Cerebrovascular events after transcatheter mitral valve repair or guideline-directed medical therapy in patients with mitral regurgitation and heart failure in the COAPT trial

Background

Our knowledge regarding the risk of cerebrovascular events (CVE) in patients with heart failure (HF) and severe secondary mitral regurgitation (SMR) treated by transcatheter mitral valve repair (TMVr) is limited.

Purpose

To examine the incidence, predictors, timing, and prognostic impact of CVE in patients with heart failure and SMR treated with TMVr vs guideline-directed medical therapy (GDMT) alone.

Methods

In the COAPT trial, 614 patients with HF with moderate-to-severe or severe SMR were randomized to TMVr with the MitraClip + GDMT vs GDMT alone. After 2 years, patients who were randomized to GDMT alone could crossover and undergo TMVr. CVE (defined as stroke or TIA) were adjudicated by an independent clinical events committee.

Results

A total of 43 CVE occurred in 42 patients within 3-year follow-up (34 strokes and 9 TIAs; 1 patient had both). CVE occurred in 10.0% (n=20) of patients randomized to TMVR and 11.3% (n=22) of patients randomized to GDMT alone (p=0.53) (Figure). Of the 22 CVE in the GDMT alone group, 3 occurred after the patient had crossed over to TMVr. The incidence rates in the TMVr and GDMT groups were similar within the first 3 months (incidence rate ratio [IRR] 0.78, 95% CI 0.17–3.48, p=0.74) and between 3 months and 3 years (IRR 0.83, 95% CI 0.43–1.60, p=0.58) after randomization. After multivariable adjustment, baseline estimated glomerular filtration rate (eGFR) was associated with CVE in the overall population (HR per 5 ml/min increase in eGFR 0.91, 95% CI 0.84–0.99, p=0.03). Peripheral vascular disease was associated with CVE in patients treated by GDMT (HR=3.21, 95% CI [1.35, 7.67]) but not TMVr (HR 0.53 95% CI 0.12–2.24; p-interaction=0.04). In contrast, baseline chronic oral anticoagulation use was associated with a reduced risk of CVE in patients in the TMVr group (HR 0.18, 95% CI 0.05–0.63) but not in the GDMT alone group (HR 1.66, 95% CI 0.70–3.94; p-interaction=0.004). In a time-adjusted multivariable analysis, CVE was associated with a higher risk of death (HR 2.51, 95% CI 1.54–4.08; p=0.0002), a risk that was marked in the first 30 days after the event (HR 14.21, 95% CI 7.30–27.97, p<0.0001), and declined thereafter (HR 1.37, 95% CI 0.72–2.59, p=0.34).

Conclusions

In patients with HF and severe SMR, CVE at 3 years was not infrequent, increased linearly over time, was similar after treatment with the MitraClip and GDMT alone, and was associated with a marked increase in all-cause death. Whether anticoagulation is especially effective at preventing CVE in patients treated by TMVr, as suggested by this report, warrants further study.

Funding Acknowledgement

Type of funding sources: Private company. Main funding source(s): Abbott Figure 1