Abstract
Aims
The aim of this study was to analyse the feasibility, safety, and clinical benefit of the CardioMEMS HF System in different healthcare systems outside the United States of America.
Methods and results
Prospective, open-label registry of NYHA class III patients with at least one heart failure hospitalization (HFH) within 12 months before enrolment, regardless of left ventricular ejection fraction. The primary safety endpoints assess the freedom from device/system-related complications (DSRC) and freedom from pressure sensor failure (PSF) at 2 years post-implant. The primary efficacy endpoint was the rate of HFH one year before and one year after implantation.
Three hundred and four patients from 37 centres in 6 countries underwent a CardioMEMS implant procedure, which was successful in 98.3% of the cases. At 2 years, there were no DSRCs and only 1 PSF. There were 517 HFH in the year before implant compared with 144 HFH in the year post-implant (risk reduction: 69% (RR: 0.31 95% CI [0.25–0.37]; P < 0.0001). Pulmonary artery (PA) pressures were significantly lowered (mean PA pressure reduction: −3.07 ± 5.91 mmHg, P < 0.0001) with a significant improvement in functional class and quality of life (mean EQ-5D-5L visual analogue score increase 8.1 ± 22.7, P < 0.0001).
Conclusion
The results of the COAST study demonstrate that the CardioMEMS HF System is a reliable device, with no device-related complications and very few pressure sensor failures. Its use is associated with a substantial HFH risk reduction, with a significant reduction in PA pressures, an improvement in NYHA classification, and an improvement in quality of life.
Clinical registration number
ClinicalTrials.gov Identifier: NCT02954341.
Keywords: CardioMEMS HF System, Haemodynamic remote monitoring, Pulmonary artery pressure, Heart Failure Hospitalization reduction, Clinical trial results
Graphical Abstract
Graphical Abstract.
Graphical abstract. Prospective, open-label registry in 304 HF patients in NYHA class III with at least one heart failure hospitalization (HFH) within 12 months. No device/system-related complications (DSRC), and 1 patient experienced a pressure sensor failure (PSF) at 2 years. There was a 69% risk reduction in HFH in the year before implant compared with the year post-implant (RR: 0.31 95% CI [0.25–0.37]; P < 0.0001). Significant pulmonary artery (PA) pressures decreased with a significant improvement in functional class and quality of life.
Introduction
In recent years, remote monitoring has become an important part of the management of heart failure (HF) patients. However, the results of international randomized studies are disappointing, particularly for non-invasive remote monitoring,1,2 as confirmed by meta-analyses.3 In contrast, several studies have demonstrated the effectiveness of implantable remote monitoring devices, such as the CardioMEMS HF System. The pivotal CHAMPION trial (CardioMEMS Heart Sensor Allows Monitoring of Pressure to Improve Outcomes in NYHA Class III Heart Failure Patients) trial4 enrolled 550 patients with NYHA Class III and a recent HF hospitalization (HFH). At 6 months, CardioMEMS-guided management reduced HFH by 28% compared to standard care (hazard ratio (HR) = 0.72 [0.60–0.85], P = 0.002). The subsequent GUIDE-HF trial (haemodynamic-guided management of heart failure)5 expanded inclusion to 1000 patients with NYHA Class II–IV symptoms and either a prior HFH or elevated natriuretic peptides. Overall, the primary composite endpoint of all-cause mortality and HF events was not significantly reduced (HR = 0.88 [0.74–1.05]). But, in a prespecified pre-COVID-19 analysis, CardioMEMS-guided care achieved a 19% reduction in the primary endpoint and a 24% reduction in HF events (HR = 0.76 [0.61–0.95]). The third randomized trial, the MONITOR-HF (remote haemodynamic monitoring of pulmonary artery pressures in patients with chronic heart failure),6 which randomized 348 NYHA class III patients and recent HFH. After one year, CardioMEMS-guided management led to a 44% reduction in HFH (secondary objective, HR = 0.56, 95% CI 0.38–0.82) with an improvement in quality of life (primary objective). Successive observational studies as the MEMS-HF (CardioMEMS European Monitoring Study for Heart Failure),7 the CardioMEMS Post Approval Study8 reported positive results. A recent meta-analysis of observational studies in 3308 patients,9 confirms the consistency of the results, with a 61% of HFH [HR = 0.39, 0.31–0.47] with a remarkable safety profile of the device.
The CardioMEMS HF System Post-Market Study (COAST) was an open-label study implemented to examine the real-world feasibility, safety, and clinical benefit of a haemodynamic-guided management strategy for patients with HF in different medical systems within Europe and Australia. Three previous publications presented the results of subgroups of patients from the COAST trial, which included patients from the United Kingdom (UK),10 France,11 and Australia.12 This publication reports the complete COAST trial results.
Methods
The COAST study was designed as a prospective, open-label registry with the aim of examining the safety and feasibility of symptomatic HF patient remote management using pulmonary artery (PA) pressures assessed via the CardioMEMS HF System in varying healthcare systems outside the United States of America (US). Details concerning the rationale and the design of this study have been reported previously.13 The study protocol was approved by all responsible ethics committees and was conducted in accordance with the Declaration of Helsinki principles. All participants provided written informed consent before any study-related procedure. Patients were required to be in New York Heart Association (NYHA) class III and have undergone at least one heart failure hospitalization (HFH) within 12 months before enrolment, regardless of left ventricular ejection fraction (LVEF). Subjects with reduced LVEF (defined as < 40%, HFrEF) should have been under beta blocker treatment for at least 3 months and an angiotensin-converting enzyme inhibitor (ACEi) or an angiotensin receptor blocker (ARB) or an angiotensin receptor blocker/neprilysin inhibitor (ARNI) for 1 month, unless the investigator deemed the patient to be intolerant to such therapy. Patients with LVEF above 40% were considered to have a preserved LVEF according to the definition of ESC guidelines.14 The complete study inclusion and exclusion criteria are listed in Supplementary material online, Table S1 of the Supplementary Materials.
The COAST study consented a total of 321 subjects at 38 sites in 6 countries (Australia, Belgium, Denmark, France, Italy, and United Kingdom, participating centres list in Supplementary Materials), of which 304 underwent a CardioMEMS implantation, which was successful in 299 cases (Figure 1). These patients were instructed on how to operate the CardioMEMS Patient Electronics System to obtain daily pressure measurements from the implanted sensor. Guidelines regarding euvolemic ranges and general strategies to achieve haemodynamic stability were provided to the Investigators via the study protocol (see Supplementary Materials—Study Definitions). The rationale behind these thresholds indicates volume shifts as a contributing factor to PA pressure variations over time, with a fluid overload or reduction resulting in a PA pressure increase or decrease outside the patient’s baseline, respectively.
Figure 1.
Consort diagram: the consort diagram reports the disposition of patients at each follow-up interval. The COAST study included two populations: out of the 321 patients signing the informed consent, 304 underwent a CardioMEMS implant procedure and constituted the Safety Endpoint Population, while the 299 patients who had a successful implant constituted the Effectiveness Endpoint population.
Primary endpoints
All patients with an attempted sensor implant constituted the Safety Endpoint Population, while all subjects with a successful implant contributed to the Effectiveness Endpoint Population. Subjects with attempted but unsuccessful implantations were followed for 30 days to evaluate post-procedural safety. The Safety Endpoint was a combination of freedom from device/system-related complications (DSRC) and freedom from pressure sensor failure (PSF) at 2 years post-implant (details on DSRC and PSF are provided in Supplementary Materials) compared to a pre-specified performance goal of 80% and 90%, respectively. The effectiveness endpoint compared the rate of HFH during the 12-month period before the sensor was implanted to the 12-month period after implantation.
Secondary/exploratory endpoints
Subject’s functional class and quality of life were collected via NYHA class assessment and the EQ-5D-5L questionnaire, administered at baseline, 6-month, 12- and 24-month follow-up visits.
The PA pressure change over time was evaluated using absolute change and with the Area Under the Curve (AUC) methodology, which quantifies the duration of time that a patient spends at a pressure lower (or higher) than their baseline PA pressure.
Statistical analysis
Data were summarized using univariate statistics (e.g. N, mean, standard deviation, median, minimum, and maximum) or frequency (e.g. N, %) as appropriate for continuous or categorical variables, respectively. Enrolment was defined as having a successful sensor implant. The primary safety endpoints were analysed at 24 months post enrolment while the primary efficacy endpoint was assessed at 12 months post enrolment.
The primary safety analysis was based on the following objective performance criteria: (a) the lower limit of the one-sided 97.5% confidence limit interval on the freedom from DSRC rate at 24 months is greater than 80% and (b) the lower limit of the one-sided 97.5% confidence limit on the freedom from PSF rate at 24 months is greater than 90%. Both performance goals for DSRC and sensor failure were based on the CHAMPION trial design15 and results,16 so that the populations of these two studies could be pooled for additional analyses. The COAST study was deemed to have provided positive safety results if both primary safety endpoints were statistically significant (i.e. P < 0.025).
The primary efficacy analysis compared the annualized HFH rate at one year after implant vs. the HFH rate in the year prior to enrolment using a one-sample, one-sided Poisson rate test. The primary effectiveness endpoint was considered met if the one-sided, upper 97.5% confidence limit for the rate parameter reported after enrolment was lower than the rate in the year before enrolment.
Additional analyses included PA pressure changes (systolic, mean, and diastolic) for the effectiveness population and subjects with HF with preserved ejection fraction (HFpEF) and HF with reduced ejection fraction (HFrEF), home reading compliance (defined as the number of days with a reading divided by the total number of days of patient follow-up spent outside the hospital), and the weekly compliance (defined as the number of weeks with at least one daily reading divided by the total weeks of the follow-up period spent outside the hospital). Events were reported by investigators responsible for the patients’ clinical care.
Comparisons between quantitative variables were performed by a paired or an unpaired Student’s t test.
Role of the funding source
The COAST study was sponsored and funded by Abbott. Data were collected in an electronic database developed by the sponsor. The sponsor also performed the analysis for this manuscript following the Authors’ directives and decisions. The COAST authors of this manuscript had access to the study data and contributed to its preparation by interpreting the data, determining the modality of analysis, and presenting the results.
Results
Between July 2016 and December 2021, 321 patients were consented to participate in the COAST Study. Seventeen (17) subjects were withdrawn from the study before undergoing a CardioMEMS sensor implant, mainly due to consent withdrawal or change in medical condition, while 304 underwent an attempted procedure and constituted the study’s Safety population. Of these 304 patients, two hundred ninety-nine (299, 98.3%) received a CardioMEMS sensor, representing the study’s Effectiveness population. Baseline characteristics of the Safety and Effectiveness populations are reported in Table 1, with the latter further detailed between HFpEF and HFrEF in Supplementary material online, Table S2.
Table 1.
Baseline data: safety and effectiveness populations
| Safety Population (N = 304) |
Effectiveness Population (N = 299) |
|
|---|---|---|
| Age (year) | 68.4 ± 11.7 (304) | 68.5 ± 11.7 (299) |
| Male | 73.0% (222/304) | 72.9% (218/299) |
| Ischaemic Cardiomyopathy | 45.4% (138/304) | 45.8% (137/299) |
| Preserved Ejection Fraction (%) | 36.8% (109/296) | 36.4% (106/291) |
| BMI (kg/m2) | 28.87 ± 5.98 (304) | 28.85 ± 5.94 (299) |
| Hypertension | 54.6% (166/304) | 54.8% (164/299) |
| Diabetes mellitus | 43.1% (131/304) | 43.1% (129/299) |
| Chronic obstructive pulmonary disease | 22.4% (68/304) | 22.4% (67/299) |
| Chronic kidney disease | ||
| Stage 3 | 61.8% (188/304) | 62.5% (187/299) |
| Stage 4 | 7.6% (23/304) | 7.7% (23/299) |
| Stage 5 | 0.3% (1/304) | 0.3% (1/299) |
| Glomerular filtration rate (mL/min/1.73 m2) | 51.82 ± 19.84 (304) | 51.53 ± 19.85 (299) |
| Beta Blocker | 81.6% (248/304) | 82.9% (248/299) |
| ACE inhibitor/ARB/ARNi | 74.0% (225/304) | 75.3% (225/299) |
| Beta Blocker + ACE inhibitor/ARB/ARNi | 66.4% (202/304) | 67.6% (202/299) |
| Mineralocorticoid Receptor Antagonists (MRA) | 63.2% (192/304) | 64.2% (192/299) |
| Diuretics | 95.1% (289/304) | 96.7% (289/299) |
| Loop Diuretic | 94.7% (288/304) | 96.3% (288/299) |
| Thiazide Diuretic | 15.5% (47/304) | 15.7% (47/299) |
| Other* | 3.0% (9/304) | 3.0% (9/299) |
| CRT/CRT-D | 29.6% (90/304) | 29.1% (87/299) |
| ICD | 23.7% (72/304) | 23.7% (71/299) |
| NYHA Class | ||
| Class I | 0.0% (0/304) | 0.0% (0/299) |
| Class II | 0.0% (0/304) | 0.0% (0/299) |
| Class III | 100.0% (304/304) | 100.0% (299/299) |
| Class IV | 0.0% (0/304) | 0.0% (0/299) |
| EQ-5D-5L VAS | 60.0 ± 21.1 (289) | 60.0 ± 21.1 (289) |
Data are reported as mean ± SD (n) for continuous or discrete variables and % (n/total) for categorical variables. BMI: Body Mass Index, CRT: Cardiac Resynchronization Therapy, CRT-D: Cardiac Resynchronization Therapy with Defibrillation backup, ICD: Implantable Cardiac Defibrillator ACE: Angiotensin-converting-enzyme, ARB: Angiotensin receptor blockers, ARNi: angiotensin receptor/neprilysin inhibitor, NYHA: New York Heart Association. *Other diuretics include: potassium-sparing diuretics and carbonic anhydrase inhibitors.
Sensor implant was unsuccessful in five patients due to haemoptysis (n = 1), anatomical constraint (n = 2), inability to gain venous access (n = 1) or inability to advance the delivery system beyond the pulmonary valve (n = 1). These patients were followed for 30 days for safety prior to being withdrawn from the study. Successive follow-up visits at 1, 6, 12, 18, and 24 months were completed by 294, 268, 239, 208, and 183 patients, respectively (Figure 1).
The study’s primary Safety Endpoints, calculated at 24 months post-implant as freedom from DSRC in the Safety population (n = 304) and freedom from sensor failure in the Effectiveness population (n = 299), were significantly higher than the performance goals pre-specified in the study protocol (80% and 90%, respectively). There were no reported DSRC at two years, therefore freedom from DSRC was 100% (P < 0.0001), and 298 out of 299 patients were free from sensor failure, resulting in a 99.7% freedom from PSF (P < 0.0001).
The primary Effectiveness endpoint was calculated in the Effectiveness population as the annualized rate of HFH in the 12 months prior to implant compared to 12 months post-implant. There were 517 reported events in the 12 months prior to implant (annualized 1.59 events/year [1.47–1.71]), reduced to 144 in the 12 months post-implant (annualized 0.49 events/year [0.40–0.59], respectively). The study therefore reported an annualized HFH rate reduction of 69% (RR 0.31 (95% CI: 0.25, 0.37, P < 0.0001; Figure 2).
Figure 2.
Primary effectiveness endpoint—heart failure hospitalization (HFH) reduction: the heart failure hospitalization reduction, calculated for the effectiveness endpoint population, compares the annualized HFH rate in the 12 months prior to CardioMEMS implant (1.583 events per pt-year) with the annualized HFH rate in the 12 months post-implant (0.488 events per pt-year). The HFH rate reduction was 69%.
The overall survival rate in the study was 86.7% at one year and 69.1% at 2 years (see Supplementary material online, Figure S1). A total of 89 (29.3%) patient deaths—mainly due to pump failure—occurred in the Safety population as reported by the study Investigators (see Supplementary material online, Table S3). An additional analysis was performed to determine if the relatively high death rate observed in the study would impact the Effectiveness endpoint. A conservative approach to this assessment assumes death as a failure in the effectiveness of the CardioMEMS device, therefore combining HFH and death. In the 12 months post implant, the combination of HFH and deaths was of 187 events (annualized 0.63 event/year [0.53 0.76]) compared to 517 events [1.57 event/year [1.46–1.70], indicating a rate reduction of 60% (RR 0.40 (95% CI: 0.33, 0.49, P < 0.0001), demonstrating no impact of death rate on the effectiveness endpoint. Cumulative incidence of HFH and combined HFH and death are reported in Supplementary material online, Figures S2 and S3. A subgroup analysis assessing the impact of baseline demographics and clinical characteristics was also performed which showed no impact of sex, age (< 60, between and 75, ≥ 75 years old), reduced or preserved ejection fraction, HF aetiology (ischaemic vs. non-ischaemic), presence or not of diabetes, renal insufficiency subgroups (stage 1–2, 3 or 4–5), BMI subgroups (≥ 30 or < 30 kg/m²), and presence of implantable defibrillator or resynchronization therapy (ICD/CRT-D) on the HFH reduction (Table 2, Figure 3).
Table 2.
Effectiveness endpoint—subgroups analysis
| 12 Months prior to Implant |
12 Months after to Implant |
Rate Ratio (95% CI) | Interaction Term P-Value |
||
|---|---|---|---|---|---|
| Male (n = 218) | Number of Events | 358 | 103 | 0.32 (0.25, 0.41) | P = 0.4621 |
| Annualized Event Rate (95% CI) |
1.50 (1.39, 1.63) |
0.48 (0.38, 0.61) |
|||
| Female (n = 81) | Number of Events | 159 | 41 | 0.28 (0.20, 0.39) | |
| Annualized Event Rate (95% CI) |
1.81 (1.54, 2.12) |
0.50 (0.35, 0.70) |
|||
|
Age < 60
(n = 58) |
Number of Events | 87 | 31 | 0.37 (0.25, 0.55) | P = 0.4549 |
| Annualized Event Rate (95% CI) |
1.39 (1.17, 1.64) |
0.51 (0.34, 0.76) |
|||
|
Age between 60 and 75
(n = 140) |
Number of Events | 253 | 62 | 0.27 (0.20, 0.36) | |
| Annualized Event Rate (95% CI) |
1.65 (1.49, 1.84) |
0.45 (0.34, 0.60) |
|||
|
Age ≥ 75
(n = 101) |
Number of Events | 177 | 51 | 0.33 (0.23, 0.47) | |
| Annualized Event Rate (95% CI) |
1.60 (1.40, 1.83) |
0.53 (0.37, 0.75) |
|||
| LVEF ≥ 40% (N =106) | Number of Events | 190 | 45 | 0.25 (0.17, 0.36) | P = 0.1144 |
| Annualized Event Rate (95% CI) |
1.65 (1.45, 1.88) |
0.41 (0.28, 0.61) |
|||
| LVEF < 40% (N =185) | Number of Events | 316 | 98 | 0.35 (0.28, 0.44) | |
| Annualized Event Rate (95% CI) |
1.56 (1.42, 1.72) |
0.55 (0.44, 0.69) |
|||
| Ischaemic (N =137) | Number of Events | 251 | 63 | 0.28 (0.20, 0.39) | P = 0.4276 |
| Annualized Event Rate (95% CI) |
1.68 (1.50, 1.89) |
0.47 (0.34, 0.66) |
|||
| Non-Ischaemic (N =162) | Number of Events | 266 | 81 | 0.33 (0.26, 0.42) | |
| Annualized Event Rate (95% CI) |
1.50 (1.36, 1.66) |
0.50 (0.39, 0.63) |
|||
|
With Diabetes
(n = 129) |
Number of Events | 228 | 62 | 0.30 (0.23, 0.39) | P = 0.7697 |
| Annualized Event Rate (95% CI) |
1.62 (1.46, 1.80) | 0.48 (0.37, 0.63) | |||
|
Without Diabetes
(n = 170) |
Number of Events | 289 | 82 | 0.32 (0.24, 0.42) | |
| Annualized Event Rate (95% CI) |
1.56 (1.40, 1.74) | 0.49 (0.37, 0.65) | |||
|
Renal Insufficiency Stage 1 or 2
(n = 88) |
Number of Events | 136 | 34 | 0.27 (0.19, 0.39) | P = 0.6902 |
| Annualized Event Rate (95% CI) |
1.41 (1.22, 1.61) | 0.38 (0.26, 0.56) | |||
|
Renal Insufficiency Stage 3
(n = 187) |
Number of Events | 331 | 94 | 0.31 (0.24, 0.41) | |
| Annualized Event Rate (95% CI) |
1.63 (1.47, 1.79) | 0.51 (0.40, 0.65) | |||
|
Renal Insufficiency Stage 4 or 5
(n = 24) |
Number of Events | 50 | 16 | 0.37 (0.19, 0.69) | |
| Annualized Event Rate (95% CI) |
1.92 (1.58, 2.33) | 0.70 (0.39, 1.28) | |||
|
BMI ≥ 30 Kg/m²
(n = 118) |
Number of Events | 207 | 63 | 0.33 (0.24, 0.45) | P = 0.5407 |
| Annualized Event Rate (95% CI) |
1.61 (1.43, 1.81) | 0.53 (0.40, 0.71) | |||
|
BMI < 30 Kg/m²
(n = 181) |
Number of Events | 310 | 81 | 0.29 (0.22, 0.38) | |
| Annualized Event Rate (95% CI) |
1.57 (1.42, 1.73) | 0.46 (0.35, 0.60) | |||
| With ICD/CRT-D (N = 151) | Number of Events | 257 | 83 | 0.36 (0.27, 0.47) | P = 0.1077 |
| Annualized Event Rate (95% CI) |
1.56 (1.40, 1.75) |
0.56 (0.43, 0.73) |
|||
| Without ICD/CRT-D (N =148) | Number of Events | 260 | 61 | 0.26 (0.19, 0.34) | |
| Annualized Event Rate (95% CI) |
1.61 (1.45, 1.78) |
0.42 (0.31, 0.55) |
Figure 3.
Primary effectiveness endpoint—subgroup analysis: the forest plot of the annualized heart failure hospitalization reduction between different subgroups confirmed the significant reduction in HFH regardless of the analysed conditions. HFpEF: Heart Failure with preserved Ejection Fraction; LVEF: Left Ventricular Ejection Fraction; HFrEF: Heart Failure with reduced Ejection Fraction; CRT-D: Cardiac Resynchronization Therapy with Defibrillation backup; ICD: Implantable Cardiac Defibrillator
Over the 24 months of haemodynamic guided care follow-up, pulmonary pressures were significantly lowered in PA Systolic (−3.90 ± 7.88 mmHg, P < 0.0001), PA diastolic (−2.51 ± 4.77 mmHg, P < 0.0001), and PA mean pressures (−3.07 ± 5.91 mmHg, P < 0.0001, Table 3 and Supplementary material online, Figure S4). The AUC (mmHg-day) reduction at two years was significant for all three PA pressure parameters (−2355.66 ± 4996.70 systolic; −1561.52 ± 3020.45 diastolic; and −1892.21 ± 3726.16 mmHg-days, mean; P < 0.0001, Table 3 and Figure 4).
Table 3.
PA pressures and AUC changes at 24 months
| Baseline | 24 Months | P-value | |
|---|---|---|---|
| PA Systolic Pressure | 49.02 ± 15.79 (297) | 42.47 ± 14.44 (121) | − |
| Baseline to 24-Month Average Pressure Change | n/a | −3.90 ± 7.88 (297) | P < 0.0001 |
| Baseline to 24-Month AUC | n/a | −2355.66 ± 4996.70 (297) | P < 0.0001 |
| PA Mean Pressure | 33.66 ± 10.84 (297) | 28.89 ± 10.34 (121) | − |
| Baseline to 24-Month Average Pressure Change | n/a | −3.07 ± 5.91 (297) | P < 0.0001 |
| Baseline to 24-Month AUC | n/a | −1892.21 ± 3726.16 (297) | P < 0.0001 |
| PA Diastolic Pressure | 23.57 ± 8.11 (297) | 20.03 ± 8.20 (121) | − |
| Baseline to 24-Month Average Pressure Change | n/a | −2.51 ± 4.77 (297) | P < 0.0001 |
| Baseline to 24-Month AUC | n/a | −1561.52 ± 3020.45 (297) | P < 0.0001 |
Data are presented as mean ± SD (n). AUC: Area Under the Curve; PA: Pulmonary Artery.
Figure 4.
Pulmonary pressure: area under the curve (AUC) changes: the area under the curve (AUC) represents the duration of time that a patient spends at a pressure lower (or higher) than their baseline PA pressure.
There was a significant improvement in NYHA classification over the 24-month follow-up. All patients were required to be classified as NYHA Class III at baseline, with 31.8% improving to Class II and 6.0% improving to Class I at 24 months (see Supplementary material online, Table S4).
The changes to the patient’s Quality of Life, measured using the EuroQoL Five Dimensions Questionnaire (EQ-5D-5L) and administered at each follow-up visit, are reported in Table 4. Compared to Baseline, results at 12 and 24 months indicate a significantly better overall state (VAS, P = 0.0008 and P < 0.0001). No changes were observed in the individual questionnaire components (Mobility, Self-Care, Usual Activities, Pain/Discomfort, and Anxiety/Depression).
Table 4.
Patient's quality of life (EQ-5d-5L) 12 and 24 months paired analysis
| Baseline | 12 Month | 12 Month Change from Baseline |
24 Month | 24 Month Change from Baseline |
P-Valueb | |
|---|---|---|---|---|---|---|
| EQ-5D VAS Score | 0.0008c < 0.0001d |
|||||
| Mean ± SD (n) | 60.1 ± 22.0 (147) | 66.8 ± 20.9 (147) | 6.7 ± 23.6 (147) | 68.2 ± 20.1 (147) | 8.1 ± 22.7 (147) | |
| Median (Q1, Q3) | 60.0 (45.0, 75.0) | 70.0 (50.0, 80.0) | 0.0 (−5.0, 20.0) | 70.0 (50.0, 80.0) | 5.0 (−10.0, 20.0) | |
| Range (min, max) | (1, 100) | (15, 100) | (−60, 97) | (10, 100) | (−40, 97) | |
| [95% Confidence Interval]a | [56.5, 63.7] | [63.4, 70.2] | [2.9, 10.5] | [64.9, 71.5] | [4.4, 11.8] | |
| EQ-5D Index Score | 0.0489c 0.3241d |
|||||
| Mean ± SD (n) | 0.70 ± 0.22 (132) | 0.67 ± 0.23 (132) | −0.04 ± 0.21 (132) | 0.68 ± 0.28 (132) | −0.02 ± 0.25 (132) | |
| Median (Q1, Q3) | 0.73 (0.59, 0.84) | 0.69 (0.56, 0.82) | −0.01 (−0.16, 0.07) | 0.73 (0.55, 0.89) | −0.00 (−0.15, 0.13) | |
| Range (min, max) | (0.14, 1.00) | (0.03, 1.00) | (−0.59, 0.55) | (−0.26, 1.00) | (−0.75, 0.66) | |
| [95% Confidence Interval]a | [0.67, 0.74] | [0.63, 0.71] | [−0.07, −0.00] | [0.63, 0.73] | [−0.06, 0.02] | |
| Mobility Score | 0.2108c 0.6160d |
|||||
| Mean ± SD (n) Median (Q1, Q3) |
2.2 ± 1.0 (151) | 2.3 ± 1.1 (151) | 0.1 ± 1.0 (151) | 2.1 ± 1.2 (151) | −0.0 ± 1.1 (151) | |
| Range (min, max) | 2.0 (1.0, 3.0) (1, 4) |
2.0 (1.0, 3.0) (1, 5) |
0.0 (0.0, 1.0) (−3, 3) |
2.0 (1.0, 3.0) (1, 5) |
0.0 (−1.0, 1.0) (−3, 3) |
|
| [95% Confidence Interval]a | [2.0, 2.4] | [2.1, 2.5] | [−0.1, 0.3] | [2.0, 2.3] | [−0.2, 0.1] | |
| Self-Care Score | 0.6781c 0.8683d |
|||||
| Mean ± SD (n) | 1.5 ± 0.8 (151) | 1.4 ± 0.8 (151) | −0.0 ± 0.8 (151) | 1.5 ± 0.9 (151) | 0.0 ± 1.0 (151) | |
| Median (Q1, Q3) | 1.0 (1.0, 2.0) | 1.0 (1.0, 2.0) | 0.0 (0.0, 0.0) | 1.0 (1.0, 2.0) | 0.0 (0.0, 0.0) | |
| Range (min, max) | (1, 5) | (1, 5) | (−3, 3) | (1, 5) | (−3, 4) | |
| [95% Confidence Interval]a | [1.3, 1.6] | [1.3, 1.6] | [−0.2, 0.1] | [1.3, 1.6] | [−0.1, 0.2] | |
| Usual Activities Score | 0.4961c 0.8065d |
|||||
| Mean ± SD (n) | 2.1 ± 1.0 (151) | 2.2 ± 1.1 (151) | 0.1 ± 1.2 (151) | 2.1 ± 1.2 (151) | 0.0 ± 1.3 (151) | |
| Median (Q1, Q3) | 2.0 (1.0, 3.0) | 2.0 (1.0, 3.0) | 0.0 (−1.0, 1.0) | 2.0 (1.0, 3.0) | 0.0 (−1.0, 1.0) | |
| Range (min, max) | (1, 5) [1.9, 2.3] |
(1, 5) [2.0, 2.4] |
(−3, 3) [−0.1, 0.3] |
(1, 5) [1.9, 2.3] |
(−3, 4) [−0.2, 0.2] |
|
| [95% Confidence Interval]a | 2.0 ± 1.0 (152) | 2.1 ± 1.1 (152) | 0.1 ± 1.2 (152) | 2.0 ± 1.1 (152) | −0.0 ± 1.1 (152) | |
| Pain/Discomfort Score | 0.1642c 0.9407d |
|||||
| Mean ± SD (n) | 2.0 (1.0, 3.0) | 2.0 (1.0, 3.0) | 0.0 (−1.0, 1.0) | 2.0 (1.0, 3.0) | 0.0 (−1.0, 1.0) | |
| Median (Q1, Q3) | (1, 4) | (1, 5) | (−2, 4) | (1, 5) | (−3, 3) | |
| Range (min, max) | [1.8, 2.1] | [1.9, 2.3] | [−0.1, 0.3] | [1.8, 2.1] | [−0.2, 0.2] | |
| [95% Confidence Interval]a | 1.6 ± 0.9 (150) | 1.7 ± 0.9 (150) | 0.1 ± 0.9 (150) | 1.7 ± 1.0 (150) | 0.1 ± 1.0 (150) | |
| Anxiety/Depression Score | 0.0794c 0.0628d |
|||||
| Mean ± SD (n) | ||||||
| Median (Q1, Q3) | 1.0 (1.0, 2.0) | 1.0 (1.0, 2.0) | 0.0 (0.0, 0.0) | 1.0 (1.0, 2.0) | 0.0 (0.0, 0.0) | |
| Range (min, max) | (1, 5) | (1, 5) | (−2, 4) | (1, 5) | (−3, 4) | |
| [95% Confidence Interval]a | [1.5, 1.7] | [1.6, 1.9] | [−0.0, 0.3] | [1.6, 1.9] | [−0.0, 0.3] |
aBy normal approximation.
bFrom paired t-test.
cComparing 12-Month Change from Baseline.
dComparing 24-Month Change from Baseline.
Patients’ compliance, in terms of daily and weekly upload rates, remained high throughout the study: 81.7% of the patients transmitted readings daily, and 93.3% provided weekly pressure readings (see Supplementary material online, Figure S5).
At 24 months, there was a total of 4484 recorded changes to medications. Most of these changes were in diuretics, accounting for 3298 changes (74%, of which 1781 were dosage increases, 1517 decreases), and changes in ACE/ARB/ARNi accounted for 494 changes (276 titration increases and 218 decreases). Approximately half of the titration changes (2177 out of 4484, 48.5%) were due to PAP variations outside of the suggested euvolemic ranges that were identified by remote haemodynamic monitoring: 1520 changes were implemented following an increase and 667 following a decrease in PAP (see Supplementary material online, Table S5 and Supplementary material online, Table S6).
Discussion
The COAST Study demonstrated that the CardioMEMS is a reliable device with no DSRC and a very low rate of PSF after 2 years of follow-up. Moreover, the implantation procedure is safe with a complication rate lower than 2%. These results corroborate prior trial results. As this is an open, non-randomized study with no control group, the clinical effectiveness was analysed as a comparison of event rates before and after device implantation, and its interpretation should include the effects of all the medical management around the device. The significant reduction in HFH demonstrated in multiple countries with different healthcare access, management, and clinical practice suggests a possible improvement in the global patient care of the CardioMEMS HF System, as demonstrated in previous randomized and observational studies4,16 We found an impressive 69% annualized HFH rate reduction after the implantation of the CardioMEMS. However, it is important to note that the impact of the CardioMEMS HF System on HFH is more pronounced in the open registries than in the three randomized controlled trials. The absence of a control group is undoubtedly one of the main factors explaining this difference.
The remote monitoring of trends in pulmonary pressure allows timely adjustments in the pharmacological treatment and optimizes volume status, which can improve the patients’ quality of life by avoiding exacerbation of HF symptoms. In patients with optimized treatment, the adjustment of diuretics according to the evolution of pulmonary pressures constituted 77% of all the therapeutic changes recorded in this study, both in up- and down-titration, undoubtedly contributing to the improvement in the quality of life of the patients. It should be noted that this device is not intended to quickly modify the pathophysiology of heart failure, unlike beta-blockers, renin-angiotensin system blockers, mineralocorticoid receptor antagonists, and gliflozins. However, when the euvolemia is obtained, there is room for haemodynamic-guided optimization of HF treatments. A recent analysis of a large CardioMEMS population of more than 4300 patients17 determined that changes in PA diastolic pressure (from baseline to 6 months) are an independent predictor of a change in all-cause mortality. For each 1 mmHg increase or decrease in PA diastolic pressure, there is a 3% of mortality risk increase or decrease in mortality risk, regardless of the ejection fraction. While a significant effect on mortality is not expected in a short time, a strong correlation between the number of HFH and mortality risk has been demonstrated,18 therefore reducing repeated hospitalizations due to HF can subsequently have positive effects on patient’s survival.
The significant improvement in NHYA classification and quality of life is also consistent with previous findings, although different questionnaires were used, such as the Kansas City Cardiomyopathy Questionnaire (KCCQ) in the GUIDE-HF19 and in the MONITOR-HF6 trials. The rapid improvement in NYHA classification observed from the first month can be explained by therapeutic optimization that is not limited to the diuretic dose alone. However, a placebo-like effect on a subjective parameter cannot be ruled out, linked to a feeling of security experienced by the patient thanks to the daily monitoring of an invasive parameter.
Another important, consistent, surprising result is the high rate of compliance with the CardioMEMS in most of the studies, either randomized or not. This was not the case with non-invasive telemonitoring, where the compliance with monitoring devices is limited. For example, in the Optimization of the Ambulatory Monitoring for Patients With Heart Failure by Telecardiology, the OSICAT trial, the adherence to daily self-measurement of body weight was 59.9 ± 35.5% or in the Better Effectiveness After Transition–Heart Failure (BEAT-HF) trial, the adherence during the first month was greater than 50% in only 55.4% of the patients.2,20 The reason behind such a difference is not certain, but it may be due to patient’s selection, the innovative nature of the device, the experience of the nurses or doctors, or perhaps lack of confidence in the information obtained by non-invasive devices. Further studies are required to improve our understanding of compliance with devices.
Limitations
One limitation of the present study is the design as an open-label, single-arm, unblinded registry without a control comparison. In this open registry, there was not an independent adjudication committee; however, major events and pre-enrolment hospitalizations data were monitored both locally and remotely to ensure a high level of reliability. We cannot exclude competitive risk regarding mortality. Furthermore, this study population represents patients with severe HF symptoms treated by HF specialists in selected investigational sites, which could have decreased the risk of DSRC or implant failure and/or improved the overall management of HF at baseline. This limits the generalizability of these results for the average practitioner. Precise medication and titration changes that occurred during the follow-up are not in the scope of this paper, which is focused on the primary study endpoints and patient-reported outcomes. Additional analysis is needed to assess the relationship between drug class, titration, HF rates, symptoms, and its impact on pulmonary pressure.
Conclusion
The results of the COAST study demonstrate not only the safety of the CardioMEMS HF System, with safe implantation and very few device complications at 24 months, but also its effectiveness with a significant reduction in pulmonary pressures and annualized HFH rates, in conjunction with improvements in NYHA class and quality of life. In addition, a high patient’s compliance rate was maintained throughout the duration of the study.
Supplementary Material
Contributor Information
Pascal de Groote, CHU Lille, Service de Cardiologie, Lille F-59000, Institut Coeur Poumon, Boulevard du Professeur J Leclercq, CHU, 59000 Lille, France; Inserm U1167, Institut Pasteur de Lille, Lille F-59000, 1, rue du Professeur Calmette, 59019 Lille Cedex, France.
Scott McKenzie, The Prince Charles Hospital, 627 Rode Road, Chermside, QLD 4032, Australia.
Andrew Flett, University Hospital Southampton NHS Foundation Trust, Tremona Road, Southampton, Hampshire, SO16 6YD, United Kingdom.
Paul Foley, Wiltshire Cardiac Centre, The Great Western Hospital, Marlborough Road, Swindon, Wiltshire, SN3 6BB, United Kingdom.
Kasper Rossing, Rigshospitalet, Blegdamsvej 9, 2100 Copenhagen, Denmark.
Michele Ciccarelli, Department of Medicine, Surgery and Dentistry. University of Salerno, Via S. Allende, Baronissi (Salerno), 84081 Italy and Unit of Cardiac Rehabilitation and Heart Failure, AOU OO. RR. San Giovanni di Dio Ruggi d’Aragona, Largo Città Ippocrate, Salerno, 84131, Italy.
Anne-Catherine Pouleur, Cardiovascular Department, Cliniques Universitaires Saint-Luc, Avenue Hippocrate, 10, Brussels 1200, Belgium.
Carlo Gazzola, Abbott, 100 Abbott Park Rd, North Chicago, IL 60064, USA.
Eunyoung Park, Abbott, 100 Abbott Park Rd, North Chicago, IL 60064, USA.
François Roubille, PhyMedExp, INSERM, CNRS, Cardiology Department, INI-CRT, Université de Montpellier, Montpellier 34295, France.
Data availability
The data from this study will not be made publicly available. However, the authors and the sponsor are committed to working with other investigators on projects of mutual interest that might include data sharing. Any such inquiries should be made to the sponsor, Abbott.
Supplementary material
Supplementary material is available at European Heart Journal Open online.
Funding
The COAST study was funded by Abbott.
Lead author biography
Pascal de Groote, MD, PhD, FESC is a French cardiologist specializing in heart failure, cardiomyopathies, cardiac orphan diseases and pulmonary hypertension. He heads one of the cardiology departments in the Heart and Lung Institute, University Hospital of Lille. He is Involved in research, member of the Inserm U1167 research unit at the Institut Pasteur de Lille. Member of the French Society of Cardiology and the European Society of Cardiology.
References
- 1. Koehler F, Koehler K, Deckwart O, Prescher S, Wegscheider K, Kirwan BA, Winkler S, Vettorazzi E, Bruch L, Oeff M, Zugck C, Doerr G, Naegele H, Stork S, Butter C, Sechtem U, Angermann C, Gola G, Prondzinsky R, Edelmann F, Spethmann S, Schellong SM, Schulze PC, Bauersachs J, Wellge B, Schoebel C, Tajsic M, Dreger H, Anker SD, Stangl K. Efficacy of telemedical interventional management in patients with heart failure (TIM-HF2): a randomised, controlled, parallel-group, unmasked trial. Lancet 2018;392:1047–1057. [DOI] [PubMed] [Google Scholar]
- 2. Galinier M, Roubille F, Berdague P, Brierre G, Cantie P, Dary P, Ferradou JM, Fondard O, Labarre JP, Mansourati J, Picard F, Ricci JE, Salvat M, Tartiere L, Ruidavets JB, Bongard V, Delval C, Lancman G, Pasche H, Ramirez-Gil JF, Pathak A, Investigators O. Telemonitoring versus standard care in heart failure: a randomised multicentre trial. Eur J Heart Fail 2020;22:985–994. [DOI] [PubMed] [Google Scholar]
- 3. Scholte NTB, Gurgoze MT, Aydin D, Theuns D, Manintveld OC, Ronner E, Boersma E, de Boer RA, van der Boon RMA, Brugts JJ. Telemonitoring for heart failure: a meta-analysis. Eur Heart J 2023;44:2911–2926. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Abraham WT, Adamson PB, Bourge RC, Aaron MF, Costanzo MR, Stevenson LW, Strickland W, Neelagaru S, Raval N, Krueger S, Weiner S, Shavelle D, Jeffries B, Yadav JS, Group CTS. Wireless pulmonary artery haemodynamic monitoring in chronic heart failure: a randomised controlled trial. Lancet 2011;377:658–666. [DOI] [PubMed] [Google Scholar]
- 5. Lindenfeld J, Zile MR, Desai AS, Bhatt K, Ducharme A, Horstmanshof D, Krim SR, Maisel A, Mehra MR, Paul S, Sears SF, Sauer AJ, Smart F, Zughaib M, Castaneda P, Kelly J, Johnson N, Sood P, Ginn G, Henderson J, Adamson PB, Costanzo MR. Haemodynamic-guided management of heart failure (GUIDE-HF): a randomised controlled trial. Lancet 2021;398:991–1001. [DOI] [PubMed] [Google Scholar]
- 6. Brugts JJ, Radhoe SP, Clephas PRD, Aydin D, van Gent MWF, Szymanski MK, Rienstra M, van den Heuvel MH, da Fonseca CA, Linssen GCM, Borleffs CJW, Boersma E, Asselbergs FW, Mosterd A, Brunner-La Rocca HP, de Boer RA; MONITOR-HF investigators . Remote haemodynamic monitoring of pulmonary artery pressures in patients with chronic heart failure (MONITOR-HF): a randomised clinical trial. Lancet 2023;401:2113–2123. [DOI] [PubMed] [Google Scholar]
- 7. Angermann CE, Assmus B, Anker SD, Asselbergs FW, Brachmann J, Brett ME, Brugts JJ, Ertl G, Ginn G, Hilker L, Koehler F, Rosenkranz S, Zhou Q, Adamson PB, Bohm M; MEMS-HF Investigators . Pulmonary artery pressure-guided therapy in ambulatory patients with symptomatic heart failure: the CardioMEMS European monitoring study for heart failure (MEMS-HF). Eur J Heart Fail 2020;22:1891–1901. [DOI] [PubMed] [Google Scholar]
- 8. Shavelle DM, Desai AS, Abraham WT, Bourge RC, Raval N, Rathman LD, Heywood JT, Jermyn RA, Pelzel J, Jonsson OT, Costanzo MR, Henderson JD, Brett ME, Adamson PB, Stevenson LW; CardioMEMS Post-Approval Study Investigators . Lower rates of heart failure and all-cause hospitalizations during pulmonary artery pressure-guided therapy for ambulatory heart failure: one-year outcomes from the CardioMEMS post-approval study. Circ Heart Fail 2020;13:e006863. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Kapelios CJ, Liori S, Bonios M, Abraham WT, Filippatos G. Effect of pulmonary artery pressure-guided management on outcomes of patients with heart failure outside clinical trials: a systematic review and meta-analysis of real-world evidence with the CardioMEMS heart failure system. Eur J Heart Fail 2025;27:1857–1865. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Cowie MR, Flett A, Cowburn P, Foley P, Chandrasekaran B, Loke I, Critoph C, Gardner RS, Guha K, Betts TR, Carr-White G, Zaidi A, Lim HS, Hayward C, Patwala A, Rogers D, Pettit S, Gazzola C, Henderson J, Adamson PB. Real-world evidence in a national health service: results of the UK CardioMEMS HF system post-market study. ESC Heart Fail 2022;9:48–56. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. de Groote P, Thuny F, Blanchart K, Gueffet JP, Habib G, Salvat M, Leclercq C, Mouquet F, Roncalli J, Sebbag L, Cassagneau R, Peyrol M, Sabatier R, Gazzola C, Henderson J, Adamson PB, Roubille F. Remote haemodynamic-guided heart failure management in France: results from the CardioMEMS HF system post-market study (COAST) French cohort. Arch Cardiovasc Dis 2024;117:624–632. [DOI] [PubMed] [Google Scholar]
- 12. McKenzie SPS, O’Donnell D, Gazzola C, Adamson PB. Heart Failure Hospitalisation Reduction With Remote Pulmonary Artery Pressure Monitoring in Australia: Results of the CardioMEMS HF System OUS Post-Market Study. Heat Lung and Circulation 2024;33:S266. 72nd Annual Scientific Meeting of the Cardiac Society of Australia and New Zealand (CSANZ). Perth, Australia; 2024. [Google Scholar]
- 13. Cowie MR, de Groote P, McKenzie S, Brett ME, Adamson PB; CardioMEMS Post-Market Study Investigators . Rationale and design of the CardioMEMS post-market multinational clinical study: COAST. ESC Heart Fail 2020;7:865–872. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. McDonagh TA, Metra M, Adamo M, Gardner RS, Baumbach A, Bohm M, Burri H, Butler J, Celutkiene J, Chioncel O, Cleland JGF, Crespo-Leiro MG, Farmakis D, Gilard M, Heymans S, Hoes AW, Jaarsma T, Jankowska EA, Lainscak M, Lam CSP, Lyon AR, McMurray JJV, Mebazaa A, Mindham R, Muneretto C, Francesco Piepoli M, Price S, Rosano GMC, Ruschitzka F, Skibelund AK; ESC Scientific Document Group . 2023 focused update of the 2021 ESC guidelines for the diagnosis and treatment of acute and chronic heart failure. Eur Heart J 2023;44:3627–3639. [DOI] [PubMed] [Google Scholar]
- 15. Adamson PB, Abraham WT, Aaron M, Aranda JM Jr, Bourge RC, Smith A, Stevenson LW, Bauman JG, Yadav JS. CHAMPION trial rationale and design: the long-term safety and clinical efficacy of a wireless pulmonary artery pressure monitoring system. J Card Fail 2011;17:3–10. [DOI] [PubMed] [Google Scholar]
- 16. Abraham WT, Stevenson LW, Bourge RC, Lindenfeld JA, Bauman JG, Adamson PB; CHAMPION Trial Study Group . Sustained efficacy of pulmonary artery pressure to guide adjustment of chronic heart failure therapy: complete follow-up results from the CHAMPION randomised trial. Lancet 2016;387:453–461. [DOI] [PubMed] [Google Scholar]
- 17. Zile MR, Abraham WT, Stevenson LW, Costanzo MR, Angermann CE, Mehra MR, Desai AS, Ducharme A, Johnson N, Henderson J, Lindenfeld J. Relationship between remote, ambulatory pulmonary artery pressures, and all-cause mortality in patients with chronic heart failure. Circ Heart Fail 2025;18:e012754. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Setoguchi S, Stevenson LW, Schneeweiss S. Repeated hospitalizations predict mortality in the community population with heart failure. Am Heart J 2007;154:260–266. [DOI] [PubMed] [Google Scholar]
- 19. Lindenfeld J, Abraham WT, Maisel A, Zile M, Smart F, Costanzo MR, Mehra MR, Ducharme A, Sears SF, Desai AS, Paul S, Sood P, Johnson N, Ginn G, Adamson PB. Hemodynamic-GUIDEd management of heart failure (GUIDE-HF). Am Heart J 2019;214:18–27. [DOI] [PubMed] [Google Scholar]
- 20. Ong MK, Romano PS, Edgington S, Aronow HU, Auerbach AD, Black JT, De Marco T, Escarce JJ, Evangelista LS, Hanna B, Ganiats TG, Greenberg BH, Greenfield S, Kaplan SH, Kimchi A, Liu H, Lombardo D, Mangione CM, Sadeghi B, Sadeghi B, Sarrafzadeh M, Tong K, Fonarow GC; Better Effectiveness After Transition–Heart Failure (BEAT-HF) Research Group . Effectiveness of remote patient monitoring after discharge of hospitalized patients with heart failure: the better effectiveness after transition—heart failure (BEAT-HF) randomized clinical trial. JAMA Intern Med 2016;176:310–318. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data from this study will not be made publicly available. However, the authors and the sponsor are committed to working with other investigators on projects of mutual interest that might include data sharing. Any such inquiries should be made to the sponsor, Abbott.