Left Impella^®-device as bridge from cardiogenic shock with acute, severe mitral regurgitation to MitraClip^®-procedure: a new option for critically ill patients

Abstract

Introduction

Acute, severe mitral regurgitation (MR) occurs as a result of an abrupt failure of the mitral valve apparatus, such as rupture of a papillary muscle or chordae, resulting in flail mitral leaflets.¹ Though rare, it is a medical emergency due to abrupt deterioration to cardiogenic shock (CS). Patients develop severe decompensated heart failure due to an acute reduction in cardiac output, caused by a major proportion of the blood pumped by the left ventricle retrogradely into the left atrium instead of flowing forward through the aortic valve (Figure 1A, left panel; forward failure). This leads to an acute cardiac pressure and volume overload and pulmonary oedema and hypoxia ensues.

Figure 1

Impella^®-CP as bridge from acute severe mitral regurgitation to MitraClip^®-procedure. (A) (left): Severe mitral regurgitation causing left atrial congestion and low cardiac output (forward failure); (middle): left ventricular unloading by Impella^®-CP reduces atrial/pulmonary congestion and increases total cardiac output; (right): Impella^®-CP-supported MitraClip^®. (B) Pulmonary congestion (Patient 6) with acute, severe mitral regurgitation before (left), and after (right) Impella^® supported MitraClip^®-procedure. (C) (left): Severe (ischaemic) mitral regurgitation (Patient 6); (middle and right): reduction of mitral regurgitation after MitraClip^®-procedure (^#Impella^®-CP, *2 XTR MitraClips^®).

Open in new tabDownload slide

Acute severe MR is traditionally treated surgically, but significant intra- and postoperative risks exist, in particular where CS and pulmonary oedema are present. Conventional management of CS includes using intravenous fluids and vasoactive agents; however, these may result in greater MR and worsening pulmonary oedema with only minimal improvement in forward flow, and/or exacerbation of hypotension.^2,3 Even if stabilization using these methods is achieved, the risks from open surgical intervention remain, even if minimally invasive techniques are utilized.

This exploratory study analyses whether the combined and staged approach of mechanical circulatory support (MCS) and percutaneous mitral valve intervention offer promise in the management of this complex and rare condition. MCS decreases preload, cardiac workload and—depending on the device used—also afterload whilst simultaneously increasing coronary perfusion and cardiac output. In non-septic patients with preserved right ventricular function presenting with acute severe MR, a percutaneous left ventricular device (pVAD) such as Impella^® could offer a viable bridge to further therapy.^3,4 A combined approach with percutaneous mitral interventions such as MitraClip^® (approximation of the anterior and posterior mitral leaflets via trans septal approach, restoring coaptation of the leaflets)⁵ could be a safer treatment strategy in this acute CS group of patients with high surgical risk.

Aims

We explored whether in patients with CS related to acute MR and were considered too high risk for immediate surgical intervention, the combined approach of stabilizing CS by unloading the left ventricle with an Impella^® device (Abiomed, Danvers, USA) and MitraClip^®-guided (Abbott Vascular, USA) MR-reduction was a feasible and possibly safer alternative to mitral valve surgery.

Methods

We analysed all patients at two tertiary cardiac intensive care unit (ICU) centres between August 2017 and January 2020 who had presented with acute, moderate-to-severe or severe MR, been declined for immediate surgical intervention (EURO-II score ≥8% + pulmonary oedema necessitation mechanical ventilation and/or haemodynamic instability), and had been selected for percutaneous MCS (Impella^®)-assisted support and subsequently referred for percutaneous mitral valve repair (MitraClip^® procedure). At both sites, the patients had been assessed by a heart team comprising intensivists, heart failure specialists, interventional cardiologists experienced in Impella^® and MitraClip^® insertion and a cardiac surgeon. Vital parameters were assessed for all patients until discharge and freedom from heart failure hospitalization or survival was followed up until 6 months post-discharge. Results are shown as mean ± standard deviation. This study was conducted in accordance with Declaration of Helsinki tenets. This study was approved by the ethics committee of the university hospitals Leuven (S62048).

Results

Between August 2017 and January 2020, six patients matching our inclusion criteria were identified. Baseline patient characteristics are listed in Table 1. The mean age was 66.8 ± 4.9 years. All but one patient presented with de novo CS. All patients underwent a coronary angiogram with percutaneous coronary intervention (PCI) carried out when indicated. All patients presented in INTERMACS-1 (interagency registry for mechanically assisted circulatory support scale) CS with pulmonary oedema due to high filling pressures (PCPW > 20 mmHg; all patients had a pulmonary artery catheter inserted) necessitating mechanical ventilation. None of these patients had severe right ventricular failure (exclusion for isolated left-sided mechanical support) or septic shock in addition to their acute MR. One patient was in multiple organ failure (late referral; lactate 9.3 mmol/L) at presentation. The mean overall cardiac operative risk assessment (EURO-II) score indicated a mean perioperative mortality risk of 39 ± 18.6% (SD 18.6%).

The MR severity and echocardiography findings at initial presentation, plus supportive measurements are outlined in Table 2. Five (84%) of the patients presented with severe (Grade 4) MR, two due to ischaemia (33%), and three (50%) due to non-ischaemic/degenerative chordal rupture. Cardiac output was severely reduced (mean left ventricular outflow tract velocity time index 9.8 ± 1.8 cm; forward failure) with preserved right ventricular function (mean tricuspid annular plane systolic excursion 2.7 ± 0.3 cm). All patients received inotropic support and were supported by an Impella^®-CP axial-flow pump (mean flow 2.9 ± 1.8 L per minute; mean support 9.7 ± 6.0 days), inserted as quickly as possible after admission. Fifty percent of our critically ill population needed continuous kidney replacing therapy during the ICU-stay. A MitraClip^® procedure was successfully performed in all cases [MR reduction from moderate–severe or severe to mild; 1.8 ± 0.8 clips per patient; 67% of the inserted clips was of the last generation (XTR or NTR); 7.2 ± 5.0 days after Impella^®-insertion]. There were no cases where the presence of the Impella^®-CP device complicated the percutaneous mitral valve repair technique nor did the presence of the pump require a different approach for clip-insertion. The time of pump-removal after mitral valve repair was 2.6 ± 2.1 days. The combination of medical and mechanical support, plus MitraClip^®-procedure resulted in a pulmonary capillary wedge pressure (PCWP) of less than 15 mmHg (normal range) in every patient.

Outcomes are outlined in Table 3. There were no device complications seen and no complications related to the MitraClip^® procedure. Plasma free haemoglobin levels, reflecting the degree of haemolysis, were low due to precarious monitoring of the pump position. In three (50%) of the patients, primarily those with ischaemic MR, bleeding necessitating transfusion (BARC 3) occurred presumably as they required both heparin for pMCS anticoagulation and antiplatelet drugs post-coronary intervention.

Five (86%) of the patients survived to discharge. The non-survivor died of multi-organ failure that was present at the time of referral (MCS-support and correction of the MR was ultimately unable to correct the metabolic and organ dysfunction that had developed prior to this). Overall, MR was reduced from severe to mild. Of the five patients that survived to discharge, all survived 6 months after discharge and only two patients were readmitted; both due to mild decompensated heart failure secondary to atrial fibrillation (AF), successfully treated with diuretic therapy and rhythm control.

Discussion

Shock due to MR carries a poor prognosis and has been shown to be an independent predictor of 1-year mortality in ST-elevation myocardial infarction (STEMI) patients who present with shock on admission.⁶ This exploratory study evaluated the feasibility of a combined percutaneous mechanical support plus mitral valve repair in acute CS patients presenting with acute, severe MR who were considered too critically ill for urgent surgical valve repair. There have previously only been individual case reports of the use of Impella^® in unloading strategies in patients with acute MR, mostly as a bridge to later surgery or (rarely) to stable medical therapy.⁴

In patients with acute severe MR, left ventricular function is not the driver for limitation of cardiac output but high backward flow and forward failure (Figure 1) are the causing factors of shock. In patients such as these, the acute volume overload of the non-distended and non-compliant left ventricle and atrium both result in acute pulmonary oedema and forward failure (Figure 1), and is associated with an extremely high mortality between 35% and 50%.¹ In these patients, quick restoration of forward flow (to preserve end organ perfusion) and prevention of retrograde flow (to prevent pulmonary congestion) is crucial. Surgery has previously been considered the only viable salvage therapy in this acutely ill population, although this is associated with high perioperative risks.² Technological improvements in percutaneous MCS and transcatheter valve therapies have the potential to explore new perspectives for this critically unwell patient group.

We used the updated version of the European System for Cardiac Operative Risk Evaluation (EURO-II) score to predict the mortality after cardiac surgery in our patient population.⁷ Scores were high reflecting the urgent presentation with emergent need for surgery, recent myocardial ischaemia and a critical preoperative state.⁸ The EURO-II score is a well-validated risk assessment (receiver operating characteristic curve of 0.8095) for in-hospital mortality after cardiac surgery. A EURO-II score of ≥8% in-hospital is considered as high-risk.^7,9

CS is a frequently lethal syndrome, and disappointing results from standard medical therapy to support the circulation has led to increasing interest in MCS as a potential management option.¹⁰ In line with this, there has been a significant increase in the implementation of short-term pMCS; its ability to reduce cardiac workload and myocardial oxygen demand while maintaining systemic and coronary perfusion is a great asset in stabilizing CS patients. Despite extensive registry data (in particular for Extra-Corporeal Membrane Oxygenation and axial-flow pumps) high-quality scientific data are limited to only a couple of randomized trials¹¹; The IMPella vs. IABP Reduces mortality in STEMI patients treated with primary PCI in Severe cardiogenic SHOCK (IMPRESS-in-Severe-SHOCK) trial, included patients with severe CS requiring mechanical ventilation and randomized them to Impella^®-CP vs. IABP. Neither this trial nor a meta-analysis of 148 patients randomized to Impella^® or TandemHeart versus IABP showed any difference in all-cause mortality after 30 days.^11,12 Recently, a propensity-matched registry-based retrospective cohort study of 1680 patients with acute myocardial infarction complicated by CS undergoing PCI, reported a significantly higher risk of in-hospital mortality associated with use of an axial-flow pump (45.0%) vs. IABP (34.1%, P < 0.001), likely related to a higher risk of in-hospital major bleeding (31% vs. 16%, P < 0.001). The validity of these findings however is unclear, as overall in-hospital mortality of IABP appeared uncharacteristically low, being the lowest reported in the literature, including randomized controlled trials.¹³ The DanGer Shock (Danish-German cardiogenic shock) Trial (NCT01633502) is currently randomizing 360 patients with acute myocardial infarction and CS to Impella^®-CP or conventional guideline-driven treatment.¹⁴

Transcatheter mitral valve edge-to-edge repair is another novel, minimal invasive technology. MitraClip^® has shown promising results in patients with moderate-to-severe (3+) secondary MR and left ventricular ejection fraction ≥20% with left ventricular end-systolic diameter <70 mm,^15–17 although the implementation of percutaneous repair techniques in the setting of acute valve failure due to chordal or papillary rupture remains unexplored. From a conceptual perspective, however, a favourable impact on hemodynamic is to be expected whenever systolic dysfunction is temporary. As shown in the EVEREST-studies, MitraClip^®-repair has been shown to be non-inferior to surgical repair in terms of survival, while conferring decreased risk from the procedure.¹⁸

For the first time, we report a group of six patients at high-risk for cardiac surgery, presenting with CS and acute, severe MR all successfully haemodynamically stabilized by Impella^®-CP (mean flow of 2.9 L per minute for an average of 9.7 days) prior to MitraClip^® procedure (Figure 1). MitraClip^® procedure was successfully carried out in all six patients, with MR grade reduced [from moderate (3+) or severe] to mild (1+) or absent in all cases. Post-procedure, PCWP in all patients was below 15 mmHg further allowing rapid ventilator and sedation weaning.

A sub study of 98 patients of the SHOCK-trial (Should we use emergently revascularize Occluded Coronaries in cardiogenic shock) registry, performed in patients developing severe MR complicating acute myocardial infarction and leading to CS, reported an in-hospital death of 40% in patients undergoing mitral valve surgery (comparable with the EURO-II score in our group: 39 ± 18.6%) and 71% in the non-cardiac surgery group. Overall, 75% of these CS patients died because they could not be stabilized prior to surgery or due to multi-organ failure secondary to CS.¹⁹ During the defined study time, a total of 11 CS patients underwent emergency mitral valve repair at our institutions (mean EURO-II score 35%; in hospital mortality 54%). Survival to discharge in our study was high at 83% and early referral for MCS support and valve repair were key in achieving this. None of our patients had to be upgraded to a permanent left ventricular assist device. Readmission rates were low with only two patients required readmission within 6 months, both for mild decompensated heart failure secondary to AF.

Conclusion

A combined strategy of Impella^® and MitraClip^® appears to be a novel, feasible alternative for weaning patients presenting with CS and acute, severe MR unable to proceed to a surgical corrective procedure at presentation due to instability caused by the left ventricular forward flow failure. In these cases, early initiation of pVAD-support combined with optimal ICU-management is essential to reduce the risk of development of irreversible end organ damage and dysfunction. Upfront stabilization with Impella^® facilitates bridging to MitraClip^®-procedure to achieve successful weaning from sedation and ventilator support.

Acknowledgements

We sincerely thank the nursing staff of the adult intensive care units and the coronary angiography labs at both institutions.

Funding

CV ALSO received a research grant from the "Klinische Onderzoeks- en opleidingsraad Leuven" (KOOR, UZ Leuven).

Conflict of interest: C.V. received a research grant from the European Society of Cardiology working group on Thrombosis. B.M. and T.A. received a research grant from Abiomed.

References