Rapid Infusion of Cold Normal Saline During CPR for Patients With Out-of-hospital Cardiac Arrest

INTRODUCTION:

Pre-hospital cardiac arrest is common and associated with a poor prognosis and only about 8% of patients have a good outcome. For patients who are initially successfully resuscitated by ambulance paramedics and transported to hospital, there is considerable mortality and morbidity. This is largely due to the anoxic brain injury sustained during the cardiac arrest.

One current therapy for the severe anoxic brain injury following out-of-hospital cardiac arrest is therapeutic hypothermia (TH) induced after resuscitation. This treatment was shown to improve outcomes in two clinical trials. The International Liaison Committee on Resuscitation now recommend TH (33°C for 12-24 hours) for patients who remain comatose after resuscitation from cardiac arrest.

However, the optimal timing of TH is still uncertain. The above clinical studies used surface cooling and this delayed TH until after arrival at the hospital. On the other hand, laboratory data has suggested that there is significantly decreased neurological injury if cooling is initiated during CPR.

The current ideal technique for induction of TH during CPR is a rapid intravenous infusion of a large volume of ice-cold fluid. This technique has become established as the cooling method of choice in the pre-hospital setting, the Emergency Department (14) and the Intensive Care Unit.

Previously, we conducted a randomised, controlled trial of paramedic cooling after CPR compared with cooling in the emergency department (ED) (NHMRC study number 236879). This study enrolled 396 patients between 2005 and late 2007. There were 234 patients with an initial cardiac rhythm of ventricular fibrillation (VF) and 163 patients with an initial cardiac rhythm of non-VF (asystole or pulseless electrical activity).

In the post-VF arrest patients, there was a good outcome (discharge to home or to rehabilitation) in both groups of 50%. In the post-non-VF arrest patients, there was a good outcome at hospital discharge of 12% in the paramedic cooled group compared with 9% in the standard care group. The study was stopped at the interim analysis due to a lack of difference in the primary outcome measure (outcome at hospital discharge) between the two groups (futility).

Further analysis of this data revealed that paramedics infused up to 1000mL ambient temperature fluid during CPR prior to enrolment as part of standard paramedic treatment. In addition, the rapid infusion of cold fluid was commenced en-route to hospital. Thus, cooling commenced approximately 30 minutes after paramedic arrival and only just prior to hospital cooling. Although there was a decrease in the core temperatures of the patients allocated to pre-hospital cooling on arrival at the ED, this was a transient effect lasting only approximately 20 minutes. Subsequently, the cooling curves of the patients in both groups were identical. Thus, it was considered unlikely that this transient difference in core temperature could have a measurable effect on outcomes.

Laboratory data suggests that a rapid intravenous infusion of cold fluid during CPR effectively decreases core temperature. Nordmark et al. studied the induction of hypothermia with a large volume of intravenous ice-cold fluid after cardiac arrest during CPR in anaesthetised piglets who were subjected to eight minutes of VF. The mean temperature reduction was 1.6°C in the hypothermic group and 1.1°C in the control group (p=0.009).

There is also laboratory data that cooling during CPR may increase the rate of successful defibrillation. Boddicker et al. examined the success rates of defibrillation in swine cooled to different temperatures. After 8 minutes of VF (with no CPR), the animals were defibrillated with successive shocks as needed and underwent CPR until resumption of spontaneous circulation or no response. First-shock defibrillation success was highest in the hypothermia (33°C) group (6/8 hypothermia versus 1/8 normothermia; P=0.04). None of the 8 animals in the normothermia group achieved resumption of spontaneous circulation compared with 7/8 moderate hypothermia (P=0.001). Coronary perfusion pressure during CPR was not different between the groups, thus this beneficial effect of hypothermia was not due to alteration of coronary perfusion pressure but likely due to changes in the electrophysiological properties of the myocardium. Thus, it appears that mild hypothermia may have a beneficial anti-arrhythmic effect, as well as a neuroprotective effect.

More recently, pilot clinical trials have been undertaken in Europe that suggest that cooling during CPR by paramedics is feasible. For example, Bruel et al studied the feasibility and safety of a rapid infusion of 2000mL of normal saline at 4°C during CPR. A total of 33 patients were included in the study of whom eight patients presented with VF as the initial cardiac rhythm. After intravenous cooling, the temperature in the patients decreased by a mean of 2.1°C.

In a similar pilot study, Kämäräinen et al. cooled seventeen adult patients with out-of-hospital cardiac arrest during CPR. A return of circulation was achieved in 13 patients (76%). The temperature of the patients on hospital admission was a mean of 33.8°C and the mean infused volume of cold fluid was 1571mL. The authors concluded that induction of therapeutic hypothermia during prehospital CPR was feasible and apparently well tolerated.

More recent data specifically examining respiratory function in 52 patients treated with large volume, ice cold saline has indicated that there is no adverse effect on respiratory function.

Given this supportive laboratory and preliminary clinical data, and our previous experience in running a large pre-hospital trial in cardiac arrest patients, we propose to conduct a definitive randomized, controlled trial of paramedic cooling during CPR compared with standard care including cooling after arrival at the hospital.

In the treatment arm, paramedics will undertake immediate cooling during cardiac arrest, using a large volume (20mL/kg, followed by 10mL/kg) intravenous bolus of ice-cold saline. This strategy will overcome the delay in treatment that was found in our previous study. Thus, cooling will commence significantly earlier, possibly resulting in significantly improved rates of resuscitation and better neurological outcomes.

In the control arm, patients will be resuscitated using current protocols and be cooled after arrival at the hospital (the current standard of care).

The trial will run as two parallel clinical trials because of the marked difference in outcomes between patients with VF as the initial cardiac rhythm and patients with non-VF as the initial cardiac rhythm.

STUDY DETAILS

HYPOTHESIS:

That paramedic core cooling during CPR using a rapid infusion of ice-cold (4 degrees C) large volume (30mL/kg total) normal saline improves outcome at hospital discharge compared with standard care in patients with out-of-hospital cardiac arrest.