Withdrawal of GDMT in patients with a recovered LV and reversible etiology: a scenario in valvular regurgitation

Abstract Background Long-term guideline-directed medical therapy (GDMT) is recommended for patients with reduced left ventricular ejection fraction (LVEF), and TRED-HF study demonstrated a quick relapse after medication withdrawal in those patients with dilated cardiomyopathy who had a recovered LVEF. However, it needs to be answered whether the scenario differs for patients with reversible etiologies, like those with primary valvular regurgitation who received surgical correction. Purpose It aimed to investigate the safety of treatment withdrawal in patients with residual LV remodelling after surgery for primary aortic regurgitation (AR) and/or mitral regurgitation (MR), who received GDMT and displayed a complete recovery. Methods This was a randomized, open-label, parallel trial that recruited patients with a LV end-diastolic diameter (LVEDD) 60 mm or LVEF 50% by echocardiography at 7-14 days after surgery for primary AR and/or MR. They were given GDMT and scheduled follow-up at 1, 3, 6, 9, and 12 months. Those patients with a complete recovery of LV remodelling (i.e., LVEDD 55 mm and LVEF ≥ 55% at two consecutive visits) were randomized into the treatment withdrawal or continued group. Monthly on-site visit or phone call conducted for 6 months. The primary endpoint was the rate of LV remodelling relapse (LVEDD 60 mm or LVEF 50%). The secondary endpoints included the change in 6-minute walk distance (6MWD) and Kansas City Cardiomyopathy Questionnaire (KCCQ) scores, and the rate of composite clinical outcomes. Results Eighty patients (55±13 years, 81.3% male) were enrolled who experienced a complete LV recovery after GDMT for 10±3 months. They were randomized into treatment withdrawal group (n=39) and treatment continued group (n=41) with comparable baseline characteristics. GDMT for HFrEF was stopped in all patients of the withdrawal group, 51.3% of whom used calcium channel blockers to hypertension control instead. During the 6-month follow-up, no patient died, but 5 patients experienced non-cardiovascular rehospitalization, with comparable rates between the two groups (5.1% vs. 7.3%, p=1.000). There was not a single patient who showed a LV remodeling relapse in both groups, while the LV volume and LVEF remained unchanged compared with baseline. Similar findings were also depicted in the secondary endpoints. (Table 1 and Figure 1) Conclusion Medication withdrawal appeared safe and feasible in those patients who had a complete LV structural recovery after surgery and limited period of GDMT. Certainly, it warrants a longer follow-up to differentiate between those who have already developed valvular cardiomyopathy and those who have not.Table 1 Figure 1