The impact of atrial flow regulator implantation on hemodynamic parameters in patients with heart failure edema; heart failure; interatrial shunt; quality of life

Background Left atrial decompression has emerged a new option to treat patients with heart failure and dyspnea at rest or during exercise. Here we report the impact of atrial flow regulator (AFR) implantation on hemodynamic parameters in patients at our center with heart failure and with reduced (HFrEF) or with preserved left ventricular ejection fraction (HFpEF). Material and methods The PRELIEVE trial is designed to assess the safety and efficacy of the AFR in patients with HFrEF or HFpEF. Patients with left ventricular end-diastolic pressure ≥15 mmHg at rest or ≥25 mmHg during exercise and with an ejection fraction ≥15 % were enrolled. Echocardiographic data, 6-min walking distance, Kansas City Cardiomyopathy Questionnaire, and brain natriuretic peptide levels were assessed pre- and post-AFR implantation and at 3 mos. Invasive hemodynamic assessments were also performed pre- and post-AFR implantation and at 3 mos. Results 27 12 this study. A significant decrease was observed in pulmonary arterial wedge pressure regardless of EF (p=0.007 HFrEF and p=0.03 HFpEF). No significant difference of mean pulmonary arterial pressure, right arterial pressure and cardiac output (CO) existed at 3 months compared with pre-implantation baseline values. AFR implantation led to decrease in left ventricle filling pressure without the deleterious impact on CO and right heart function regardless of ejection fraction.


Introduction
Increased left atrial pressure (LAP) secondary to elevated left ventricular end-diastolic pressure (LVEDP) leads to pulmonary congestion that is responsible for dyspnea at rest or during exercise in patients with heart failure and with reduced (HFrEF) or preserved left ventricular ejection fraction (HFpEF) [1]. Decreased left ventricular ejection fraction (LVEF) in HFrEF and impaired myocardial relaxation in HFpEF cause elevated LVEDP and LAP [2,3]. Diuretics and vasodilators are used to reduce LVEDP and LAP [4], but as heart failure progresses, the effects of medical therapy abate. The prevalence of HF is 1-2 % in the general population and reaches >10 % in persons aged 70 and older [5]. Given the high prevalence of HF in the older population, drug-resistant scenarios are inevitable.
Recently, several novel device implants to treat heart failure symptoms using left atrial decompression have been tested successfully. They include the Ventura device (V-Wave Ltd., Or Akiva, Israel), the IASD (Corvia Inc., Tewksbury, MA, USA), the AFR (Occlutech, Schaffhausen, Switzerland), and the Transcatheter Atrial Shunt System (a left atrium-to-coronary sinus shunt device by Edwards Lifesciences, Irvine, California). The latter device is placed by atriotomy, whereas the other three devices are deployed in the interatrial septum. The IASD has been investigated in HFpEF patients [6,7], and the Ventura device was tested in patients with HFrEF [8]. Both devices were proven to be safe and showed initial beneficial hemodynamic and clinical outcomes. Last year, 3-mos results of the PRELIEVE trial in both HFrEF and HFpEF patients were published in Eurointervention [9]. All three interatrial shunting devices are approved (CE-marked) for use in patients with HF.
These devices create passive left atrial decompression. In a computer simulation study, an 8-mm interatrial shunt, which was identical to the bAFR device shunt, shifted the left atrial pressure-volume loop leftward and downward [10]. This caused a minimal decrease in left ventricle output while mildly increasing right ventricle output. These results were coupled with a marked reduction in pulmonary artery wedge pressure (PAWP) (~3 mmHg at rest and ~11 mmHg at peak exercise). Right atrial and pulmonary artery pressures did not significantly increase. The effects of interatrial shunt on pulmonary hemodynamics have been investigated in a preclinical study [11]. That study ОРИГИНАЛЬНЫЕ СТАТЬИ § demonstrated the favorable effects of creating an interatrial shunt on pulmonary hemodynamics in rats with HFpEF.
Here we report the hemodynamic changes following AFR implantation at 3 mos in patients with HFrEF and HFpEF.

Material and methods
PRELIEVE is a non-randomized, prospective, multicen ter, open label pilot study of the AFR. PRELIEVE has been approved by local and national ethics committees, and 19 clinical centers in Turkey, Belgium and Germany are part of the study. Prior to patient recruitement study protocol was approved by sponsor and local ethic committee in Bezmialem Vakif University (Date:29 / 03 / 2017, No: 71306642-050.01.04).
Quality of life (QoL, assessed by the Kansas City Cardio myopathy Questionnaire (KCQQ)), New York Heart Association (NYHA) class, 6-min walking distance (6MWD), and transthoracic echocardiography (TTE) parameters are assessed during follow-up according to the protocol ( Figure 1).
HFrEF and HFpEF patients were enrolled in the study. Patients with left ventricular ejection fraction (LVEF) equal to or greater than 15 % and less than 40 %, and with documented elevated left ventricular filling pressure (PAWP ≥15 mmHg at rest or ≥25 mmHg during exercise) were included in the HFrEF group. Patients with LVEF equal to or greater than 40 % and with documented elevated left ventricular filling pressure were included in the HFpEF group. Inclusion and exclusion criteria are summarized in Table 1.
The study follow-up period was planned to be completed in 12 m and consisted of eight clinical visits, i.e., screening, implantation, and six follow-up visits. 6MWD performance and QoL (KCCQ) were assessed after 1, 3, 6, and 12 mos. Transesophageal echocardiography (TEE) and right heart catheterization were performed at the 3-mon follow up visit. The study is ongoing. We reported here 3-month data.
The primary safety endpoint was the presence of serious adverse device effects (SADEs) at 3 mos, SADEs were defined as device dislocation or embolization, device-related injury of mitral or tricuspid valve, device-related intractable  Figure 1. Brief review of the study protocol ОРИГИНАЛЬНЫЕ СТАТЬИ § arrhythmia, or any circumstance that required device removal. The secondary safety endpoints were the rate of all serious device events (SAE) and the presence of SADEs during 12 mos after implantation.

AFR device description
The AFR device is a self-expanding, double-disc, circular device made of nitinol wire mesh with «superelastic properties» (Figure 2). A waist with a central shunt connects the discs. A welded ball structure located on the proximal disc surface serves as an adapter for the pusher cable during implantation. The AFR is available with different waist-shunt diameters, waist heights and disc diameters to provide shunts of different diameters and to accommodate varying atrial septal anatomy. Depending on the size of the AFR, the manufacturer recommends using the Occlutech Delivery System (ODS) with sizes ranging from 8F to 14F.

Procedural details
The procedure began with local anesthesia and sedation. Right and left cardiac catheterizations were performed. PAWP, pulmonary artery and right heart chamber pressures, central venous, and aortic and left ventricular pressures were recorded. Blood samples for gas analysis were obtained. Cardiac output (CO), pulmonary and systemic vascular resis tance were calculated. Hemodynamic findings were used to confirm subjects' study eligibility. Device sizing was performed according to hemodynamic and clinical data as well as atrial septal thickness ( Figure 3). Transseptal puncture was performed with TEE guidance and under general anesthesia. After the transseptal puncture, unfractionated heparin was given intravenously to achieve an activated clotting time (ACT) >250 sec, and a stiff wire was placed in the left upper pulmonary vein. The puncture site in the septum was predilated by balloon inflation to a diameter 2 mm larger than the intended shunt, i.e., the AFR device diameter. After the AFR device was loaded onto the pusher, the AFR was advanced through the delivery sheath into the left atrium. Following appropriate positioning of the left atrial disc, the right atrial disc was deployed. Before releasing the device from its pusher, the Minnesota maneuver [10] was used to test device stability, and TEE was performed to confirm correct device position and patency. Following device placement, right and left heart catheterizations were repeated.
During the first 3 mos after implantation, patients received oral clopidogrel 75 mg and acetylsalicylic acid 100 mg OD. Thereafter, acetylsalicylic acid 100 mg OD was recom mended for patients not on anticoagulation. If a patient needed anticoagulation for any reason, daily clopidogrel 75 mg was added. Standard endocarditis prophylaxis was administered during the procedure and for a minimum of 6 mos following implantation.

Study Populations and Baseline Characteristics
A flowchart showing patient enrollment is shown in Figure 4. A total of 48 patients were screened and 39 patients were enrolled in the study at our center. Twenty-seven (69.2 %) patients were in the HFrEF group and 12 (30.8 %) patients were in the HFpEF group. Eighteen (66.7 %) HFrEF patients and 7 (58.3 %) HFpEF patients were male. The median age of the HFrEF population was 68.3±7.1 yrs and 70.6±7.2 yrs in the HFpEF population. Additional study population information and characteristics are listed in Table 2.

Procedural results
Procedural data are summarized in Table 3. Overall catheterization times were similar in both groups (HFrEF: 86 (70-103) min and HFpEF: 89 (71-105) min. The AFR device with an inner fenestration diameter of 8 mm (HFrEF: 81 % and HFpEF: 92 %) and a height of 5 mm (HFrEF: 100 % and HFpEF: 83 %) was used in most of the study population ( Table 4). The final opening size was assessed by intraprocedural TEE in terms of device recoiling. We did not observe any recoiling of the devices. Left-to-right shunting through the device was documented immediately after deployment in all patients, and device patency was 100 % at the 3-mo follow up, as assessed by TEE or TTE.

Safety events at the 3-mo follow up examination
Safety data are summarized in Table 5. Adverse devi ce effects were seen in two HFrEF patients (injection site reaction and paresthesia) and all had resolved. No adverse device events and deaths were seen in the HFpEF patients. One HFrEF patient died within 1 wk after device implantation due to pneumonia and septicemia. SAEs were observed in 12 (44.4 %) HFrEF and 4 (33.3 %) HFpEF patients. Worsening heart failure was observed in 3 (11.1 %) patients, all in the HFrEF group. Iliac vein thrombosis occurred in one HFrEF patient. Acute arterial deoxygenation was not reported, and all patients were discharged after implantation. Procedure related serious adverse effects (SAEs) were observed in 2 patients (5.1 %), all in the HFrEF group. No strokes / transient ischemic attacks (TIA), myocardial infarctions, or complications requiring device removal were seen in either group. Device patency was maintained in all patients at the 3-mo follow-up examination.   Table 6. Compared with pre-implantation, PAWP and LVEDP were significantly reduced at 3 mos (p=0.007 and p=0.01, respectively). There was no difference between pre-implantation and 3-mos cardiac output (p=0.45). No significant difference existed between post-implantation Qp / Qs and Qp / Qs calculated at 3 mos (1.23±0.29 vs. 1.32±0.39, respectively). Mean pulmonary pressure decreased mildly but not significantly (p>0.99). There was no increased in RA pressure at 3 mos compared with baseline (median 10.00 vs. 9.33, respectively, p=0.70) ( Figure 5).

HFpEF population
PAWP was significantly improved at 3 mos, from median 18 to 10 mmHg (p=0.03). There was no change in cardiac output at 3 mos compared with baseline, median 4.67 vs 4.99 l / min, respectively (p=0.67). Median mean pulmonary arterial pressure decreased from 28 to 23 mmHg, but it was not statistically significant (p=0.39). No significant difference was observed between pre-implantation and 3-month mean right arterial pressure (p=0.80) ( Figure 6).

Discussion
We assessed the impact of AFR device implantation on hemodynamic parameters in both HFrEF and HFpEF patients. Device implantation were performed without any complications in either group. No events requiring device removal occurred after implantation. At 3 mos, device patency was present in all assessed cases. A left-to right shunt was seen immediately after device placement and at 3 mos. No acute or chronic arterial deoxygenation was observed after AFR device deployment nor during the 3-mo follow up period.
Two different situations show that left atrial decompres sion may be effective in heart failure. The first is Lutembacher's syndrome, which is defined as the concomitant rheumatic mitral stenosis and atrial septal defect [11]. This combination is associated with a relatively small increase in left atrial pressure during rest and effort, and a delay in the onset of dyspnea compared to isolated mitral  ОРИГИНАЛЬНЫЕ СТАТЬИ § stenosis. The same volume of blood may lead to lower pressure in the right atrium compared to the left atrium due to greater right atrial distensibility [12]. This theory was supported by Little's anatomical study in isolated canine hearts [13]. Similar results were reported in post-MitraClip (Abbott, Chicago, USA) cases performed by the transseptal route. Recent publications demonstrated that hemodynamic and clinical results may be effective in selected MitraClip cases with persistent iatrogenic septal defects. In a study by Ikenaga et al., although left atrial pressure was higher in post-MitraClip patients with persistent iatrogenic atrial septal defect (iASD) compared to those without iASD, the lack of difference between NYHA FC and brain natriuretic peptide (BNP) values indicated that left atrial decompression may be beneficial [14]. In the MITHRAS study, percutaneous closure of the iASD following transcatheter mitral valve repair was compared with conservative treatment. Patients with Qp / Qs >1.3 were included in the study. At 5-mo follow-up, there were no differences in terms of 6MWDT, NHYA class and peripheral edema between groups [15]. The second situation is the effect of atrial septal defect (ASD) closure on left ventricular and atrial hemodynamics in adults [16], where it was shown that ASD closure triggers left heart failure in patients with high left ventricular enddiastolic pressures. In light of the data obtained from these two contrasting situations but with similar results, it can be postulated that interatrial shunt devices may be effective in treating isolated left heart failure [17]. The main concerns regarding left atrial decompression with interatrial shunt devices are: 1) successful implantation and stability of the device; 2) long-term shunt patency; 3) the risk of paradoxical embolism; 4) right heart volume overload and increased pulmonary pressure at the long-term follow-up; 5) decrease in left ventricle output.
The shunt diameter of the Ventura V-Wave device is 5 mm. The first generation version of this device contained a bioprosthetic valve, and stenosis or occlusion of the shunt was seen in 50 % of subjects at 1-yr follow up [18]. After examination of an explanted heart, it was found that the occlusion / stenosis was associated with the bioprosthetic valve rather than due to thrombus. V-Wave designed a second generation device without a bioprosthetic valve. Shunt patency was maintained with this device for up to 1 yr [19].
The InterAtrial Shunt Device (IASD) has a shunt diameter of 8 mm. This device has been studied in HFpEF patients [20]. Shah et al. could not find any evidence of shunt stenosis / shunt occlusion at 12 mos [21].
To date, no cases of paradoxical embolism have been reported in post-MitraClip patients despite having relatively larger iASD [22]. In studies of two different interatrial shunt devices, Del Trigo et al. [8] and Feldman et al. [23] reported no events associated with paradoxical embolism or device thrombus. An increase in pulmonary artery pressure as a result of volume overload created by the left-right shunt is another concern. However, as known from congenital heart disease patients, small defects, e.g., an ASD of 10 mm, are not associated with deleterious hemodynamic effects during a long-term follow-up period [24]. These data were confirmed in a simulation study by Kaye et al. [25]. They observed that the interatrial shunt with a diameter of 8-9 mm provided a significant decrease in PAWP without serious volume loading in the right ventricle or a serious decrease in left ventricular output [25]. The lumen diameters of the AFR device are 8 and 10 mm. In the present study, 33 (85 %) patients were implanted with an AFR device with a diameter of 8 mm. An AFR with a diameter of 10 mm was implanted in patients whose resting left ventricular end-diastolic pressure was less than 15 mmHg but with exercise LVEDP equal to or greater than 25 mmHg. The Qp / Qs ratio was 1.32±0.39 in the HFrEF group and 1.12±0.32 in the HFpEF group at the end of 3 mos. There was no significant decrease in resting cardiac output in either group at the end of 3 mos (HFrEF

Study limitations
This study has several limitations. First, this was an open label, non-randomized trial. The results are limited by being conducted at one center and with a small sample size. Second, the follow-up period was restricted to 3 mos after the procedure. The PRELIEVE study is ongoing, and further results will be available in the future. Third, the study was a single-arm trial, and we could not compare these results with placebo therapy.

Conclusion
Regardless of left ventricular ejection fraction, AFR implantation decreased left ventricle filling pressure without a deleterious impact on cardiac output or on right heart function.

No conflict of interest is reported.
The article was received on 14 / 03 / 2021