Implementation of a POst-Arrest CAre Protocol - Sunde et al.
At the time of this study there was on-going recognition that the outcomes after cardiac arrest were persistently bad. However, there were signs that certain interventions might actually be able to help with this bleak picture e.g. some promising RCTs on therapeutic hypothermia (See here and here). The authors set out to see if an organised care bundle could help bring these interventions together to improve the poor outcomes they were seeing in their own patients.
What did they do?
Design: An observational cohort study, before and after implementation of a package of care for out of hospital cardiac arrest (OOHCA) patients.
Control: Routine care before the implementation of the care package. Patients managed in one of 5 different ICUs in a single large teaching hospital. No protocol for management.
Intervention: Implementation of a standard treatment protocol. This comprised of setting standards for; optimising haemodynamics and ventilator status, early reperfusion therapy, therapeutic hypothermia, blood glucose control, treatment of seizures, heart rate control, and other parameters. Patients managed in one of 2 specific ICU at the same large teaching hospital.
Population: All patients with ROSC following OOHCA admitted to a large teaching hospital in Oslo, Norway.
Primary Outcome: Survival to hospital discharge with a favourable neurological outcome (defined as a cerebral performance category score of 1 or 2)
Control: Routine care before the implementation of the care package. Patients managed in one of 5 different ICUs in a single large teaching hospital. No protocol for management.
Intervention: Implementation of a standard treatment protocol. This comprised of setting standards for; optimising haemodynamics and ventilator status, early reperfusion therapy, therapeutic hypothermia, blood glucose control, treatment of seizures, heart rate control, and other parameters. Patients managed in one of 2 specific ICU at the same large teaching hospital.
Population: All patients with ROSC following OOHCA admitted to a large teaching hospital in Oslo, Norway.
Primary Outcome: Survival to hospital discharge with a favourable neurological outcome (defined as a cerebral performance category score of 1 or 2)
What did they find?
Numbers: 61 patients recruited in the 20.5 month intervention period. 58 patients recruited from the 24 month control period.
Results: Significantly more patients survived with a good outcome in the intervention cohort (34 of 61 (54%)) compared with the control cohort (15 of 58 (26%)), P <0.001.
All patients with a good neurological outcome survived to 1 year.
Mean age was lower in the intervention cohort (63 +/- 14) compared with the control cohort (68 +/- 12), p = 0.008, and with a greater proportion below the age of 70 year old. Other baseline variables were not significantly different.
The intervention group had some significantly different physiology parameters during the initial part of their ICU stay.
The blood glucose was lower in the intervention group at 12 and 24 hours (mean 8.0 and 6.4 mmol/l) compared with the control group (mean 9.7 and 7.5 mmol/L), p = 0.033 and 0.028 respectively.
The temperature was lower in the intervention group at 12 and 24 hours (mean 33.9 and 34.0 oC) compared with the control group (mean 37.5 and 38.4 oC), p < 0.001 for both times.
The heart was lower in the intervention group at 12 and 24 hours (mean 67 and 68 bpm) compared with the control group (mean 85 and 87 bpm), p < 0.001 for both times.
There was a significantly higher rate of reperfusion treatment in the intervention cohort (49%) than in the control cohort (3%), p<0.001.
Results: Significantly more patients survived with a good outcome in the intervention cohort (34 of 61 (54%)) compared with the control cohort (15 of 58 (26%)), P <0.001.
All patients with a good neurological outcome survived to 1 year.
Mean age was lower in the intervention cohort (63 +/- 14) compared with the control cohort (68 +/- 12), p = 0.008, and with a greater proportion below the age of 70 year old. Other baseline variables were not significantly different.
The intervention group had some significantly different physiology parameters during the initial part of their ICU stay.
The blood glucose was lower in the intervention group at 12 and 24 hours (mean 8.0 and 6.4 mmol/l) compared with the control group (mean 9.7 and 7.5 mmol/L), p = 0.033 and 0.028 respectively.
The temperature was lower in the intervention group at 12 and 24 hours (mean 33.9 and 34.0 oC) compared with the control group (mean 37.5 and 38.4 oC), p < 0.001 for both times.
The heart was lower in the intervention group at 12 and 24 hours (mean 67 and 68 bpm) compared with the control group (mean 85 and 87 bpm), p < 0.001 for both times.
There was a significantly higher rate of reperfusion treatment in the intervention cohort (49%) than in the control cohort (3%), p<0.001.
Is it any good?
Overall: Mixed – methodologically poor but powerful results.
Strengths: A fairly pragmatic study looking at bundle of care implementation. Clinically significant outcomes. The aims of the protocol seem to have been achieved i.e. the physiologic targets.
Weaknesses: Multiple confounding factors, including changes to pre-hospital care and the younger age of patients in the intervention cohort. High potential for Hawthorne effect. No blinding in outcome assessment.
Strengths: A fairly pragmatic study looking at bundle of care implementation. Clinically significant outcomes. The aims of the protocol seem to have been achieved i.e. the physiologic targets.
Weaknesses: Multiple confounding factors, including changes to pre-hospital care and the younger age of patients in the intervention cohort. High potential for Hawthorne effect. No blinding in outcome assessment.
Final THoughts
Overall this is a very interesting paper that demonstrates a clear improvement in outcome for these patients. Though the quality is poor in terms of the number of confounding factors and risk of bias, this is perhaps not that important in terms of the message that this paper gives, as I think they are unlikely to be solely responsible for the effects seen. It would suggest that if we are aggressive and interventional with our patients then we can improve their outcomes regardless of where exactly this improvement comes from.
Written: Tom Heaton
Reviewed: Not done
11th May 2016
Reviewed: Not done
11th May 2016
References
- K Sunde et al. Implementation of a standardised treatment protocol for post resuscitation care after out-of-hospital cardiac arrest. Resuscitation. 2007. 73:29-39.
