Extracorporeal CPR Hits Snag in INCEPTION, but Some See Unmet Promise

Lack of a protocol and the prolonged times to start of care may have torpedoed a promising study, one expert says.

Extracorporeal CPR Hits Snag in INCEPTION, but Some See Unmet Promise

Compared with cardiac arrest patients assigned to conventional CPR, those assigned to receive extracorporeal CPR (eCPR) after arrival at the hospital had similar rates of survival with favorable neurologic outcome in the Dutch INCEPTION trial.

The addition of extracorporeal membrane oxygenation (ECMO) to CPR is meant to restore perfusion and limit brain injury when there is no return of spontaneous circulation after standard advanced life support. But evidence to support its value in clinical practice has been inconclusive, note Martje M. Suverein, MD (Maastricht University Medical Center, the Netherlands), and colleagues.

They conducted the multicenter INCEPTION trial with the expectation, based on prior studies, that eCPR would increase 30-day survival to 30%, far beyond the typical 8% to 10% survival rate seen with standard CPR in OHCA.

Those hopes fell short, however. Instead, Suverein and colleagues found a nonsignificant 4% difference between groups, suggesting that despite the appeal of eCPR, major gains are “not self-evident when extracorporeal CPR is pragmatically implemented, even in cardiosurgical centers where providers are experienced in its use.”

Their findings stand in contrast to the ARREST trial, a small single-center US trial that used a highly standardized eCPR pathway that was associated with 36% higher survival compared with conventional CPR.

“It's very common in many centers that these patients are considered dead when they arrive,” Demetris Yannopoulos, MD (University of Minnesota Medical School, Minneapolis), the lead author of ARREST, told TCTMD. He added that “the devil is in the details” when it comes to setting up and applying a cohesive system for eCPR, which should include a reorganization of emergency response infrastructure and centralization of care.

Suverein and colleagues acknowledge in the paper that they did not have a protocol for emergency medical services (EMS) and hospital staff, which Yannopoulos called “mind-boggling” and said is likely a major factor in why the trial was unsuccessful. The authors say although it may be seen as a limitation, this feature “may also increase the generalizability of the resuscitation procedures and extracorporeal CPR.”

Yannopoulos disagreed vehemently.

“INCEPTION had the potential, if it was done in a very well-organized system, to show remarkable results because the mean time between cardiac arrest and arrival to the hospital was 36 minutes. The average time that it should take for a patient to be cannulated and put on pump support is about 7 to 10 minutes. So, that would have put people in the 45 minutes total CPR duration before being on ECMO,” he said. “Our study and others have shown that if you put people on ECMO within 45 minutes, their chance for survival is close to 70%.”

The median time from start of arrest to start of ECMO flow in INCEPTION was 74 minutes. In the paper, Suverein and colleagues acknowledge that their times to initiation of cannulation and time to start of oxygen flow were longer than those of ARREST and another single-center trial conducted in Prague that was terminated early for futility when eCPR did not significantly improve outcomes. They say this  is a reflection of “ differences in such factors as team experience, logistics, and caseload.”

In an accompanying editorial, John F. Keaney Jr, MD (Brigham and Women’s Hospital, Boston, MA), and Thomas Münzel, MD (University Medical Center Mainz, Germany), contend that the mixed findings from the three trials should not be taken as an indication that eCPR is not useful in some patients.

“Collectively, results of all three randomized trials showed a numerical advantage in their respective primary outcomes in favor of extracorporeal CPR as compared with conventional therapy,” they write. “Thus, it is possible that larger trials are needed for the statistical power necessary to capture any benefit associated with extracorporeal CPR as compared with conventional CPR.”

Few Between-Group Differences

For INCEPTION, patients (mean age 54 years in the eCPR group and 57 years in the conventional CPR group; 90% and 89% men, respectively) were randomized to receive e CPR (n = 70) or conventional CPR (n = 64) after arriving via EMS at one of 10 participating cardiosurgical centers.

All had witnessed OHCA, CPR started within minutes, and an initial ventricular fibrillation, ventricular tachycardia, or a shockable rhythm detectable on defibrillator, but no return of spontaneous circulation despite 15 minutes of advanced CPR. The cause of arrest was acute MI in 73% of those in the eCPR group and 81% in the conventional CPR group.

At 30 days, the percentage of patients meeting the primary outcome of survival with a Cerebral Performance Category score of 1 or 2 (normal or disabled but independent) was 20% in the eCPR group and 16% in the conventional CPR group (P = 0.52).

Compared with standard CPR, more of those who received eCPR survived until admission to the ICU. However, the number patients who actually survived to discharge from the ICU was small and similar between groups, as was the number who survived to hospital discharge.

At 6 months, survival with a favorable neurologic outcome also showed no significant difference between treatment groups.

A Call for Training and Systems of Excellence

According to Suverein and colleagues, another difference between their study and the ARREST and Prague trials is a far lower caseload at the participating hospitals. During the study period of May 2017 through February 2021, there were 51 cannulations attempted across the 10 centers, with 46 being successful. The small number of patients points to another issue, say Keaney and Münzel, namely that “we have not adequately identified the specific subsets of patients who may benefit from  extracorporeal CPR.”

But Yannopoulos told TCTMD that a more important future endeavor is understanding and perfecting the aggressive cycle of care needed once eCPR begins. He believes many patients are withdrawn from support too soon at the discretion of neurologists based on the conclusion that they cannot recover. In his experience, it’s not uncommon to have reversible bleeding or traumatic complications that can be managed and pave the way for survival.

“I really think with this type of therapy, it's a matter of training that isn’t really happening right now,” he said. “In addition, you have to create systems of excellence, and effectively, you need to build a program slowly by looking at the idiosyncrasies of your own hospital, your own city, etc, to be successful.”

Sources
Disclosures
  • Suverein, Münzel, and Yannopoulos report no relevant conflicts of interest.
  • Keaney reports being employed by the New England Journal of Medicine as an associate editor.

Comments

1

Zend Mayin

1 year ago
"Suverein and colleagues acknowledge in the paper that they did not have a protocol for emergency medical services (EMS) and hospital staff, which Yannopoulos called “mind-boggling” and said is likely a major factor in why the trial was unsuccessful." This likely means that ECPR centers have their own set of protocols, not one defined by INCEPTION. The trial looked at the effects of a therapeutic (ECMO) for OHCA meeting certain conditions, not an ECPR package from EMS until good neurologic survival. Hospitals with an ECPR program could participate in the trial and enroll a set of their patients for the study and randomize them at the ED. Currently, 13 hospitals provide ECPR for OHCA in The Netherlands. 10 of these hospitals participated in INCEPTION. These are all large academic or large non-academic teaching hospitals in The Netherlands that are designated heart centers. There are 15 designated heart centers spread at the country all located at large academic or teaching hospitals. Of which 13 provide ECPR, and 2 (MCL Leeuwarden - far north, and MST Enschede - far east) do not as of yet. These centers specialize into complex cardiac care and are the only centers that are allowed to do open heart surgery. They also provide PCI capability, together with 15 other designated PCI centers. 10 hospitals are in the ELSO registry, of which 4 are centers of excellence (1 platinum, 2 gold, 1 silver). However, many hospitals in The Netherlands stopped considering applying and paying for registries as it is generally accepted they are a waste of resources (time and money), so this doesn't mean a lot. So all of these hospitals should have ECMO experience, but not necessarily ECPR experience. So a hospital could have entered the study that is completely unknown with ECPR, and other centers could have an ECPR program already. For some hospitals OHCA ECPR was new, but already did IHCA ECPR. The important thing is that 4 hospitals (lead by Maastricht MUMC+ academic hospital in the far south) started the program, with the remaining hospitals joining after the study protocol was released. Thus most hospitals only started enrolling patients later into the study and might not have even be fully up and running by then if they were centers new to OHCA ECPR. Some hospitals even enrolled patients for less than a year. You could also question that in a very high performing system like in The Netherlands, which already achieves far higher survival rates with the stay and play model than most other western societies, the benefits of ECPR and transport could decrease as fewer ECPR viable patients, and thus the most viable patients, having achieved ROSC already on scene or before ECPR initiation, while these patients would be more likely to not have achieved ROSC in a lower performing system, like the one in Minnesota (state where ARREST trial took place). This could lead to lower ECPR exposure and thus experience, and it could lead to the most viable ECPR patients in other trials not needing ECPR in a better performing system. At the hospital in my region, it ran their own ECPR program and made separate agreements and protocols for the sole EMS agency in the region. It was new to ECPR for OHCA and only started using it since early 2018. These EMS regions informed their staff on these protocols, and the ones of the surrounding ECPR centers. The INCEPTION trial ran paralal to this, and for that they enrolled and randomised patients as well. This is one of the smaller non-acadeamic participating centers, also in a farily ECPR saturated area, with 2 large academic ECPR centers nearby. They are one of the centers that joined the study later after the study protocol became known. And thus they also started enrolling patients for the trial quite late into it. On its own, they expect they'd need ECPR around 20 times per year for such calls. As there was no clear protocol for hospitals and EMS attached to the trial it makes sense. They at least did that in my region, I cannot confirm how it was done elsewhere. But it was the hospitals setting up an ECPR program and making agreements and protocols together with the regional EMS organizations. The study didn't look into setting up an ECPR program, it had the goal of looking at the performance of ECPR in an RCT at multiple centers with their own ECPR program.