Effects of the Cardio First Angel TM on chest compression performance

1. Introduction

Out-of-hospital cardiac arrest affects 20.9–186.0 patients/100.000/year and survival rates remain poor [1, 2, 3]. Global morbidity as well as mortality are immense and socioeconomic impact is high [2, 4]. As first response measures, established guidelines and current analyses highlight the importance of recognizing cardiac arrest early, promptly placing the emergency call, performing cardiopulmonary resuscitation (CPR) and defibrillation [1, 5, 6]. Even though randomized trials in this specific setting are hardly feasible and corresponding data are thus lacking, studies indicate that bystander CPR performed before the arrival of emergency medical services (EMS) is associated with profoundly increased survival rates and similarly, associations between the time from collapse to start of CPR and survival have been emphasized [1, 5].

Whilst the role of bystanders is thus widely deemed crucial, the knowledge, skills and confidence of the latter will vary according to the circumstances, level of training and prior experience [6]. Furthermore, even though CPR trainings are offered widely spread, performance especially of laymen deteriorates soon after training [7]. Adjunct devices assisting and guiding bystanders could potentially improve CPR performance. The Cardio First Angel TM (Cardio First Angel UG, Munich, Germany/INOTECH, Nubberg, Germany) device was recently shown to be associated with beneficial clinical results when being used in an intensive care unit setting [8]. Here, we focus on laymen and examine performance in chest compression comparing usage of the Cardio First Angel TM device to standard basic life support (BLS).

OPEN IN VIEWER

Altogether, 294 subjects participated in the study. Due to organizational difficulties, n = 46 medical students did not answer the questionnaire (i.e., n = 82 with completed questionnaire). Demographic data are depicted in Table 1. Table 2 provides detailed results of the remaining items asked for in the questionnaire. General willingness to perform CPR was high throughout all groups and by far exceeded 90%. The majority of laymen (without respectively with explanation of the device) ranked their own resuscitation skills as being on an intermediate level (76.8% and 81.3% within the range from 3 to 7, respectively). The vast majority of subjects had participated in a first aid course before (81.4% and 90.6% of laymen, 98.8% of medical students and 92.6% of medical staff, respectively). Interestingly, a substantial proportion of all groups expressed fears of contact. This included medical professionals, even though the percentage in this group was only 18.5% and thus lower than the percentages within the other groups all of which exceeded one third.

OPEN IN VIEWER

Performing conventional chest compression according to the individual state of knowledge of the participant in comparison to using the Cardio First Angel TM did not show any significant differences regarding correctness of compression frequency for all groups. Of note, proportions of participants with correct compression frequency were low throughout all groups. Medical students represented the only group with more than 50% of the subjects achieving correct compression frequencies within the reference range of 100–120/min. Compression frequency results of all groups during conventional chest compression and when using the Cardio First Angel TM are illustrated in Fig. 3a. Both, laymen without and with explanation of the device showed a significantly better success rate regarding correct compression depth when using the Cardio First Angel TM device (23.3% vs. 55.8%, p = 0.004 and 25.0% vs. 52.1%, 0.001, respectively). Similarly, positive effects were observed for medical students (22.7% vs. 42.2%, p = 0.004) while no effect was seen in medical staff ( p = 0.39). For detailed results of compression depth see Table 3 and Fig. 3b. Neither age nor sex were significantly associated with the effect of using the Cardio First Angel TM device versus conventional chest compression ( $p>$ 0.05 for both, frequency and depth, Wald test for interaction effect in GEE). Furthermore, the benefit of using the Cardio First Angel TM device did not differ significantly in laymen with or without explanation. The improvement yielded by usage of the Cardio First Angel TM versus conventional chest compression regarding frequency was significantly higher for probands with fears of contact in general and for probands with fear of physical contact/blood/body fluids, respectively ( p = 0.02 and p = 0.03; Wald test for interaction effect in GEE model with adjustment for group). All other test results concerning fears were non-significant.

Even though rates of correct hand positioning were high throughout all groups during conventional chest compression (ranging from 76.7% to 95.3%), when using the Cardio First Angel TM correctness of position was 100% in all groups, i.e. no mistakes in hand positioning occurred when using the device. Detailed data on chest compression performance are given in Table 3.

A broad majority of the participants stated that they would use a device that assists regarding compression strength, frequency and locating the correct position (83.7–95.8% for the different groups, respectively). Approval of the Cardio First Angel TM device in particular was likewise high among laymen and medical staff (81.5–97.7%). 69.5% of the medical students stated that they would use the device. Especially laymen would resuscitate rather using the device than conventionally performing CPR by hand only (83.7% and 72.9%, respectively, Table 2).

In the chain of survival, early recognition and intervention in cardiac arrest are of uppermost importance. In case of out-of-hospital cardiac arrest, essential bridging until the arrival of EMS most frequently has to be performed by laymen who thus potentially influence survival, prognosis and outcome of the patient [1, 5, 6, 9]. Effective chest compressions are a key element of successful CPR. Recently, their importance has been further highlighted and changes in guidelines included a focus on chest compression with more as well as deeper compressions and initialization of CPR with chest compression instead of ventilation [6, 8]. Even though evidence available is not sufficient to change current practice and to generally promote chest compression-only CPR, according to the ERC Guidelines for Resuscitation 2015 in endorsement of a recommendation of the International Liaison Committee on Resuscitation (ILCOR), all CPR providers should perform chest compressions for all victims in cardiac arrest, CPR providers trained and able to perform rescue breaths should combine chest compressions and rescue breaths [6].

In our analysis, general willingness of laymen to perform CPR was high, but more than one third of the participants expressed fears of contact. Analysis of chest compression parameters revealed that the success rates of laymen regarding compression depth and frequency were extremely low when conventional CPR was performed. To a lesser extent, the latter was also true for medical students and staff. Quality of CPR is essential to improve outcome [6]. Too slow compression results in an insufficient circulatory effect and in inadequate perfusion whilst too high compression rates decrease cardiac output because of impaired venous return aside from potentially impairing coronary perfusion. While both, too deep and too shallow compressions likewise impact arterial pressure and seem to be associated with poorer outcomes, wrong hand positioning or inaccurate compression depth and force can result in injuries such as fractures, organ lacerations or pneumothorax with associated mortality and morbidity [5, 10, 11, 12]. When using the Cardio First Angel TM device, a remarkable 100% of correct hand positioning was achieved throughout all groups. Additionally, especially laymen showed a significantly improved and approximately doubled success rate of correct compression depth when using the device. Besides these direct effects on chest compression performance, the device might additionally aid to overcome fears. The vast majority of participants stated that they would use an assisting device and a substantial number had expressed fears of contact. Improvement in compression frequency when using the Cardio First Angel TM was significantly higher for probands with fears of contact in general and for probands with fear of physical contact/blood/body fluids. Besides the actual fear, translation of skills from training environments to a real life situation which is stressful, mentally demanding and potentially disorganized imposes additional obstacles [11, 13]. Holmberg et al. reported that bystander CPR was attempted in only 36% of patients suffering out-of-hospital cardiac arrest and Wissenberg et al. reported a rate of 21.1% which increased to 44.9% after national initiatives were taken to improve cardiac arrest management [5, 9]. Potential indirect effects could include increasing the rate of performed bystander CPR via overcoming fears, raising the attention towards CPR as well as shortening the time interval from collapse to starting resuscitation and thus minimizing no-flow and ischemic time. Additional studies will be needed to undermine these hypotheses. The device seems to be self-explanatory even when used by laymen as we saw no differences in its beneficial effect achieved in laymen with versus without explanation.

Besides clear beneficial effects in laymen, we saw significantly more chest compressions performed with correct depth in medical students when using the Cardio First Angel TM . Regarding compression frequency and correctness of hand positioning, medical students showed a high success rate even without the device which was most likely due to recent extensive CPR training. Since especially correctness of compression frequency by far outnumbered the rates of the other groups, this clearly underscores the importance of CPR training.

Wik et al. analyzed the quality of out-of-hospital real life CPR performed by ambulance personnel, as measured by adherence to CPR guidelines. They found that chest compressions were not delivered in a substantial proportion of time and that most compressions were too shallow. All involved personnel had undergone a refresher course in ALS prior to the study period [11]. Likewise, Abella et al. analyzed the quality of CPR during in-hospital cardiac arrest and found that, even though performed by well-trained hospital staff, the quality of multiple parameters of CPR was inconsistent and often did not meet guideline recommendations [13]. This is of particular interest since only a few studies examined CPR performance in real life situations and it demonstrates that even resuscitation as performed by professionals is far from ideal. The findings underline the importance of high-quality and high-frequency recertification of medical staff in BLS and ALS. In our analysis, medical staff likewise showed clear shortages in chest compression performance. Whilst correctness of hand positioning was 92.6% without and 100% with the CFA, no significant differences were observed for compression depth and frequency. Vahedian-Azimi et al. performed a randomized controlled clinical trial investigating standard manual CPR versus CPR using the Cardio First Angel TM in patients with cardiac arrest in mixed medical-surgical intensive care units (ICU) of academic teaching hospitals. Before the start of the study, all ICU nurses (who performed the chest compressions) received standardized CPR training in addition to formal training with the Cardio First Angel TM device. Even though, for example, absolute compression depth and frequency are not provided, adherence to CPR guidelines as well as CPR quality as determined by dedicated scores were significantly improved in the intervention group and return of spontaneous circulation (ROSC) was observed significantly more often. Furthermore, a decrease in rib fractures was reported [8]. A direct comparison of the results of Vahedian-Azimi et al. with those of our analysis is not feasible since first, the data were obtained in a training setting versus in real-life situations and second, they used a more comprehensive model for the analysis of CPR performance. However, it may be assumed that also medical professionals could benefit from assisting devices especially in real-life situations, either via direct or via indirect effects, and further analyses are clearly warranted.

A number of devices to assist in CPR and to ensure consistency and quality of chest compressions have been proposed [8]. In general, they comprise feedback and/or prompting functions and intend to improve CPR quality with the aim to increase ROSC and ultimately survival. Proposed forms of feedback include voice prompts, metronomes, visual dials, numerical displays, waveforms, verbal prompts and visual alarms, i.e. technology ranges from simple metronomes to more complex devices that monitor and provide real-time audiovisual feedback. They may or may not be associated with automated external defibrillators (AEDs) and even smartphone applications have been developed [6, 8, 14]. A number of devices that are positioned between the resuscitator’s hand and the patient’s chest, comparable to the Cardio First Angel TM , have been tested in simulation settings. Several studies reported beneficial results for various devices including the CPR-plus TM (Kelly Medical Products, Inc., Princeton, USA), CPREzy TM (Health Affairs, London, UK) and the Laerdal CPRmeter TM (Laerdal, Stavanger, Norway) [15, 16, 17, 18, 19]. Contrastingly, Zapletal et al. investigated the efficiency of chest compressions comparing three CPR feedback devices (Zoll PocketCPR ® (Zoll Medical, Chelmsford, USA), Laerdal CPRmeter TM , the iPhone app Zoll PocketCPR ® (Zoll Medical, Chelmsford, USA)) and standard BLS in an open, prospective, randomized controlled trial in a simulated setting with a testing period of 8 minutes. Even though there were several differences between the feedback devices and standard BLS, none of the devices was able to achieve an improvement in compound parameters for chest compression quality compared to standard BLS [14].

Most of the devices depend on sufficient battery capacities. The Cardio First Angel TM provides a mechanically generated feedback of compression depth and release via springs. It does thus not depend on battery capacities or algorithms to determine compression depth. Furthermore, the auditory feedback provided as soon as sufficient compression and decompression have been achieved does not require focusing on potentially small displays or lights [8]. Another concern raised by Zapletal et al. is that usage of an adjunct device may delay starting CPR [14]. Whether this proves to be significant in a real-life setting and if so, whether this might be counterbalanced by an overall increased rate of attempted CPR due to secondary effects of the devices, remains to be demonstrated. Altogether, current guidelines do not recommend any assisting device since clear evidence consistently voting for their usage is lacking. Several studies report conflicting data, parts of the discrepancies may arise from using single key parameters such as compression depth, frequency and position versus more comprehensive models. It still has to be determined which approach will be more suitable for analysis of CPR and especially which will prove better correlation to performance in real-life situations and to relevant effects such as improving ROSC and survival. Additional factors limiting comparability include the test setting, the tested cohorts and differences in the tested devices themselves. Altogether, there seems to be a fundamental need for guidance and feedback during bystander CPR as otherwise there is no option for the resuscitator to evaluate whether chest compressions are satisfactory. Clear indicators that adjunct devices may have beneficial effects justify further structured analyses to eventually provide guidelines on their usage.

5. Limitations

Main shortages of our analysis include that it did not test for decompression and pauses or interruptions in chest compression. Furthermore, the analyzed period was rather short not representing the delay to be expected until EMS arrive. We did not analyze rescuer fatigue and differences in fatigue with and without the device. Additionally, due to the crossover design of the study, a potential individual learning effect of uncertain magnitude cannot be ruled out. Only one group received an explanation of the device which might have led to hypervigilance serving as a confounder in this group. Further considerations related to the device itself that have been raised include that it does not account for more complex aspects such as compressibility of the surface on which the patient is lying and changes in chest wall compliance and elasticity during CPR [8]. Additionally, this was a manikin-study under simulation conditions and results may not necessarily be extrapolated to real-life scenarios. Clearly, further studies are warranted.

6. Conclusion

Bystander CPR seems to be of utmost importance but human factors and human error will influence its quality in both laymen and healthcare professionals. Providing guidance and support via adjunct devices has been proposed. With growing evidence that chest compressions are becoming an increasingly important part of CPR, the Cardio First Angel TM as a new and totally mechanical device has been developed to increase chest compression quality, efficiency and consistency. In our analysis, correctness of manual chest compression in laymen was significantly improved with regard to compression depth when using the device. Furthermore, with the device, no cases of incorrect hand positioning occurred. Additional and especially clinical studies are required to assess the relevance and potential benefit of assisting devices.