August 31st, 2011

New Resuscitation Strategies Fail to Improve Outcomes After Cardiac Arrest

Two trials from the Resuscitation Outcomes Consortium (ROC) investigators were unable to demonstrate meaningful improvements to resuscitation strategies after cardiac arrest. The two trials, one testing an impedance threshold device and the other examining a strategy of early versus late rhythm analysis, have been published in the New England Journal of Medicine.

In the first trial, 8718 patients were randomized to treatment with an active or sham impedance threshold device (ITD) intended to improve venous return and cardiac output during CPR. There was no significant difference between the groups in the percentage of subjects who survived to hospital discharge with satisfactory function:

  • 6.0% (260 patients) in the sham-ITD group and 5.8% (254 patients) in the active-ITD group

In the second trial, which utilized a cluster-randomized design, 9933 patients with out-of-hospital cardiac arrest received either CPR for 30-60 seconds or CPR for 180 seconds prior to ECG analysis. Once again, no difference in the primary outcome of survival to hospital discharge with satisfactory functional status was observed:

  • 5.9% (273 patients) in the later-analysis group and 5.9% (310 patients) in the early-analysis group

There were no significant differences in any of the secondary outcomes or among subgroups in either of the trials.

In an accompanying editorial, Arthur Sanders writes that although the early versus late rhythm analysis trial ruled out any significant difference between the two groups, it did not answer “the more critical question” of whether any CPR before rhythm analysis is beneficial.

More generally, Sanders argues that “there are fundamental tensions between the principles of randomized trial design and the practice of resuscitation that make the conduct of any clinical trial of out-of hospital cardiac arrest challenging.” Out-of-hospital cardiac arrest should be considered a public health problem rather than a disease process and might be better tackled with a continuous-quality-improvement model, he writes.

2 Responses to “New Resuscitation Strategies Fail to Improve Outcomes After Cardiac Arrest”

  1. Graham Nichols, a member of the ROC consortium and a co-author of the papers, sent the following comment to CardioExchange:

    I dont think that the editorial raises critical (as in essential) issues.

    Medicine is replete with examples of interventions that appear to work in observational studies but fail to work in randomized trials. Hormone replacement therapy is one; it appears that the impedance threshold valve and CPR before defibrillation are others. In Arizona and Wisconsin, and North Carolina, implementation of multiple change in resuscitation practices (including but not limited to adopting continuous compressions) were associated with improved outcomes in EMS systems that had not previously invested much effort in monitoring resuscitation practices. Do these interventions work? Or are the observed improvements because of Hawthorne effects? We do not know.

    I note that the benefit observed in the unblinded, industry-sponsored trial of the ITD (published in Lancet, first author Aufderheide) is similar to the pooled effect observed in unblinded vs. blinded trials.
    Does the emperor have any clothes on? We will find out, as the Resuscitation Outcomes Consortium has begun enrollment in a large randomized trial to test the intervention advocated by Dr. Sanders.
    On one level, whether intervention works or whether there were Hawthorne effects is a moot point as survival improved.

    On another level, resources are always constrained, so we must strive to develop good evidence of what works so we can improve outcomes as efficiently as possible.

  2. This week’s NEJM study found that use of an impedance threshold device (ITD)with standard CPR did not improve survival outcomes for out-of-hospital cardiac arrest. This differs from the recent Lancet study published earlier this year, in which ITD use was reported to be associated with higher survival. However, as mentioned by Dr. Graham Nichol, it is important to keep in mind that the Lancet study was unblinded. Moreover, in the Lancet study, the intervention was a combination of active compression-decompression CPR (with a suction cup and metronome) plus use of an ITD. Unfortunately, the Lancet study had withdrawn their third comparison group with standard CPR and an ITD.

    This leaves unanswered whether the current NEJM study differs from the Lancet study because of the lack of blinding in the Lancet study OR whether the positive survival findings from the Lancet study was due to the effect of active compression-decompression CPR and not from the ITD device.

    Competing interests pertaining specifically to this post, comment, or both: