Thanks to defibrillators, burly-armed EMTs, speedier cardiac revascularization, and induced hypothermia, the mortality rates after ventricular tachycardia or fibrillation have improved markedly for both in- and out-of-hospital cardiac arrests. But mortality rates after PEA and asystole remain stubbornly steady, seemingly resistant to any of the above interventions.
Background: People suffering cardiac arrest in an ICU have the advantage of immediate response, but the disadvantage of being sick enough to be in an ICU already. A national database review in AJRCCM 2010 showed that the likelihood of surviving to discharge after a first cardiac arrest in an ICU was 16% overall, but this was highly dependent on 1) whether the rhythm was shockable, i.e., VT/VF, and 2) whether the patient was already on vasopressors:
- 41% with VT/VF, not initially on pressors, survived to discharge, with more than half (21%) going home. These were the best outcomes; being on pressors cut these rates in half.
- Of those with PEA/asystole arrests, 15% survived to discharge if not on pressors at time of arrest. Only 6% survived if on pressors; only 1.7% had a favorable neurologic outcome at the time of discharge.
That paper may be among the most important ever published in critical care, in my humble opinion. However, its major limitation was a follow-up limited to hospital discharge. Mortality and functional status after discharge were not reported, but are critical for understanding and communicating the real-world prognosis for a patient after in-ICU cardiac arrest. (I.E., how many deaths in LTACs at 45 days are there, and how many people recover from cognitive impairment and physical debility at discharge to semi-independent living at 6 months?)
Kutsogiannis et al add a piece to that epidemiologic puzzle. They retrospectively reviewed the short- and long-term survival of 517 patients resuscitated after a cardiac arrest (and including ~11 patients with a primary respiratory arrest). Importantly, only 38% of their patients were in a general MICU or SICU, while 62% were in a coronary care unit or cardiovascular surgery ICU. Patients were a median 70 years old.
Results: Their survival statistics were significantly better than those reported by Tian et al above.
- 59% survived the arrest;
- 27% survived to hospital discharge;
- 24% were alive at one year;
- 16% were alive at 5 years.
These results varied widely by the type of arrest and location of ICU. They had a high proportion of shockable and paceable rhythms: 34% had ventricular tachycardia or fibrillation, 18% had SVT, bradycardia or AV block. Only 46% had PEA/asystole. (2% had respiratory arrest.)
36% of patients with the electrically-modifiable arrests (VT/VF, SVT, bradycardia, block) were alive at one year.
Only 11% of those with PEA or asystole were alive at one year.
The outcomes were just as starkly different when considering the location of arrest:
- 54% of patients arresting in a cardiovascular ICU survived to hospital discharge;
- 29% of those arresting in a coronary care unit survived to discharge;
- 19% of those arresting in a general ICU survived to hospital discharge.
They did not stratify by whether the patient was on pressors at the time of arrest. APACHE II score was not predictive unless it was over 30. As authors point out, these survival rates after in-ICU PEA/asystole cardiac arrest are about the same as those reported over the past 20 years. Unfortunately, we don't know the surviving patients' neurologic or functional outcomes, only that they were alive.
What we need to know now is, how "alive" are these poor souls at one and 5 years after cardiac arrest? What percentage are living independently or with minimal assistance, and how many require 24-hour care? With that information, stratified by type of arrest, we could provide meaningful, accurate information to distressed family members as they seek to choose the right treatment course for their loved one while struggling with their own painful emotions.
Kutsogiannis DJ et al. Predictors of survival after cardiac or respiratory arrest in critical care units. CMAJ 2011;183(14):1589-1595.