Use of mechanical circulatory support in the management of cardiac arrest has generated considerable discussion in recent years. Much of this discussion has revolved around extracorporeal membrane oxygenation (ECMO)., In this issue of Resuscitation, Vase et al. introduce the Impella® device as a novel technology for restoration of spontaneous circulation (ROSC). The percutaneous Impella CP® was used to augment blood flow in the left ventricle during cardiac arrest. Impella®, in contrast to venoarterial (VA) ECMO, may create a more physiologic type of support by propelling blood antegrade from the left heart through the aorta. In cardiogenic shock, studies have supported the use of Impella®,, , , , but in cardiac arrest, the data has been limited to porcine models., This human case series challenges us to assess whether Impella® is efficacious, safe even feasible for cardiac arrest resuscitation.
In terms of efficacy, this paper has two main contributions. First, it demonstrates that Impella® placement can be done in cardiac arrest. The authors were able to successfully deploy Impella® in all eight cardiac arrest patients. Having been involved in an intra-arrest Impella® insertion myself, the fluoroscopic manipulation needed to advance the device beyond the aortic valve during active chest compressions is an impressive task. Time to insertion of the device was not documented, but will be a key consideration moving forward.
The second contribution is that some of these patients will survive neurologically intact. The author’s cohort was heterogeneous in both arrest location and duration. Half of the cardiac arrests occurred in the cardiac catheterization laboratory (cath lab) and six of eight had a duration of chest compressions of 30?min or less. Comparison of this highly selective cohort to other cardiac arrest cohorts is difficult. The authors compare their survival to Wang et al.’s ECMO arrest data. This offers little useful comparative value given the substantial dissimilarity of the cohorts. The expected survival in Vase’s cohort is difficult to ascertain, but the 50% survival is promising.
In terms of complications, the study cited significant vascular problems. Of the eight cardiac arrest patients, half had vascular complications, including one extremity amputation and one retroperitoneal bleed requiring massive transfusion. Additionally, the difficulty in diagnosing vascular injuries among patients that die must be considered. Of the four survivors, three had vascular complications. The authors compare their complication rate in cardiogenic shock with patients receiving ECMO for refractory cardiac arrest (CHEER Trial). This comparison is problematic in that catheters placed during chest compressions will intuitively portend a higher chance of complication. Also, the CHEER Trial’s “complications” were primarily related to the insertion of a distal perfusion catheter. This is empiric practice in many ECMO centers to prevent limb ischemia, with some data supporting empiric placement with Impella® as well. The complications noted by the authors support my fears that intra-arrest Impella® placement would have a high risk of complication. Without autopsy data, safety concerning the aortic valve, left ventricle and proximal aorta is only conjecture.
While vascular injury is a common complication of mechanical devices, an often overlooked problem is the failure to run a high-quality standard resuscitation. A patient who develops ventricular fibrillation does not require ECMO or Impella®, they need defibrillation. While mechanical devices may provide benefit to cardiac arrests overall, initially noninvasive strategies may be more efficacious. This applies not only to patients developing cardiac arrest in the cath lab, but also those being considered for early transport from out-of-hospital cardiac arrests.
In terms of feasibility, the authors assert that Impella® is easier to deploy than VA-ECMO. While Impella®requires one less cannula insertion, the overall procedure requires the patient to be transported to a fluoroscopy equipped room with physicians trained at passing a catheter across the aortic valve during active chest compressions. Overall, these additional resources and training requirements for Impella® insertion will limit its utility in many centers. The authors, however, were able to demonstrate two survivors who suffered in-hospital arrests, were placed on a mechanical chest compressor, and transported to the cath lab. The same institution concurrently published a case report involving Impella® insertion on a patient who survived after being transported from an outside hospital in cardiac arrest. These cases will certainly make us re-think this perceived limitation of Impella®.
Looking forward, management of Impella® during cardiac arrest will need to be optimized. Placement of Impella® in the left ventricle (LV) has a significant disadvantage in that it relies on a functional right ventricle (RV) for adequate flow. In cardiac arrest, the RV is often not only non-functional but may be the origin of the cardiac arrest, e.g. pulmonary embolism or RV infarction. Suction events occurred in all of the Impella® patients suggesting more RV output is required. Improving intra-arrest LV volumes could potentially improve the efficacy of an Impella® device but the question of how to accomplish this remains. Similar to the concept of modified chest compressions during left ventricular assist device (LVAD) arrests, continued chest compressions after Impella® insertion may be of benefit. The optimal duration of concomitant chest compressions with Impella®placement is an interesting physiologic question that will require more investigation.
The future of resuscitation may involve offsetting the limitations of one device with the strengths of a second device. One of VA-ECMO’s limitations in cardiac arrest is differential hypoxemia. After ROSC, the heart will pump blood from lungs whose function has been compromised from prolonged chest compressions. With a functioning LV, the well oxygenated blood from the ECMO circuit will preferentially perfuse the lower body leaving relatively deoxygenated blood going to the coronaries and carotids. This combined with the increased afterload an VA-ECMO circuit places on the heart makes combinations of Impella® and ECMO theoretically advantageous. Deploying VA-ECMO in combination with a LV Impella® may improve venting of the LV. The use of venovenous ECMO (VV-ECMO) with left ventricular augmentation using Impella® may provide oxygenation, ventilation and perfusion. Well oxygenated blood from the VV circuit would be directed anterograde through the aorta into the coronaries and carotids. Even the use of isolated VV-ECMO using a pulmonary artery return catheter may be a better use of mechanical support for cardiac arrest of RV origin. The introduction of the RV Impella® increases the possible support permutations even more.
Overall, the authors should be commended on continuing to push the boundaries of resuscitation technology. This is an exciting time to be part of resuscitation, and this study shows just the tip of the iceberg for new and improved processes in the mechanical management of refractory cardiac arrest.
No financial disclosures.
No conflict of interests.