Study participants were informed about the general aim of the study but were blinded to the outcome assessments and the specific hypothesis stated. They were instructed to administer CPR with optimum performance. Each participant performed both CPR techniques in a successive order that was randomly assigned. Both techniques were performed for 2 min. Before the actual evaluation, both techniques were practiced for 1–2 min, with a resting period of 4 min between 2 sets.
The manikin (Resusci® Anne, Laerdal Medical, Stavanger, Norway) was connected to a computer running an analysis software (Laerdal PC SkillReporting® software, version 1.0.0). Data were recorded during a 2-min period of the BLS test. The printout of the test period was further manually analyzed by a blinded person to assess the quality of every chest compression and ventilation (a representative printout is shown in Fig. 3). The software measures the quality of the BLS performance (ventilation and chest compressions) and classifies these variables as correct or incorrect, according to the published International Liaison Committee on Resuscitation guidelines (1). The following variables were recorded for each individual ventilation or chest compression: correct ratio of ventilations-to-compressions (2:15); ventilation: correct tidal volume (400–800 mL), correct duration of inflation (1–2 s), and correct inspiratory flow rate (<500 mL/s); and chest compressions: correct depth (38–51 mm), correct rate per min (80–120/min), correct hand position (automatically recorded by device, specified according to published guidelines) (1), and correct release (complete release of pressure during chest compressions). Data were not released to the participants.
Because of the crossover design of this study, normally distributed continuous variables were analyzed with the two-tailed paired t-test. Dichotomous variables from 2 × 2 contingency tables (e.g., correct/incorrect) were analyzed with either the χ2-test (including Yates correction for continuity, where appropriate) or the two-tailed Fisher's exact test. A value of P < 0.05 was considered statistically significant.
Over-the-head CPR was compared with the standard protocol for professional BLS (2) in a manikin simulation study. Sixty-seven EMTs participated in this trial; 62 data sets were complete, and 5 were partially incomplete because of recording problems. The demographic characteristics of the participants were the following: of 67 participants, 53 were men and 14 women; the median age was 29 yr old (range, 18–55 yr); the average height was 180 cm (range, 154–200 cm); and the average weight was 72 kg (range, 50–150 kg).
Over-the-head CPR was superior to standard CPR in the quality of ventilation (Table 1); there were significantly more adequate and correct inflations in the over-the-head CPR group, which was almost exclusively caused by more frequent correctly applied tidal volume and not-too-rapid insufflations. Although not statistically significant, the time required to apply 2 ventilations (one cycle) was on average 0.5 s shorter for the over-the-head CPR group.
The quality of chest compressions did not differ between the two study groups; neither rate, mechanical performance, nor correct 2:15 ratio were different (Table 2). Most of the chest compressions classified as incorrect were caused by too shallow of compression depth and not the wrong hand positioning.
Our simulation study found that over-the-head CPR with bag/mask ventilation is as effective as standard CPR with respect to hemodynamic support but provides superior ventilation. Not only is over-the-head CPR superior in delivering adequate and effective ventilation, but it also provides additional advantages over standard CPR with mouth-to-mouth ventilation.
First, the use of a self-inflating bag enables the rescuer to administer a larger inspiratory oxygen concentration (16% for mouth-to-mouth, as compared to up to 100% for bag-valve masks with demand valve, an example of which is the Life Support Products Demand Valve; Allied Healthcare Products Inc, St. Louis, MO), which allows a reduction of the tidal volume administered, from 700–1000 mL to 500 mL (8). The reduction of tidal volume reduces the risk of stomach inflation (9) and thereby the risk of regurgitation, aspiration, and pneumonia. Second, avoiding mouth-to-mouth ventilation dramatically reduces the risk of contamination from the victim (4), a risk that is the major reason health care workers are reluctant to administer mouth-to-mouth ventilation (10,11). Although the risk of transmission of an infectious disease (human immunodeficiency virus, tuberculosis, etc.) seems to be small, most health care professionals would not give mouth-to-mouth ventilations to an unknown victim (10). Third, using a self-inflating bag instead of giving mouth-to-mouth ventilations is, although never tested, likely to reduce the fatigue of the rescuer, which could, in turn, improve the overall performance of CPR. Although a possible option, we did not test pocket mask ventilation instead of mouth-to-mouth ventilation because the use of pocket masks is not common in most Austrian emergency medical service (EMS) systems.
It could be argued that there must be other options for a single professional rescuer other than regular BLS or over-the-head CPR. Using a self-inflating bag-valve mask from the side of the patient is theoretically possible but is, in our opinion, very ineffective and barely controllable. A second option, changing positions during CPR from the head of the patient to the side, is also possible, but we know from clinical experience that this is impractical and time consuming.
However, there is a third option that has received much attention in recent years: administering continuous chest compressions without ventilating the patient. This method is not recommended in the current BLS guidelines but seems like a suitable alternative, as many health care professionals are very reluctant to give mouth-to-mouth ventilations (10). Furthermore, studies in humans (12) and swine (13,14) have shown no benefit of ventilation during BLS, but an increase in survival in the chest-compression-only technique (15). The authors explain their findings with superior myocardial perfusion during uninterrupted chest compressions compared with standard BLS, spontaneous gasping during CPR, and a larger initial oxygen delivery during the first minutes of CPR.
However, other studies show that ventilation is an indispensable part of successful CPR (8,16,17), in particular, when the airway is obstructed. Moreover, chest-compression-only CPR has been advocated mostly in witnessed cardiac arrests and situations in which the interval to initiation of CPR is short because oxygen tension levels are still high in the cardiac arrest victim. This situation does not represent the usual circumstances when professional health care providers, such as paramedics or prehospital emergency physicians, arrive at the scene. Most of the time, several minutes have passed and the arterial oxygen content of the patient has decreased to low levels, thereby requiring delivery of oxygen via ventilation (18,19). A recent multicenter CPR trial with more than 1000 patients from several German, Austrian, and Swiss EMS reported an average duration of untreated cardiac arrest of approximately 8 minutes (20) compared to 3.1 minutes in the Seattle metropolitan area (12).
For a single professional rescuer, ventilation could be accomplished using the new over-the-head CPR technique, because it allows the rescuer to ventilate a patient with a self-inflating bag from the head of the patient without the need to change positions during CPR and without having to give mouth-to-mouth ventilations, as recommended in the current guidelines for BLS (1). In particular, this might be even more the case in situations with prolonged CPR.
Three other studies have evaluated at the quality of over-the-head CPR (5–7). The first compared over-the-head CPR with standard two-rescuer CPR in an ALS scenario and found that the quality of ventilations and chest compressions did not differ between the two techniques and that the time until administration of the first defibrillation was significantly shorter in the over-the-head CPR group (7). However, because only one rescuer delivered CPR in the over-the-head group compared to two rescuers in the standard CPR group, the number of delivered chest compressions was significantly less in the over-the-head group, which led the authors to conclude that over-the-head CPR is not a suitable alternative for professional two-rescuer CPR (7). The setting of our study was a professional single-rescuer scenario, and therefore, the results of the two trials are complementary rather than comparable.
The two other studies, both small simulation studies in manikins, compared over-the-head CPR with standard CPR in a single-rescuer BLS scenario (5,6). In contrast to our study, which focuses on professional rescuers with experience in bag-valve mask ventilation, in both other trials, the participants had no previous experience with over-the-head CPR and used a pocket mask for ventilation (no bag-valve mask). Handley and Handley (6) focused on trained laypersons performing CPR in a confined space and found that there was no difference between the two methods except for more frequent wrong hand position in the over-the-head CPR group. Importantly, the quality of ventilation was not assessed in this trial. Perkins et al. (5) also found no major difference in quality between over-the-head CPR and standard CPR with the exception that hand positioning during chest compressions was better in the over-the-head group. Both studies conclude that over-the-head CPR might be a suitable alternative technique. However, both studies failed to exploit one of the greatest advantages that over-the-head CPR might offer: the use of a self-inflating bag instead of reliance on the rescuer's own breath to ventilate the patient.
The findings of our study strongly support the notion that the use of a self-inflating bag for ventilation during over-the-head CPR allows the single rescuer to deliver better ventilation than with a pocket mask or mouth-to-mouth ventilation. One possible explanation for the different findings of our study is that in our trial, all participants had previous theoretical and practical experience with over-the-head CPR.
Despite clear results, our study has several limitations. First, this study was performed in manikins, not patients. Although the technique of over-the-head CPR has been used in real patients, there are no published reports on efficacy and outcome. Therefore, the first step was to test this method under well-described circumstances of a CPR simulation comparing BLS performed according to published guidelines and over-the-head CPR. Second, the overall quality of ventilations and chest compressions was poorer than expected, but this might reflect the average performance of professional rescuers in CPR simulation studies, as one study suggests (21). Third, this study obviously compares two different ventilation methods, which is, in theory, a limitation of this study but does, in fact, represent clinical practice and is one of the major reasons this study was performed. In our study, we compared the two techniques during a two-minute CPR cycle, an interval that might be too short compared with real-life cardiac arrest situations. Thus, differences between the two CPR methods that are caused by increasing fatigue during prolonged CPR might be much larger than our study could show.
In conclusion, this study has shown that over-the head CPR may be an efficient alternative to the BLS technique for health care professionals when a single rescuer has to perform CPR. It offers superior ventilation compared with standard BLS but cannot be recommended for laypersons unfamiliar with bag-mask ventilation.
The authors would like to thank Doris Wundsam for her support of this study, Boris Steiner for his help in data analysis, and all members of the study center and participating EMTs from St. John's Ambulance, Vienna, Austria. Furthermore, we are indebted to Professor Volker Wenzel, University of Innsbruck, Austria, for his support in improving the quality of the manuscript.
1. Guidelines 2000 for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care: International Consensus on Science. III. Adult basic life support. Resuscitation 2000;46:29–71.
2. Guidelines 2000 for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care: International Consensus on Science. I. Introduction to the International Guidelines 2000 for CPR and ECC: a consensus on science. Resuscitation 2000;46:3–15.
3. Handley AJ, Monsieurs KG, Bossaert LL. European Resuscitation Council Guidelines 2000 for Adult Basic Life Support. Resuscitation 2001;48:199–205.
4. Mejicano GC, Maki DG. Infections acquired during cardiopulmonary resuscitation: estimating the risk and defining strategies for prevention. Ann Intern Med 1998;129:813–28.
5. Perkins GD, Stephenson BTF, Smith CM, Gao F. A comparison between over-the-head and standard cardiopulmonary resuscitation. Resuscitation 2004;61:155–61.
6. Handley AJ, Handley JA. Performing chest compressions in a confined space. Resuscitation 2004;61:55–61.
7. Gliwitzky B, Kohlmann T. “Over-head CPR”: a sensible method for advanced CPR with two rescue workers?. Notfall & Rettungsmedizin 2003;6:193–6.
8. Wenzel V, Idris AH, Dorges V, et al. The respiratory system during resuscitation: a review of the history, risk of infection during assisted ventilation, respiratory mechanics, and ventilation strategies for patients with an unprotected airway. Resuscitation 2001;49:123–34.
9. Wenzel V, Idris AH, Banner MJ, et al. Influence of tidal volume on the distribution of gas between the lungs and stomach in the nonintubated patient receiving positive-pressure ventilation. Crit Care Med 1998;26:364–8.
10. Hew P, Brenner B, Kaufman J. Reluctance of paramedics and emergency medical technicians to perform mouth-to-mouth resuscitation. J Emerg Med 1997;15:279–84.
11. Brenner B, Stark B, Kauffman J. The reluctance of house staff to perform mouth-to-mouth resuscitation in the inpatient setting: what are the considerations? Resuscitation 1994;28:185–93.
12. Hallstrom A, Cobb L, Johnson E, Copass M. Cardiopulmonary resuscitation by chest compression alone or with mouth-to-mouth ventilation. N Engl J Med 2000;342:1546–53.
13. Berg RA, Kern KB, Hilwig RW, et al. Assisted ventilation does not improve outcome in a porcine model of single-rescuer bystander cardiopulmonary resuscitation. Circulation 1997;95:1635–41.
14. Berg RA, Kern KB, Hilwig RW, Ewy GA. Assisted ventilation during ‘bystander' CPR in a swine acute myocardial infarction model does not improve outcome. Circulation 1997;96:4364–71.
15. Kern KB, Hilwig RW, Berg RA, et al. Importance of continuous chest compressions during cardiopulmonary resuscitation: improved outcome during a simulated single lay-rescuer scenario. Circulation 2002;105:645–9.
16. Dorph E, Wik L, Steen PA. Arterial blood gases with 700 ml tidal volumes during out-of-hospital CPR. Resuscitation 2004;61:23–7.
17. Dorph E, Wik L, Stromme TA, et al. Oxygen delivery and return of spontaneous circulation with ventilation: compression ratio 2:30 versus chest compressions only CPR in pigs. Resuscitation 2004;60:309–18.
18. Turner I, Turner S, Armstrong V. Does the compression to ventilation ratio affect the quality of CPR: a simulation study. Resuscitation 2002;52:55–62.
19. Safar P, Bircher N, Pretto E Jr, et al. A reappraisal of mouth-to-mouth ventilation during bystander-initiated cardiopulmonary resuscitation. Resuscitation 1998;36:75–6.
20. Wenzel V, Krismer AC, Arntz HR, et al. A comparison of vasopressin and epinephrine for out-of-hospital cardiopulmonary resuscitation. N Engl J Med 2004;350:105–13.
© 2005 International Anesthesia Research Society
21. Greingor JL. Quality of cardiac massage with ratio compression-ventilation 5/1 and 15/2. Resuscitation 2002;55:263–7.