Evolving CPR guidelines follow the growing understanding of the physiology surrounding cardiac arrest. For example, it’s been shown that the maintenance of high-quality chest compressions with minimal interruptions is a cornerstone of providing adequate CPR.1 However, the role of airway and ventilation management during CPR remains less well understood.
Although the safest and most effective airway management for the patient in cardiac arrest still remains to be ascertained, the initial approach in airway management in out-of-hospital cardiac arrest (OHCA), however, is typically the application of the bag-valve mask (BVM) to assist ventilation. The gold standard for airway management for OHCA has historically been endotracheal intubation (ETI), a procedure requiring considerable training and skills maintenance to be performed successfully.2,3
The development of additional airway adjuncts—including supraglottic airways (SGAs) such as the laryngeal mask airway (LMA) and the King Laryngeal Tube (King LT)—have offered an intermediate approach, providing an advanced airway alternative while generally requiring less training and skill than that required for ETI. Application of these approaches in the prehospital setting has varied widely, depending principally upon the treatment protocols of individual services.
Substantial published research, however, has suggested that advanced airway use in OHCA management is associated with worsened patient outcomes.4–12 This suggests that some survival advantage is associated either with the BVM device specifically, with avoiding the use of advanced airways, or perhaps reasons associated with both.
The objective of the following study was to elucidate a relationship between survival from OHCA and the type of airway technique employed during resuscitative efforts. The spectrum of resuscitative literature was studied to isolate factors associated with survival from OHCA relative to the use of BVM alone vs. advanced airways. Consideration was given to examining each study for any evidence regarding the maintenance of the quality of CPR during resuscitation efforts and its potential relationship to “the BVM effect.”
The authors conducted a literature review in July 2014 to identify papers addressing airway management during OHCA. Electronic databases PubMed and Google Scholar were searched using the keywords out-of-hospital, prehospital, emergency medical services, heart arrest, cardiac arrest, airway and survival. Relevant material was also obtained through reviewing references from articles identified in the study and by contacting subject matter experts. A total of 171 scientific studies were found.
The various airway management techniques analyzed were BVM, SGA (LMA, King LT, Combitube), esophageal obturator airway (EOA) and ETI. Generally the comparisons in the studies focused between BVM and advanced airways or among different advanced airways. Primary outcomes largely focused on survival to discharge, with some studies including neurological function post-discharge as well.
Specific attention was then directed toward criteria addressing cardiac arrest management in the out-of-hospital environment, cardiac arrest victims, comparison of advanced airway with BVM and survival outcomes.
Exclusion criteria included qualitative studies, studies focused on traumatic arrests, studies comparing only advanced airways excluding BVM, and studies solely commenting on the training or feasibility of certain airway use.
Our search found nine observational studies meeting our inclusion criteria and specifically associating survival from OHCA with the type of airway management used during CPR. These papers were published from 1997 to 2014 with patient data ranging from 1990 to 2011. Patient populations varied substantially, with the smallest sample size being 355 and the largest being 649,359.
The datasets covered many different regions across the world, including North America, Europe and Asia. Each paper employed a set of controls for confounding variables, and generally followed the collection of data using recommended Utstein guidelines. Statistical analysis generally involved multivariate regression models with some studies using propensity-score matching.
Study 1 reviewed data from 1991 through 1994 collected in a “Heartstart” program.4 Resuscitation was attempted for 8,651 patients with 3,427 (39.6%) attempts at ETI. The primary results found a survival to discharge rate for patients receiving ETI of 3.7% vs. 9.1% for patients receiving BVM alone (p < 0.001).
Of interest, the proportion of patients intubated increased with the number of defibrillatory attempts and was higher in patients with unwitnessed arrest. The trend in decreased survival with ETI vs. BVM persisted regardless of EMS witnessed, bystander witnessed or unwitnessed arrest, or number of shocks.
The EMS providers in the study followed European Resuscitation Council guidelines, which stated that intubation should only be attempted after three shocks if spontaneous circulation hadn’t been restored. However, a substantial proportion of patients shocked less than three times were intubated, suggesting that these patients regained a pulse but were either not spontaneously breathing or deteriorated into a non-shockable rhythm.4
Study 2 examined almost 11,000 patients with OHCA in which ETI was attempted in 5,118 (47.5%) patients.5This study demonstrated a lower chance of one-month survival correlated with the use of ETI during management of OHCA. Patients who were successfully intubated had a 3.6% one-month survival rate vs. 6.4% who weren’t intubated.5
The authors, in noting the limitations of their study, found that there was no way to control for experience or training in placement of airways by the medics delivering care and that their study wasn’t randomized, making control for possible confounders difficult.5
Study 3 highlighted the unique characteristics of pediatric patients and adults in the OHCA patient population.6 The authors studied 624 patients divided into three age groups: < 1 year (infants; n = 277), 1–11 years (children; n = 154), and 12–19 years (adolescents; n = 193). The study had significant power in their age comparisons and was able to show statistically significant differences among the various age groups studied.
They found that “the incidence of OHCA in infants approaches that observed in adults” and is “lower among children and adolescents.” They also found that “survival to discharge was more common among children and adolescents than infants or adults.” Finally, these authors concluded that there was no significant difference in survival among the types of airways used in pediatric OHCA victims.6
Study 4 evaluated 1,294 nontraumatic OHCA patients from 1994–2008 in southwestern Los Angeles County.7This study found that 1,027 (79.4%) patients received ETI, 131 (10.1%) received either Combitube or EOA, and 131 (10.1%) received only BVM. The overall survival to discharge rate was 4.3%. Odds ratio for survival to discharge of patients receiving BVM compared to ETI was 4.5 (95% CI: 2.3–8.9) after adjusting for bystander CPR, witnessed arrest, age, sex and location of arrest.
Interestingly, the authors found that the group receiving Combitube or EOA had no survivors. In this paper there was a reported rate of bystander CPR of 45%, and no significant association was found between the rate of bystander CPR and survival to discharge.7
Research has shown a substantial survival benefit from OHCA with the use of BVM ventilation rather than advanced airways such as ETI or SGA. Photo Courtney McCain
Study 5 analyzed 355 OHCA patients in Tokyo whose time from emergency call to hospital arrival was > 30 minutes.8 This study focused on comparing outcomes between advanced airway use and BVM use in prolonged cardiac arrest. The authors discovered a significant increase in overall return of spontaneous circulation (ROSC) and ICU admission in patients who received an advanced airway vs. BVM alone. However, no difference in prehospital ROSC or survival to discharge was apparent between the two groups.
Of note, the analysis found a similar time from the emergency call to arrival on scene in both groups, but the patients receiving advanced airways had longer on-scene management times by approximately two minutes. Thus, in OHCA where on-scene care was prolonged, performing an advanced airway may lead to a higher overall rate of ROSC without improving the overall rate of survival.8
Study 6 looked at a South Korean OHCA database including patients from 2006–2008.9 Of 5,278 patients reviewed, 250 (4.7%) received ETI, 391 (7.4%) received LMA and 4,637 (87.9%) received BVM. Overall survival to discharge was found to be 6.9%. Odds ratio for survival to discharge for ETI vs. BVM alone was 1.44 (95% CI: 0.66–3.15) and wasn’t statistically significant. The odds ratio for survival to discharge for the use of LMA vs. BVM alone was 0.45 (CI: 0.25–0.82).
It’s of interest that the study involved EMT-intermediates who were trained in airway placement through the use of manikins as opposed to training on patients in the operating room.
This training using simulation, combined with a median of two LMA uses per provider over two years, might suggest less skill proficiency, though there was no mention in the study of the annual number of ETIs by the providers.9
Study 7 examined the relationship between two advanced airways (SGA and ETI) using data from the Resuscitation Outcomes Consortium (ROC) PRIMED Trial.10 The authors studied the data from 10,455 adult OHCAs between June 2007 and October 2009. In this study, 8,487 patients (81.2%) received ETI and 1,968 (18.8%) received SGA.
The authors found that the overall survival to hospital discharge with satisfactory functional status was 4.7% with ETI and 3.9% with SGA (adjusted odds ratio: 1.40; 95% CI: 1.04–1.89). However, careful analysis of all data points (as revealed in supplemental data fields within the study) in this report reveals that patients receiving no advanced airways (BVM only or BVM after failed advanced airway attempts) had a significantly higher rate of survival to discharge. Indeed, when the final airway management used was BVM, the odds ratio of survival over the successful use of an ETI was 1.79 (CI: 1.33-2.40; p < 0.001).10
Study 8 presented an observational study of patients from the All-Japan Utstein Registry, a vast nationwide database of OHCA patients.11 This study found favorable outcomes with BVM airway management over ETI that had strong statistical significance due to the power produced by this large patient population, revealing a strong inverse relationship between the use of advanced airways and favorable neurological outcomes.
The overall unadjusted favorable neurological survival in this study was 2.2%, ranging between 1.1% for OHCA patients managed with advanced airways and 2.9% for patients managed with BVM (odds ratio: 0.38; 95% CI: 0.36–0.39). This data must be interpreted carefully with respect to the low overall survival in this observational study as compared to various urban centers across the world.11
Study 9, the most recent study found by the authors on this subject, analyzed 10,691 OHCA patients from the Cardiac Arrest Registry to Enhance Survival (CARES) in 2011.12 Of these patients, 5,991 (56%) were treated with ETI, 3,110 (29%) received SGA and 1,929 (18%) had no advanced airway placement. The data demonstrated 5.4% survival to discharge with good neurological outcomes in patients receiving ETI, 5.2% in patients receiving SGA, and 18.6% in patients receiving BVM only.
Of note, patients receiving ETI tended to be slightly older, more likely to be male, and less likely to receive defibrillation by use of an automated external defibrillator located in a public place. The authors also found that patients receiving BVM alone tended to have suffered OHCA in a public place, that the arrest tended to have been witnessed by EMS, and that the patient was more often in a shockable cardiac rhythm.12
The authors stated their beliefs that the association of improved survival with BVM alone “reflect[ed] the presence of unmeasured and immeasurable confounders.” These confounders could include short distance to the hospital, provider procedural skill, perceived health status of the patient and airway anatomic factors.12
They called for a future study that would integrate information including airway management steps such as duration of attempts, ventilation rates and procedures attempted in the ED.12
The management of OHCA patients remains one of the most difficult clinical challenges in the practice of medicine. These patients may have an arrhythmic cause of arrest that may be rapidly treated through defibrillator efforts, or they may have a cause of arrest that’s the result of a devastating event such as airway obstruction, massive pulmonary embolism, or major trauma that may often be inherently lethal.
The approach to these patients must be systematic, with the rescuer initiating CPR, performing rhythm analysis, managing the airway and providing IV therapy as indicated. The chance of survival for these patients is multifactorial, including the length of time in cardiac arrest, the cause of the arrest and the skill level of the rescuer.
Evidence has become available demonstrating that providing high-quality CPR—as measured by maintaining a high compression fraction,1 satisfactory compression depth,13 appropriate compression rate,14 and limiting peri-shock pauses15—is essential to optimizing survival with good neurological outcome. Nonetheless, growing evidence suggests an additional association with optimizing survival, namely the choice of airway used during resuscitation.
This article has brought together a number of studies that have found an association between the type of airway utilized during cardiac arrest management and survival. The question arises as to whether the choice of the type of airway utilized during resuscitation—i.e., BVM vs. advanced airway—is an independent predictor of survival or whether the airway choice is associated with other factors that may affect the chance for survival.
For example, in study 9, OHCA patients receiving BVM alone were significantly more likely to have suffered cardiac arrest in the presence of an EMS provider, arrested in a public place or been in a cardiac rhythm amenable to defibrillation.12
The underlying cause of this phenomenon remains to be determined. On the other hand, the above associations addressing the benefit of BVM alone over advanced airway, while statistically significant, weren’t orders of greater magnitude. So, the apparent benefit of BVM alone indeed may yet be found to be related to measurable or unmeasurable confounders, as suggested by study 9’s authors.
The underlying cause of this phenomenon remains to be determined.
We propose the BVM effect of enhanced survival with BVM over advanced airway use is also likely unrelated to the maintenance of the overall quality of CPR. Study 8’s authors also found this BVM effect in their enormous observational study, in spite of their overall reported rate of survival from OHCA of a bit less than 3% (only 1.0% with ETI).11
This low overall survival rate, as compared to centers in which OHCA cardiac arrest quality is carefully managed, raises the possibility of potential confounders that might affect survival, such as the quality of CPR performed. Importantly, traumatic arrests were included in this study,11 likely decreasing overall survival compared to a cardiac arrest patient population not including traumatic arrest patients.10
The BVM effect was also present in observational study 7, examining data from the ROC PRIMED trial.10 This observational analysis of the data from a prospective, randomized trial of OHCA patients included only non-trauma patients and excluded various populations (e.g., pediatrics). The study required that the participating agencies meet benchmarks for quality of CPR.
These studies reveal the existence of a BVM effect from multiple regions of the world that are quite heterogeneous. This suggests that this effect persists across regional variations in training, available equipment, attention to CPR quality and skill levels. This effect was also observed independent of the years studied (1990–2011).
Although it’s possible common underlying phenomena produce this effect, it remains to be demonstrated conclusively why OHCA patients managed with BVM alone seem to have improved survival. It has been well described that the patient in cardiac arrest suffers from over-ventilation during resuscitation.16 We propose that it’s more difficult to over-ventilate patients receiving BVM only, as compared to an advanced airway. Thus, part of the cause of the BVM effect may be due to less over-ventilation of patients.
Other possibilities for the BVM effect include less interruption of chest compressions with BVM compared with ETI and other advanced airways, and the avoidance of the risk of esophageal intubation that may occur with ETI, but further study is needed to elucidate the causes of or associations with the BVM effect.
To better understand the cause of the BVM effect, such studies should include factors such as response time, the time to first compression, the quality of CPR, initial cardiac rhythm, interruptions in chest compressions for airway placement, the rate of assisted ventilation, type of airway selected initially, final type of airway placed, time to airway attempt, number of airway attempts, on-scene time and final determined cause of the cardiac arrest. Only through the careful inclusion of these and possibly other parameters in future studies can the cause of the BVM effect be understood.
Numerous studies addressing airway management in OHCA have shown a strong association between improved survival with treatment using BVM alone rather than with advanced airways. This BVM effect appears to persist despite variations in geographical region, patient population, and CPR quality.
The underlying cause of this phenomenon remains to be determined. Of note, the survival benefit with BVM alone vs. advanced airways doesn’t appear in the pediatric population in the papers reviewed by the authors. The authors recommend that a prospective randomized study be conducted in order to explore this finding and to attempt to determine its causation.