Effects of pulmonary-artery-pressure-guided remote patient management in patients with heart failure with versus without depressive symptoms: insights from MEMS-HF

Abstract Background Depressive symptoms (DS) are common in heart failure (HF) and adversely impact prognosis and quality of life (QoL). Haemodynamic-guided remote patient management (RPM) using an implantable pulmonary artery pressure (PAP) monitoring device has proven effective in reducing mortality and HF hospitalisations (HFH) and improving patient reported outcomes (PRO). Purpose This secondary analysis from MEMS-HF (CardioMEMS European Monitoring Study for Heart Failure [1]) examined whether effects of PAP-guided RPM differed in subgroups by baseline DS status. Methods MEMS-HF was a prospective open-label post-market study, which enrolled 239 patients of whom 234 received a CardioMEMS sensor at 31 centers in Germany (n=26), The Netherlands (n=4) and Ireland (n=1). All participants had New York Heart Association Class III HF, ≥1 HFH in the past 12 months, and were on optimal HF therapy as tolerated. Patients served as their own control. At baseline, 6 and 12 months, average mean PAP (PAM) was obtained over a 14-day period. DS and QoL were assessed with the 9-item Patient Health Questionnaire (PHQ-9) and the Kansas City Cardiomyopathy Questionnaire Overall Summary Score (KCCQ-OSS), respectively. All comparisons of changes were adjusted for baseline values. Annualized HFH rates were compared 1 year prior to and after implant. Kaplan-Meier plots were used to estimate HFH-free survival. Results 225 patients completed the PHQ-9, and were stratified by PHQ-9 sum-score (DS: ≥10, n=84; non-DS: <10, n=141, respectively). Patient age, sex, clinical characteristics and PAM were similar, but DS vs non-DS patients had higher body weight (p=0.044), a more recent HFH (p=0.003), less diuretics (p=0.011), more often pulmonary obstruction (p=0.002), and a lower QoL (p<0.001). Mean PHQ-9 sum-scores were 15.1±4.0 vs 4.9±2.8, respectively, p<0.001. Through 12-month follow-up, PAM declined in both subgroups (DS: -7.8±1.1mmHg; non-DS: -6.0±0.8mmHg, both p<0.001, pcomparison=0.209). PHQ-9 sum-score decreased only in DS patients (-5.7±0.5, p<0.001; non-DS: 0.1±0.4, p=0.886, p comparison=0.112), while KCCQ-OSS improved across subgroups (DS: 19.8±2.6; non-DS 8.5±1.9, both p<0.001, pcomparison=0.598). In DS and non-DS subgroups, annualized HFH rates decreased similarly (0.59 vs 1.52 and 0.57 vs 1.54 events/patient-year, hazard ratio (HR) 0.37 (0.28, 0.50) and 0.39 (0.29, 0.52), both p<0.001 vs 1 year pre-implant, regardless of baseline PAM (Figure 1). However, HFH-free survival was lower in DS patients (HR 1.62 (1.1, 2.4), p=0.015, Figure 2A), and further analyses suggested an additional adverse effect from baseline PAM >35mmHg (Figure 2B). Conclusion Regardless of baseline DS status, PAP-guided RPM reduced HFH effectively. Consistent (baseline-adjusted) improvements in PRO were numerically greater in patients with DS. Despite these improvements, patients with baseline DS had lower HFH-free survival, and elevated PAM seemed to worsen prognosis especially in this subgroup.Figure 1 Figure 2