to GDMT is independently associated with substantially improved long-term survival, underscoring the importance of sustained, structured, and proactive optimisation of HF pharmacotherapy in this high-risk population.

Table 1. Comparison of baseline data of survivors and non-survivors.Table 1For image description, please refer to the figure legend and surrounding text.