Interested in Joining ICECAP?
The list of sites included on the initial application are posted here. If you are not part of this list, we welcome your interest through this survey. For sites not on the list, new sites will need to partner with an existing SIREN hub. We can help find matches if one is not apparent to your site.
We anticipate this trial to be run in the SIREN network, which will provide a high performing core of sites. However, we have designed the trial anticipating a need for 50 sites, which is more than SIREN is expected to provide. Therefore, we are interested in the participation of other high performing sites as well, especially those with a proven history of successful enrollments in NETT and ROC trials. Other sites with an interest in post cardiac arrest care and sufficient numbers of potentially eligible subjects are also welcome.
ICECAP requires sites that can enroll an average of 9 subjects per year. More detailed eligibility criteria are available, but in brief, eligible subjects will be adult comatose survivors of out-of-hospital cardiac arrest, successfully cooled to under 34 degrees within 4 hours of onset of cardiac arrest, with consent obtained from family/LAR, and randomization performed within 6 hours of onset of cooling. Subjects will include survivors of cardiac arrest due to either shockable or non-shockable presenting rhythms.
To enroll and perform well in this trial, sites will need study teams able to respond rapidly to the ED whenever potential subjects arrive. Study teams will need to round on enrolled subjects daily for up to one week of hospitalization, and then follow up at hospital discharge, by phone at day 30 and 180, and in person at day 90. Daily rounding is to reinforce the study protocol and clinical standardization. The 90 day visit involves a brief exam and functional assessment by an assessor blinded to intervention, and a 60 minute computerized battery of patient reported outcomes and neuropsychological tests.
Preliminary data suggest that length of stay in those comatose survivors of cardiac arrest that recover is typically driven by extrinsic events such as ability to schedule AICD placement or late revascularization procedures, or arrangements for subacute rehabilitation, rather than by time of awakening, so durations of stay are unlikely to be affected by the study protocol in this patient subset. In those that do not recover, length of stay is primarily driven by the timing of withdrawal of life sustaining care. Early withdrawal of life sustaining care after determination of brain death is unaffected by the study protocol and so durations of stay are unlikely to be affected by the study protocol in this patient subset. Timing of withdrawal of life sustaining care in those with varying degrees of poor prognosis is driven by a patient’s surrogate medical decision maker/LAR. Decision making about withdrawal of life sustaining treatment is standardized and incorporated into the study informed consent process in this protocol, but remains driven by the patient’s surrogate decision maker just like in routine care. Patient’s whose family/LAR opt for earlier withdrawal than is consistent with the protocol will not be enrolled. Therefore, the durations of stay of subjects that are enrolled are still unlikely to be affected by the study protocol even in this patient subset. Given these observations, and the limited number of eligible subjects at any given site per year, the study should not cause problems for ICU utilization, and expenses related to hospital duration of stay are not considered to be research costs.
Participating sites will be reimbursed primarily on a per-subject basis. Per subject payments will be fixed and inclusive of both direct and any indirect costs. The per subject payment proposed is based on anticipated average number of study coordinator hours of effort and investigator hours. In the grant application we will request a per-subject payment to sites of $10,000, but this is subject to funding approvals and budget revisions. We also budget a modest one time start up of $5000 to support application to the central IRB and other system initiation costs. There are no anticipated imaging or pharmaceutical costs. Eligible subjects are those cooled in routine clinical care prior to enrollment, so cooling devices are not research-related costs.
A brief synopsis of the trial, the complete draft of the study protocol, and a working in-progress draft of the clinical standardization plan are publicly available on the ICECAP study website, as well as these questions and answers. There is also a link there to the brief ICECAP site survey that should be completed for each enrolling hospital interested in participating as a study site. Interested sites meeting criteria will subsequently be contacted for a letter of support to be signed by co-investigators from Emergency Medicine and a Critical Care discipline, and to complete a chart review to provide more granular data on the number of potentially eligible subjects at your site.
No. This is not an EFIC study.
Yes. This study is conducted under an IDE from the FDA, which has already been granted.
No, we don’t think so. The European trial didn’t really do anything to answer the duration question one way or the other. The trial randomized 355 comatose survivors of cardiac arrest to 24 or 48 hours of cooling. The study was powered to detect a 15% absolute difference in survival with good neurological outcome, but observed a 5% improvement in good outcome and a 7% improvement in survival. These were not statistically significant but would have been clinically important if real and not random. It is unclear what can be learned from this underpowered neutral study. It is consistent with (but not indicative of) an effect of duration of cooling on outcome in humans. What it really shows is the need to properly design and power trials to detect clinically important effects. The power problem is obvious. Perhaps less obvious, but really important, is that the design is mismatched to the question. Duration is really a question of dose. The notion of accurately finding the optimal dose of anything with an A/B comparison of two subjectively chosen, relatively similar, doses is really silly. One needs a dose-finding methodology to do that. Our interpretation is that the Kirkegaard paper and its point estimates may be slightly encouraging, but really don’t affect the ICECAP submission in any meaningful way. Other than as an example of how not to approach the question.