First-in-Man Successful Use of the SPECTRUM Percutaneous Dual Stage Right Ventricle and Right Atrium to Pulmonary Artery Ventricular Assist Device

Abstract

Over the past decade several minimally invasive mechanical support devices have been introduced into clinical practice to support the right ventricle (RV). Percutaneous cannulas are easy to insert, minimally invasive, and treat acute RV failure rapidly. In December 2021, the Food and Drug Administration approved a new 31 French dual lumen single cannula for use as a right ventricular assist device. This cannula, manufactured by Spectrum, is dual staged. It has inflow ports positioned both in the right atrium as well as the right ventricle for maximal drainage of the right heart. The distal end of the cannula which includes the outflow port is positioned in the pulmonary artery. This design optimizes venous drainage as well as limits suck-down events. Theoretically, direct RV decompression also decreases RV dilation and wall tension, and facilitates improved transmural pressure gradient to reduce RV strain. Here we describe the first-in-man successful use of the dual stage RA and RV to PA Spectrum cannula in a patient with severe COVID ARDS and acute right ventricular failure, bridged to recovery.

Introduction

Acute right ventricular failure (ARVF) is a major cause of morbidity and mortality for patients with acute myocardial infarction or congestive heart failure. Over the past decade several minimally invasive mechanical support devices have been introduced to support the right ventricle (RV).^1–3 Percutaneous options are divided into an intracorporeal axial flow pumps and circuits using extracorporeal centrifugal pumps.^4,5 Centrifugal pump driven circuits require the cannulas to be placed surgically in central great vessels or percutaneously in the peripheral vessels. Percutaneous cannulas are easy to implant, minimally invasive and treat ARVF rapidly. Incremental design changes and optimization of the devices and cannulas has resulted in improved outcomes.

In the past, the percutaneous cannulation strategy for right ventricular mechanical circulatory support consisted of two independent cannulas, an inflow cannula inserted via the femoral vein extending to the inferior vena cava and an outflow cannula inserted via the right internal jugular terminating in the pulmonary artery. The Protek Duo (Livanova, Italy), the most commonly used workhorse dual lumen single cannula, is used as a single access percutaneous right ventricular assist device (RVAD) cannula. It is typically inserted via the right internal jugular vein. The cannula drains blood from the right atrium (RA) and returns blood through the distal tip positioned in the main pulmonary artery (PA). In this configuration any pump and oxygenator can be attached to the cannula. With an oxygenator this configuration is called an oxy-RVAD.^6,7 One of the challenges associated with these cannulation strategies is that blood that returns from the thebesian veins as well as blood that circumvents the RA into the RV may not be adequately drained by the single right atrial drainage port. In December 2021, the Food and Drug Administration approved a new dual stage dual lumen single cannula RVAD manufactured by Spectrum (Cheltenham, England). (Figure 1 and 2) The dual stage cannula has inflow ports positioned both in the right atrium as well as the right ventricle to allow for maximal drainage of the right heart. Here we describe the first-in-man successful use of the dual stage RA and RV to PA Spectrum cannula in a patient with severe COVID ARDS and ARVF with survival to decannulation and recovery.

Figure 1:

Location of the atrial and ventricular drainage ports.

Figure 2:

Figure obtained with permission from Spectrum Medical. Note the cannula curve and dual lumen layout.

Case

A healthy 30-year-old male presented to an outside hospital with shortness of breath, fever and cough and was found to have severe COVID pneumonia. The patient was treated with Tocilizumab, Remdesivir and intravenous Dexamethasone. On hospital day 9, he was intubated due to severe refractory hypoxemia. Conventional ARDS treatments failed and our extracorporeal membrane oxygenation (ECMO) team was consulted. After evaluation for appropriate ECMO criteria a mobile team was deployed to cannulate the patient at the outside hospital. He was cannulated via the right femoral vein with a 25 French Medtronic (Minneapolis, MN) multistage venous inflow cannula and with a right internal jugular vein 19 French Medtronic end-hole outflow cannula. RV function on echocardiography was mildly decreased and mildly dilated. We continued paralysis, deep sedation, and started a sequence of 10 days of daily prone positioning. Three weeks after cannulation his lung compliance had improved to 24 ml/cmH₂0, his venovenous (VV) ECMO settings were weaned to 21% FiO₂, and sweep gas of 0.5 LPM. As sedation was reduced and the patient began to spontaneously breathe, he experienced patient self-induced lung injury due to tidal volumes in the 900+ mL range, despite very low-pressure support settings. This caused rapidly worsening ARDS with a compliance of 6 ml/cm H2O, hypoxia and RV failure, forcing the team to increase his VV ECMO settings. On echocardiography his RV was moderate to severely decreased in function and moderately dilated. He required repeat paralysis and another cycle of 10 days of proning. Three weeks after this setback he was again weaned to minimal VV ECMO settings. His compliance had improved to 25 ml/cmH₂0 and his tidal volumes improved to the ~ 400 ml range. Again, upon awakening, he had profound hypoxia requiring escalation of VV ECMO support for a second time. His compliance deteriorated to single digits. Bronchoalveolar lavage demonstrated a Serratia pneumonia which was treated with Cefepime. A pulmonary artery catheter was placed and PA pressures were discovered to be 75/35 mmHg (mean 45 mmHg) during attempts to decrease sedation. His central venous pressure was 22 mmHg. The pulmonary vascular resistance was noted to be 3.5 WU. His TTE demonstrated mildly decreased RV function and severely dilated RV size. There was moderate tricuspid regurgitation. The patient was on epinephrine infusion at 2 mcg/min and inhaled epoprostenol at 50 ng/min. At this point, we suspected that during sedation liberation the patient experienced large fluctuation in intrathoracic pressures resulting in dynamic and inducible pulmonary hypertension, RV failure, VV ECMO recirculation, hypoxia and complete failure to wean from ECMO.

Device Description and Cannulation Sequence

Therefore, at ECMO day 53, he was taken to the hybrid operating room for VV ECMO conversion to an oxy-RVAD utilizing the novel Spectrum dual stage dual lumen cannula. The patient was first adequately heparinized. The patient was temporarily bridged with femoral-femoral VV ECMO. This was achieved by inserting a multi-orifice 22 French Edwards (Irvine, CA) quickdraw venous cannula via the left femoral vein with the distal tip positioned in the right atrium. The existing outflow tubing was then disconnected and reconnected to the newly placed femoral cannula and temporary femoral-femoral VV ECMO was initiated. After a brief period of stabilization, the in situ 19 French right internal jugular cannula was removed and a new vascular access site in the right internal jugular vein was obtained. A 7 French Arrow (Teleflex, NC) balloon tipped flow directed catheter was positioned into the right PA with transesophageal echocardiography and fluoroscopy guidance. Using this catheter, the RA, RV and PA pressures were measured and recorded. Hand injected contrast venogram using digital subtraction angiography was performed to confirm the caliber of the superior vena cava and location of the pulmonary valve. Next, a 260 cm Lunderquist wire with a 4 cm floppy straight tipped end (Cook Medical, Bloomington, IN) was placed through the catheter and positioned in the right interlobar pulmonary artery. Care was taken to avoid passing the wire tip distal to main artery and injuring the segmental pulmonary artery vasculature. After serial dilation of the insertion site with a Sorin dilator kit (Livanova, UK) the 31 French cannula was passed over the wire and the distal tip was positioned into the right PA. (Figure 3) Wet to wet connection with a new circuit was made in standard fashion and flows were increased to ~4.0 liters per minute. RV dilation resolved on echocardiography and the cannula migrated back 2 cm due to the reduction in circular tension with RV decompression. We then manually retracted the cannula back until it was positioned in the main PA. 2D and color doppler were used to guide retraction and confirm blood flow directed at the PA bifurcation. After suture the cannula in place, the temporary femoral-femoral VV ECMO was removed and the patient returned to the intensive care unit. (Figure 4)

Figure 3:

Diagram demonstrating the final location of the RA and RV to PA Dual stage dual lumen right ventricular assist device.

Figure 4:

Post-operative chest roentgenogram demonstrating the positioning of the cannula with radioopaque markers. The large 31 French diameter cannula can be seen extending down through the tricuspid valve into the right ventricle. The smaller inner lumen is seen extending beyond the inflow cannula extending up to the main pulmonary artery.

Clinical Course

Over the next two weeks, a third cycle of proning was commenced to help treat dense posterior consolidation of the lungs. Daily chest radiography was obtained to monitor cannula position and weekly TTE was performed to monitor right ventricular function. Anticoagulation was maintained with bivalirudin with an activated partial thromboplastin time goal of 50 – 60 seconds. The patient’s lung compliance slowly improved and oxy-RVAD FiO₂ and sweep settings were gradually reduced. Sedation weaning was accomplished this time with the patient actively able to participate in physical therapy without RV failure, insurmountable pulmonary hypertension, or hypoxia. After a period of stabilization, the oxy-RVAD (day 20) was weaned by gradually reducing the sweep to 0 LPM and weaning the FiO₂ to 21% for 48 hours prior to decannulation. Unlike traditional VV ECMO, in addition to sweep off trial and FiO₂ minimum settings, the flow was also reduced to 2 LPM and TTE and CVP was used to monitor for progressive RV failure. The CVP was maintained at less than 12 and the TTE demonstrated normal size and systolic function of the right ventricle at this reduced setting. When the patient tolerated these parameters the Oxy-RVAD was discontinued on ECMO day 73 and the patient was transitioned to a HemoLung RAS (A-Lung, Pittsburgh, PA) to treat residual hypercarbia. The HemoLung device was subsequently removed after the patient was ventilator liberated, awake, participating in physical therapy and ambulating. He was subsequently discharge to short-term rehabilitation on hospital day 99.

Discussion

We report the first-in-man successful use of the novel Spectrum percutaneous dual stage dual lumen RV and RA to PA RVAD. Increasing the percutaneous mechanical circulatory support options is highly desirable given the known morbidity and mortality associated with ARVF. Institution of RVAD support should be done, ideally prior to the onset of RV failure. More often than not, RVAD placement occurs after renal failure and liver failure has progressed to the point of irreversibility. With the advent of percutaneous options, that we can rapidly deploy directly in the intensive care unit under portable c-arm, RV MCS can be instituted much earlier in the arc of RV failure. The Spectrum cannula is unique in that it drains venous blood from the RA and RV, increasing the chances of unloading the RV. Clinically one of the challenges with single stage drainage is the inability to properly position the final position of the inflow orifices in the right atrium. Based on the patient height, access location, RV size and the location of the distal tip, the inflow orifice in a single stage cannula can be variable in position. If the inflow ports fall directly at the RA/superior vena cava junction high RVAD flows can result in collapse of the SVC, limit the flow rates, and result in repeated suck-down events. Positioning the inflow ports both in the RA as well as the RV facilitates optimal and balanced drainage to minimize risk of acute venous suck-down and circuit flow disruption.

Providing additional inflow drainage in the RV also allows for complete drainage of the coronary sinus blood which may escape the RA ports as well as the thebesian veins that directly drain into the RV. This blood represents the most hypoxic blood in the human circulation and constitutes 5 percent of total cardiac output. The average SvO₂ of the coronary sinus is 47%.⁸ In an oxy-RVAD configuration, the final SaO₂ of the pulmonary artery is dependent on the oxygenated blood that leaves the ECMO circuit and the blood that bypasses the ECMO circuit. Capturing this hypoxic blood in the ECMO drainage cannula reduces shunting. Second, direct RV decompression decreases the RV strain and provides RV protection. Furthermore, direct RV offloading and reduction in RV distension reduces myocardial wall tension. Although these factors are theoretical considerations, a prospective trial comparing the different cannulas and platforms should be considered in the future.

One of the drawbacks to the RA and RV to PA cannula includes a large lumen size that traverses the tricuspid valve and may create supraphysiologic RV inlet pressure gradients. Furthermore, significant drainage velocities and potential tricuspid valve trauma is possible if the inflow orifice abuts the valve. RV drainage relies solely on pulmonic valve competency to prevent regurgitation resulting in recirculation. Finally, the RA and RV drainage ports can provide acute flow stability during variations in venous return, but the thrombotic risk to reduced RV flows still needs to be studied. Future experience and experimentation with each cannula, whether in silico (computation fluid dynamics) or in vitro, may elucidate the thrombotic and valvular risks associated with large and/or dual-stage cannulas in the atrium, ventricle and at the tricuspid valve. Alternative strategies for extracorporeal right ventricular mechanical circulatory support exist. A similar cannula, the Protek Duo cannula, is a single stage dual lumen RA to PA cannula. This cannula exists as a 27 French and 31 French cannula and therefore is limited by its size and rigidity, in similar fashion to the Spectrum cannula. The lowest profile option is to use two independent cannulas. The inflow cannula, a 25 French multi-orifice long venous drainage can be placed through the femoral vein extending into the inferior vena cava and RA, and an independent 17 or 19 French long venous end hole right internal jugular Biomedicus (Medtronic, MN) as Oxy-RVAD outflow. This cannula was originally designed as a percutaneous PA vent catheter however we have increasingly used it as an outflow cannula in an RVAD setting.

An area of active research for right ventricular support is the use of pulsatile versus non-pulsatile mechanical support.⁹ This concept is heavily studied for left sided durable support, however data from animal experiments suggests that under continuous RV mechanical support there may be an increased risk of extravascular lung water accumulation, pulmonary edema, and pulmonary hemorrage. Taguchi et al. demonstrated in 14 mongrel dogs that non- pulsatile perfusion of the lungs during a 10-hour bypass period resulted in a greater tendency of lung water to accumulate.¹⁰ Under continuous-flow perfusion of the lungs, a significant increase in pulmonary artery pressure and an increase of pulmonary vascular resistance are observed. This is particularly true in scenarios where there is fixed pulmonary hypertension. Low RVAD flows may mitigate the potential adverse effects of high resistances in the pulmonary vasculature.¹¹

We demonstrated the successful weaning to decannulation of a patient with COVID ARDS with RV failure. Two attempted weaning trials without RV support resulted in major clinical setbacks. There are several groups using RVAD support for COVID ARDS and this remains an active area of investigation.^12–14 There is some evidence suggesting that RV support may improve survival in severe COVID ARDS in patients requiring ECMO support.¹⁴ This case report, although emphasizing the cannula design, also represents a novel strategy to support patients with severe ARDS.

Conclusion

The novel spectrum dual stage RA and RV to PA cannula provides an alternate option for RV support. This cannula may offer advantages related to reduction in RV dilation, RV strain, and improve RV transmural pressures. Future studies comparing cannulation strategies and cannula design are necessary to provide insights in which clinical conditions to use the cannula as well as to define flow parameters, anticoagulation requirements and mid-term to long term sequalae of cannula traversing the tricuspid and pulmonary valve.

Appendix:

Consent Form for Case Reports

University of Pennsylvania

Hospital of the University of Pennsylvania

Presbyterian Hospital

Pennsylvania Hospital

Department of Anesthesiology and Critical Care

Case Report: First-in-Man Successful Use of the SPECTRUM Percutaneous Dual Stage Right Ventricle and Right Atrium to Pulmonary Artery Ventricular Assist Device

Principal Investigator: Asad Usman, MD, MPH, University of Pennsylvania, 267.602.7025

You are being asked to consider allowing Dr. Asad Usman to use information about your COVID ARDS and requirement for venovenous and venopulmonary extracorporeal membrane oxygenation to write what is called a case report. Case reports are typically used to share new unique information experienced by one patient during his/her clinical care that may be useful for other physicians and members of a health care team. A case report may be published the Journal of Cardiac Surgery for others to read, and/or presented at a conference. This form explains the purpose of this case report. Please read this form carefully and take your time to make your decision and ask any questions that you may have.

The purpose of this case report is to inform other physicians that may be considering the use of right ventricular assist support during extracorporeal membrane oxygenation may understand the pros and cons of using the SPECTRUM dual stage dual lumen cannula.

Your information being used for this case report includes the course of your hospitalization.

Dr. Asad Usman is obligated to protect your privacy and not disclose your personal information (information about you and your health that identifies you as an individual e.g. name, date of birth, medical record number). When the case report is published or presented, your identity will not be disclosed.

Although your personal information collected or obtained will be kept confidential and protected to the fullest extent of the law, there is a limited risk associated with this case report that could result in a loss of confidentiality by virtue of your unique experience.

You will not directly benefit from participating in this case report. The information that can be shared with other health care professionals, however, may improve the care that is received by others in the future.

Allowing your information to be used in this case report will not involve any additional costs to you. You will not receive any compensation.

Taking part in this case report is your choice (voluntary). You may choose not to take part or you may change your mind at any time. However, once the case report is written and published, it will not be possible for you to withdraw it. Your decision will not result in any penalty or loss of benefits to which you are entitled including the quality of care you receive.

You will be told about any new information relating to this case report that may affect you.

Your signature below means that you have read the above information about this Case Report and have had a chance to ask questions to help you understand how your information will be used and that you give permission to allow your information to be used in this case report.

If you have any questions please contact Asad Usman at 267.602.7025

SUBJECT CONSENT TO PARTICIPATE

Name of Participant: Andrew Definis

Participant/Legally Authorized Representative

By signing this form, I confirm that:

• The case report has been fully explained to me and all of my questions have been answered to my satisfaction

• I have been informed of the risks and benefits, if any, of allowing my information to be used in this case report

• I have been informed that I do not have to participate in this case report

• I have read each page of this form

• I authorize access to my personal health information (medical record) as explained in this form

• I have agreed to participate in this case report

Patient Name: Andrew Definis

PHONE CONSENT Obtained at the following number 267.934.3010

Email Consent: amydefinis@comcast.net

Verbal Consent confirmed with Dr. Audrey Spelde (ECMO Attending) 267.693.3673 &

Carly Krespan 717.725.1757 (Clinical research Coordinator) as well as with Amy Definis Mother of patient

I have carefully explained to the subject the nature of the above project. I hereby certify that to the best of my knowledge the person who is signing this consent form understands clearly the nature, involved in his/her participation and his/her signature is legally valid. A medical problem or language or educational barrier has not precluded this understanding.

_Phone Consent (See above) 

Name of Person Obtaining  Signature  Date: 6/27/2022