Study design and setting
The single-center, prospective, non-randomized and observational study was conducted at a prehospital area that is covered by the Emergency Unit of Community Health Centre Ljubljana and the Rescue station of University Medical Centre Ljubljana during September 2019 and May 2022. The prehospital area has 1670 km2 and provides emergency services for around 450.000 inhabitants and additionally over 60.000 daily working migrants.
The research protocol received approval by Slovenian Medical Ethics Committee (No. 0120-334/2019/3); patients’ consent was waived because of the observational nature of the study and emergency setting. Study protocol was registered at clinicaltrials.gov (NCT04058925).
All patients with non-traumatic cardiac arrest aged 18 or more were eligible for inclusion. Excluded patient were as follows: age below 18 years, pregnant women, traumatic cardiac arrest, hypothermic patient, drowned patient, patient that was candidate for extracorporeal CPR, patients who had achieved ROSC before the placement of NIRS device probes on the skin and if it was not possible to place NIRS probes on the patient within 5 minutes after start of ALS algorithm.
The team of doctor and two medical rescuers were dispatched by health dispatcher after receiving information of patient not showing signs of life. The doctor led resuscitation according to European Resuscitation Council guidelines for Advanced Life Support (ALS).(20, 21) Immediately upon arrival, the prehospital team started with the ALS algorithm.
Tissue oxygenation measurement
As soon as possible, one of the team members placed NIRS probes on the patient. NIRS device (SenSmart Model X-100, Nonin Medical, Inc. Playmouth, Minnesota, USA), which records rSO2 every 4 seconds, and disposable self-adhesive probes (SenSmart Nonin Medical, Inc. Playmouth, Minnesota, USA) were used. Each probe was marked with color and always placed on the same part of the body: blue probe for brain and yellow for skeletal muscle. The blue probe for measuring brain rSO2 was placed on the patient’s right side of forehead and the yellow probe to the patient’s right hand thenar to measure skeletal muscle rSO2. The NIRS device screen was not covered, so the team members could fix probe position in case of bad contact. As consequence this study was unblinded. However, teams had instructions that measured rSO2 values must not influence decisions made by the resuscitation team regarding termination of resuscitation or continuing one. The measurement stopped when the patient was admitted to the Emergency department or when the doctor declared death of the patient and the CPR was stopped. ROSC was defined as return of spontaneous palpable pulse and or breathing, coughing, movement of the patient, rise of etCO2 for more than 30 second.(22)
After intervention data were downloaded from NIRS device by specially dedicated software (SenSmart, Version 220.127.116.11, Nonin Medical Inc., Minneapolis, MN USA) and paired with information from the intervention protocol. One of the graphs with measurements is presented on Figure 1. Basal rSO2 was defined as average of 4 measurements of brain and skeletal muscle rSO2 after signal stabilization just after NIRS probes placement. We also recorded maximal rSO2 value during the CPR, difference between maximal-minimal rSO2 value (delta-rSO2), value of rSO2 at ROSC or the end of CPR with no-ROSC (end-CPR rSO2).
The following additional data was recorded: basic demographic data (age, gender), the time of call to emergency telephone number (112), the time of arrival of the emergency team on scene, the time of ROSC/time of death, duration of CPR, was cardiac arrest witnessed, were eyewitnesses doing BLS (basic life support), the first ECG rhythm, use of AED, intubation status, number of defibrillations, cumulative dose of used adrenaline, ECG rhythm at the end of CPR/intervention and 28-day survival.
Primary outcome was feasibility of skeletal muscle rSO2 measurement; how demanding is it to apply NIRS probes on two sites, what are the main problems of losing signal, how to fix the probes not to lose the signal.
Secondary outcome was to find out if there are any changes in measured skeletal muscle rSO2 during CPR and before/after ROSC. We also want to test the relationship between basal skeletal muscle rSO2 and time to start ALS in patients with witnessed cardiac arrest and without BLS. We also want to test relationship between skeletal muscle and brain rSO2.
Absolute difference of mean skeletal muscle rSO2=20% (SD 10%) between patients with ROSC and no-ROSC (ratio of sample size 1:3) was estimated in the first 10 recruited patients. For estimated error (Type I. error of 0.05, Type II. error of 0.20) total sample size of 16 patients (4 in ROSC, 12 in no-ROSC group) would be necessary.
The study population was divided into 2 groups according to outcome: ROSC and no-ROSC group. Continuous data were summarized as median (interquartile range) and compared by Mann-Whitney test for independent samples. Non-continuous data were summarized as the count (percentage). Chi-square test was used to compare non-continuous data. Rank correlation with Spearman's coefficient (rho) was used to test relationships between variables. MedCalc® ver. 20.104 (MedCalc Software Ltd) software was used for the statistical analysis. P value < 0.05 was regarded as statistically significant.