Use of the CardioHELP Device for Temporary Hemodynamic Support During High-Risk Percutaneous Coronary Intervention

Background. Temporary extracorporeal membrane oxygenation (ECMO) support for high-risk percutaneous coronary intervention (PCI) has been described in select patients, and data are limited on the CardioHELP device (Maquet). The objective of this study was to assess clinical outcomes in patients undergoing elective, high-risk PCI with CardioHELP support. Methods. Fifteen consecutive patients receiving the CardioHELP device for elective, high-risk PCI treated at 2 medical centers were included. Patients with cardiogenic shock, cardiac arrest, or non-PCI indications for ECMO were excluded. Baseline demographics, angiographic variables, procedure-related variables, and in-hospital events were collected. Results. Mean age was 71 ± 11 years, 73% were male, mean ejection fraction (EF) was 29 ± 13%, 10 patients (67%) had an EF <30%, and mean SYNTAX I score was 32 ± 11. Multivessel coronary artery disease was present in 14 patients (93%) and unprotected left main coronary artery disease was present in 4 patients (27%). PCI was successful in all patients. In-hospital mortality occurred in 3 patients (20%), 7 patients (47%) received a blood transfusion, and there were no major vascular complications. Conclusion. Temporary use of the CardioHELP device for high-risk PCI is associated with acceptable short-term outcome and may be a new option for patients with complex coronary artery disease and left ventricular dysfunction.

J INVASIVE CARDIOL 2021;33(8):E614-E618.

Key words: high-risk PCI, hemodynamic support, extracorporeal membrane oxygenation

The ability to perform successful percutaneous coronary intervention (PCI) in patients with advanced coronary artery disease (CAD) and reduced left ventricular function may require the use of temporary hemodynamic support.^1,2 Although a variety of devices are currently available, randomized clinical trials have not found a clear benefit with any particular device.^3,4 Several small studies have evaluated the use of extracorporeal membrane oxygenation (ECMO) for high-risk PCI.^5-7 The benefits of ECMO include the ability to provide complete hemodynamic support and arterial oxygenation, but may be mitigated by the need for large arterial and venous cannulas with their inherent risk of bleeding and vascular complications. The initial generation of ECMO devices was bulky and difficult to initiate, and required the presence of a trained perfusionist.⁸ More recently, the CardioHELP device (Maquet) was introduced as a small, portable ECMO system that is easy to initiate, does not require a perfusionist, and can provide up to 3.5 L/min of cardiac output.⁹ The objective of this study was to assess clinical outcomes in patients undergoing elective, high-risk PCI with the CardioHELP device.

From October, 2014 to August, 2017, a total of 73 patients received the CardioHELP device at 2 medical centers in Los Angeles, California (Good Samaritan Hospital and Los Angeles County University of Southern California [USC] Medical Center). Patients with cardiogenic shock (n = 35), venous-venous ECMO (n = 12), and those requiring ECMO for non-PCI indications (transcatheter aortic valve implantation; n = 11) were excluded. Only patients undergoing elective, high-risk PCI were included (n = 15). High-risk PCI was defined as the presence of impaired left ventricular function (ejection fraction [EF] <40%) and multivessel coronary artery disease (CAD) or unprotected left main coronary artery disease or intervention on the last remaining patent vessel (native vessel or bypass graft). This retrospective study was approved by waiver of consent by the institutional review board at the USC. Independent research personnel blinded to the objectives of the study obtained the prespecified demographics, as well as procedural, angiographic, and in-hospital clinical events. Variables were entered into a dedicated study database for analysis.

All patients underwent PCI as clinically indicated by experienced interventional cardiologists. The decision to use CardioHELP for hemodynamic support was at the discretion of the treating interventional cardiologist. During the study period, intra-aortic balloon pump (IABP), Impella (Abiomed), and Impella CP (Abiomed) devices were available for hemodynamic support. Cannulation was femoral-femoral in all patients. All patients received moderate sedation for the procedure. None of the patients received antegrade perfusion of the lower extremity, except for 1 patient who required continued ECMO support following the PCI procedure. Heparin was used for anticoagulation during the procedure. All patients except 1 were decannulated in the cardiac catheterization laboratory at the completion of the PCI procedure using the “preclose” technique.¹⁰Duration of ECMO support was defined as the time from ECMO initiation to decannulation. Following the PCI procedure, all patients received aspirin (325 mg orally once daily) and either clopidogrel (300 mg or 600 mg orally followed by 75 mg once daily), prasugrel (60 mg orally followed by 5 or 10 mg once daily), or ticagrelor (180 mg orally followed by 90 mg twice daily). Aspirin was recommended indefinitely.

Study definitions. Race was self reported and characterized as White, Black, Asian, Hispanic, and other. Society of Thoracic Surgery (STS) score was calculated using the online calculator (http://riskcalc.sts.org/stswebriskcalc/calculate). Logistic EuroScore was calculated using the online calculator (www.euroscore.org). EF was measured in all patients by echocardiography prior to the PCI procedure. Cardiogenic shock was defined as systolic blood pressure <90 mm Hg with signs of impaired tissue perfusion (altered mental status, reduced urine output), the need for vasopressor agents to maintain systolic blood pressure >90 mm Hg and cardiac index <2.0 L/min/m², and pulmonary capillary wedge pressures >18 mm Hg for patients with a Swan-Ganz catheter. Chronic kidney disease was defined as creatinine clearance <30 mg/dL. SYNTAX score I was calculated using the online calculator (http://www.syntaxscore.com/calculator/syntaxscore/frameset.htm). In-hospital major adverse cardiovascular and cerebral vascular event (MACCE) was defined as the composite of death, acute myocardial infarction, stroke, and need for repeat revascularization (either repeat PCI or urgent coronary artery bypass surgery). Successful PCI was defined as final residual stenosis <30% with Thrombolysis in Myocardial Infarction (TIMI) grade 3 flow. Secondary endpoints included major vascular complications (need for emergent endovascular intervention or surgery for acute limb ischemia or vascular injury) and bleeding requiring blood transfusion.

Statistical analysis. Categorical variables are presented as number and percentage. Continuous variables are presented as mean ± standard deviation.

Mean age was 71 ± 11 years, 11 patients (73%) were men, mean STS score was 9.0 ± 8.1, mean logistic EuroScore was 24.4 ± 17.8, and 10 patients (67%) had an EF <30% (Table 1). Fourteen patients (93%) had multivessel CAD, 4 patients (27%) had unprotected left main, and 3 patients (20%) underwent PCI on the last remaining conduit. The left anterior descending coronary artery was treated in 11 patients (73%) and 11 patients (73%) had >1 vessel treated. Mean SYNTAX score I was 32 ± 11. Mean duration of support was 1.8 ± 0.9 hours; 1 patient required continued ECMO support post PCI for 9 days and was excluded from duration of support analysis.

All patients had successful PCI. In- hospital mortality occurred in 3 patients (20%) and MACCE occurred in 3 patients (20%). No patients experienced a stroke, repeat emergent revascularization, or a major vascular complication. Seven patients (47%) required a blood transfusion. An example of a patient with unprotected left main disease undergoing high-risk PCI with the CardioHELP device is shown in Figure 1. Moderate sedation and the preclose technique were used with removal of the cannulas at the completion of the PCI procedure in the cardiac catheterization laboratory.

In the current study, we evaluated a consecutive series of patients undergoing elective, high-risk PCI with the CardioHELP device. All patients had complex CAD and reduced left ventricular systolic function with high SYNTAX scores and were deemed to require hemodynamic support. The CardioHELP device was successfully applied in all patients using moderate sedation and percutaneous insertion and removal techniques. PCI was successful in all patients and short-term clinical outcome was excellent given the high-risk profile of the study cohort.

Percutaneous mechanical device support has been used in patients with cardiogenic shock, out-of-hospital cardiac arrest, and high-risk PCI.^2,3 In the current study, we chose to focus solely on the high-risk PCI cohort using the newest iteration of ECMO, the CardioHELP device. The benefits of the CardioHELP device include rapid initiation, ease of use, device size and portability, and the ability to manage patients without a dedicated perfusionist. Decannulation was done in all patients, except 1, in the cardiac catheterization laboratory following the completion of PCI using the preclose technique. Procedure-related complications were low, with no major vascular complications and an acceptable use of blood products. Use of blood transfusions in the current study was similar to contemporary studies using devices with similar caliber cannulas and sheaths (such as TandemHeart [TandemLife], Impella, and Impella CP).^1,4

Several prior studies have evaluated the use of ECMO for elective high-risk PCI (Table 2).^5-7 Tomasello et al evaluated 12 patients at a single center in Italy using the CardioHELP device.⁵ All patients had an EF <40% and multivessel CAD, 83% had distal left main disease, and mean SYNTAX score I was 30 ± 10. All PCI procedures were successful with no in-hospital MACCE or mortality. Shaukat et al described 5 patients treated at Minneapolis Heart Institute in Minneapolis, Minnesota using the CentriMag blood pump (Thoratec) and the Jostra Rotaflow pump (Maquet).⁶ Percutaneous cannula removal with the preclose technique was used in 2 patients and open vascular repair was used in 3 patients. PCI was successful in all cases, mean duration of support was 3.6 hours, and there were no in-hospital MACCE. More recently, van den Brink evaluated 14 patients at 2 centers in the Netherlands.⁷ PCI was successful in all patients, median duration of support was 2.6 hours, and in-hospital mortality was 7%. The specific ECMO device used was not detailed in the methods section and the authors commented that the only form of mechanical support available at both centers during the study period was ECMO. All patients received general anesthesia and endotracheal intubation prior to the procedure and decannulation was done in an “open” procedure by a cardiothoracic surgeon. The applicability of these findings to contemporary United States practice (moderate sedation, percutaneous removal of cannulas, etc) is not clear.

The most common devices used for high-risk PCI in the United States are the IABP and the Impella devices. Prior studies evaluating these devices for high-risk PCI include BCIS-1 (Balloon Pump-Assisted Coronary Intervention Study) and PROTECT II (the Prospective Randomized Clinical Trial of Hemodynamic Support with Impella 2.5 vs Intra-Aortic Balloon Pump in Patient During High-Risk Percutaneous Coronary Intervention).^11,12 BCIS-1 randomized 301 patients to elective IABP vs no IABP prior to high-risk PCI.¹¹ The primary endpoints of MACCE at 30 days and 6 months were similar. The rate of crossover (bail-out) IABP use was 12%, with these patients requiring a longer duration of IABP use post PCI and an increased hospital stay, but similar MACCE compared with the elective IABP group. Interestingly, at 5 years, there was a survival advantage of elective IABP vs no IABP (hazard ratio, 0.66; 95% confidence interval, 0.44-0.98; P=.04).¹³ PROTECT II randomized 448 patients to Impella vs IABP during high-risk PCI.¹² The primary endpoint at 30 days was similar in the intention-to-treat cohort, with no clear benefit of Impella vs IABP. Randomized studies for elective high-risk PCI using the TandemHeart device have not been completed.

Practice guidelines set forth by the American College of Cardiology/American Heart Association/Society of Cardiovascular Angiography and Intervention give a class IIb indication for the use of hemodynamic support devices in carefully selected high-risk patients.¹⁴ The definition of high-risk PCI in the practice guidelines is similar to the inclusion criteria for the current study and the aforementioned PROTECT II randomized trial. Given the absence of randomized trials showing a benefit for hemodynamic support, firm indications for device support remain controversial and the ability to identify appropriate patients continues to challenge physicians.^15-17

Study limitations. There are multiple limitations in the present study. Given the retrospective nature of the study design, we were unable to account for variables not present within the database. The current study includes a small number of patients treated at 2 hospitals. Both hospitals have experience with high-risk PCI and the CardioHELP device. The applicability of the current results to other centers remains unclear. Although all patients undergoing high-risk PCI with the CardioHELP device were included, we did not collect information on patients undergoing high-risk PCI with other devices, including IABP, Impella, and Impella CP devices. The specific approach to PCI was not standardized and may have included rotational atherectomy, various approaches to left main stenting (t-stenting, v-stenting, etc) and multiple combinations of antiplatelet and antithrombin therapy. This variability may have influenced the decision to use CardioHELP and in-hospital clinical events. Only short-term clinical outcomes were available.

Temporary use of the CardioHELP device for high-risk PCI is associated with acceptable short-term outcome and may be a new option for patients with complex CAD and left ventricular dysfunction.