Gasping during refractory out-of-hospital cardiac arrest is a prognostic marker for favourable neurological outcome following extracorporeal cardiopulmonary resuscitation: a retrospective study

Out-of-hospital cardiac arrest (OHCA) is a major public health issue in developed countries, with approximately 120,000 OHCA events occurring in Japan annually. Patient survival after OHCA remains low [1, 2]. Moreover, among survivors, only 3–10% achieve a favourable neurological outcome [2‐5], likely due to insufficient cerebral blood flow. Although minimised interruption of chest compressions has been emphasised to maintain the blood flow, chest compressions alone are inadequate to ensure sufficient cerebral blood flow [6]; accordingly, prolonged cardiopulmonary resuscitation (CPR) without the return of spontaneous circulation (ROSC) results in an unfavourable outcome following cardiac arrest. To restore cerebral blood flow, resuscitation using extracorporeal membrane oxygenation (ECMO) was proposed in the 1960s [7, 8]. Several recent studies have suggested that better neurological outcomes are achieved after refractory cardiac arrest in patients who receive extracorporeal cardiopulmonary resuscitation (ECPR) [9‐11].

ECPR is a recently adopted approach supported by many studies for refractory cardiac arrest; however, established inclusion criteria remain unclear. Some characteristics have been associated with good neurological outcomes in the setting of ECPR; these include presence of an initial shockable cardiac rhythm (ventricular fibrillation [VF] and pulseless ventricular tachycardia [VT]), witnessed cardiac arrest, presence of bystander CPR attempt, shorter duration of no-flow and low-flow time, cardiac aetiology, higher arterial pH, and initial lower serum lactate concentration [12‐17]. Despite these known variables, only 11–15% of patients achieve a favourable neurological outcome using ECPR following an OHCA [10, 18]. One indicator of a favourable neurological outcome after OHCA is gasping during cardiac arrest [19]. Furthermore, gasping is associated with witnessed events and VF [20]. These findings suggest that gasping during cardiac arrest may represent an early phase of the arrest.

However, it is still unknown whether gasping is a neurological prognostic factor for refractory cardiac arrest patients who fail to achieve ROSC by the usual resuscitation methods. The primary aim of this study was to examine whether gasping by the patient during transportation by emergency medical services (EMS) would be independently associated with a favourable neurological outcome for patients undergoing ECPR.

The present study conformed to the principles outlined in the Declaration of Helsinki and was performed with the approval of the institutional ethical committee of Sapporo Medical University (Number: 312-1118). Informed consent requirement was waived because of the life-threatening situation and the absence of a therapeutic alternative. Information was delivered to the patients’ relatives after inclusion, as appropriate, in a life-threatening context.

A total of 166 ECPR patients were included in the study. During transportation by EMS, 38 patients exhibited gasping, and 128 patients did not exhibit gasping (Fig. 1). Table 1 shows the patients’ characteristics based on the neurological outcome. Thirty-five patients had a favourable neurological outcome, and 131 patients had an unfavourable neurological outcome. Among patients who achieved a favourable neurological outcome, 20 exhibited gasping during EMS transportation (20/35; 57.1%). The patients in the favourable neurological outcome group were younger than those in the unfavourable neurological outcome group (Table 1). Compared with the unfavourable neurological outcome group, the favourable neurological outcome group had a higher rate of bystander CPR attempt, gasping during transportation, and achieving ROSC at admission. Compared with the unfavourable neurological outcome group, the favourable neurological outcome group had a lower dosage of adrenaline administered both during EMS transportation and in hospital, as well as a shorter duration of no-flow time. The sensitivity, specificity, positive predictive value, and negative predictive value of gasping for achieving a favourable neurological outcome were 57.1%, 86.3%, 52.6%, and 88.3%, respectively. Table 2 shows the results of the logistic regression analysis according to neurological outcome. Multivariate logistic regression analysis showed that gasping during transportation by EMS was independently associated with a favourable neurological outcome (odds ratio [OR]: 9.94; 95% confidence interval [CI] 5.49–18.00) after adjusting for initial cardiac rhythm, low-flow time, and serum lactate value.

The present study shows that gasping in patients who experienced OHCA during EMS transportation was independently associated with favourable neurological outcome in patients who underwent ECPR. The underlying reason for the relationship between gasping and favourable neurological outcome was suggested by Debaty et al., who remarked that gasping may be an important natural biomarker for the presence of the brainstem activity during CPR [19]. Accordingly, gasping during CPR directly and partially reflects the brainstem activity. In contrast, other reported prognostic factors do not directly reflect the brain activity [12‐17].

In the present study, bystander CPR and shorter duration of no-flow time were also associated with the presence of gasping during EMS transportation. A previous study suggested that in the absence of adequate CPR, gasping was short-lived and rapidly declined [22]. Preserved gasping is considered to result from at least some level of maintenance of cerebral blood flow due to chest compressions. Thus, preserved gasping may indirectly suggest that there was only a short duration during which no adequate CPR was administered. Therefore, gasping directly indicates that the brain is partially functioning and indirectly indicates a short duration without adequate CPR. This may explain the association between preserved gasping and a favourable neurological outcome in patients after ECPR.

Additionally, gasping by the patient during EMS transportation is a prognostic factor for favourable neurological outcome; therefore, this important information can be used to prepare ECPR for refractory cardiac arrest. Accordingly, during transportation to the hospital, we propose that EMS had better methods for transmitting information regarding the patients gasping during cardiac arrest. As previously indicated, there is no properly consolidated system regarding the recording, tracking, and transfer of information such as the existence of gasping during cardiac arrest [19]. Therefore, we think that it is better to include information regarding gasping in a manner that reliably delivers such information, similar to that used for initial cardiac rhythms, cardiac arrest witnessed, and the presence of bystander CPR. Such information may benefit potential patients owing to ECPR implementation. We recommend that hospitals that receive patients in refractory cardiac arrest establish such a system to prepare and better utilise ECPR.

Low-flow time, which has been associated with prognosis in many studies [14, 17], was not associated with neurological outcomes in this study. It has been shown that a shorter low-flow time increases the likelihood of favourable neurological outcomes, and awareness of resuscitation time is certainly important for EMS personnel and emergency physicians. In our facility, since we regard the presence of gasping to be associated with favourable neurological outcomes, when determining whether to use ECPR, the low-flow time may have been underestimated in relation to the presence of gasping. Therefore, this may have resulted in a biased conclusion that low-flow time was not associated with neurological outcomes in this study. Awareness of the significance of both gasping and low-flow time is important for resuscitation.

There were several limitations to this study. First, this was a single-centre, retrospective study with a relatively small sample size, which might have led to statistical error by allowing unrecognised bias. In particular, the small sample size limited the number of covariates that could be entered into the multivariate analysis. In addition, although we used IPTW by PS, since gasping is associated with other variables (such as no flow time) as well as the outcome, the current statistical analysis is insufficient to fully demonstrate the implications of gasping. Additionally, as the standardized differences of some variables were not below 0.1, the result of the IPTW was also insufficient from this point of view. Therefore, the purpose of the IPTW was to support the results of multivariate analysis. Further analysis that considers other confounding factors is necessary. Second, we used data only from patients who underwent ECPR. Although our facility has an ECPR database, it does not cover the entire OHCA. For this reason, the frequency of patients who maintained gasping but in whom ECPR was not introduced was unknown. This is a major limitation in recommending ECPR for refractory cardiac arrest with gasping. For instance, some patients who did not undergo ECPR might have been suitable candidates for ECPR. Furthermore, we were unable to assess whether patients who had preserved gasping during resuscitation but who did not undergo ECPR could have achieved a favourable neurological outcome by undergoing ECPR. Since the frequency of cardiac arrest patients who maintained gasping but in whom ECPR was not introduced was unknown, the algorithm for introducing ECPR by the presence or absence of gasping could not be reported. Third, we evaluated the outcomes at the time of hospital discharge, and long-term outcomes were not evaluated. Fourth, depending on the attending physician, ECPR was often introduced if gasping was preserved. We have experienced several ECPR cases with a favourable neurological outcome in patients with gasping preserved during cardiac arrest. Based on the experiences, for example, ECPR was sometimes introduced in cases, where gasping was preserved, even if the initial cardiac rhythm was non-shockable or low-flow time was quite long. This may be a selection bias in the study.

The present study shows that gasping by the patient during transportation by EMS was independently associated with a favourable neurological outcome in patients undergoing ECPR. Since gasping has the potential to confer a favourable neurological outcome in patients experiencing refractory cardiac arrest, it is important to raise awareness of the potential importance of gasping by informing EMS personnel and emergency physicians on the positive implications of gasping so that they do not abandon resuscitation efforts in such instances. In these cases, the use of ECPR can be considered.