As you wait not-so-patiently for your relief to show up and send you home, you consider two calls you had today: a cardiac arrest and an MVC that resulted in a severely injured patient. You wonder if there is anything new in the literature that suggests potential changes to your management of these patients. As it happens, your relief shows up with four papers in her hands, eager to talk about their implications for your practice.
These four papers—two on airway management in cardiac arrest, one on the use of epinephrine in cardiac arrest, and one on the effect of EMS-administered plasma in hemorrhaging trauma patients—are all well done and important to EMS medicine. As a paramedic, EMS physician, and data nerd, I’d like to describe these trials and share some thoughts about what they mean.
Airway Management in Cardiac Arrest: PART
Results of the Pragmatic Airway Resuscitation Trial (PART), led by Dr. Henry Wang, were published in JAMA this year.1 This was a U.S.-based multicenter, prospective, pragmatic randomized trial of adult patients with all-cause cardiac arrest. It compared the 72-hour survival of patients who had either endotracheal intubation (ETI) or a SGA (supraglottic airway; the King LT in this study) used as their first approach to airway management.
The trial involved 27 EMS agencies across the U.S., including large urban providers and smaller suburban agencies that see only a few thousand patients a year. Each agency was randomized to use either the King or ETI for its initial airway attempts for several months, then switched to the opposite strategy. The vast majority (almost 99%) of intubations were performed with direct laryngoscopy. Providers were given no additional training in airway management other than what their agency typically provided. Other than the initial airway strategy, they would treat all patients the way they normally would. In this fashion, the outcome would likely be similar if other, similar agencies adopted this strategy. This lack of additional training reflected the pragmatic approach of the trial; this is actually a strength and makes it easier to extrapolate the results to other EMS systems.
There were 3,004 patients with out-of-hospital cardiac arrest of any etiology (1,499 patients in the ETI-first group, and 1,505 in the LT-first group). The two groups were well matched in the areas likely to affect survival. More patients were alive at 72 hours in the LT group (18.3%) than the ETI group (15.4%). This 2.9% improved survival was statistically significant (95% CI, 0.2%–5.6%). The authors’ conclusion was that an initial strategy of a laryngeal tube was superior to endotracheal intubation in cardiac arrest.
While some agencies are taking these results and changing their practice to use King LTs for cardiac arrest, there are some caveats to this study that will prevent me from doing so (for now) in my agencies. There are some methodologic nits to pick, but the most important issue here is the ETI performance: First-pass success was only 51% with ETI. This is substantially lower than the FPS rate in our system. Although there was no evidence in this study to suggest this, I suspect a large part of the reason for worse outcomes with intubation can be attributed to poor performance.
Again, because of the pragmatic aspect of this study, these results are likely to be seen with many systems. Perhaps better intubation performance would erase the outcome difference seen in this study. Better performance, however, isn’t easy or cheap. If my agency is any indication, it takes a large investment in resources and training to build a QI and training system sufficient to improve endotracheal intubation. Video laryngoscopy will likely make this improvement easier for medics who do not intubate often. EMS agencies and medical directors must take a hard look and decide if their resources are best spent on improving intubation or using an SGA in cardiac arrest and focusing instead on optimizing other areas of care.
One other aspect of this trial is worth mentioning: There was a group of patients who had no advanced airway placed. Because of the analysis methodology, these patients were included in the analysis of the group to which they were randomized. This means if a patient was supposed to get an ETI but didn’t get any airway, they would be analyzed along with all the patients who got ETIs. This is called an intention-to-treat analysis and is entirely appropriate. The challenge, though, is that more of the patients who had no advanced airway attempted but were analyzed in the LT group had characteristics favorable to a good outcome than did those in the ETI group (things like witnessed arrests, short response times, shockable rhythms, or awaking prior to airway management). This may have biased the results in favor of the King LT. An additional analysis was performed to help understand the impact of this difference. It looked at survival based on the strategy actually used after eliminating those patients with no advanced airway attempted. This sensitivity analysis showed no statistical difference in survival (LT 16.0% vs. ETI 13.5%, difference of 2.5% with a 95% CI, [-0.2%]–5.2%), erasing the benefit found in the primary analysis.
It is helpful to think of why these results might be. The theoretic benefit of the King LT is that medics do not have to interrupt compressions, while endotracheal intubation has frequently been associated with such interruptions. We know from other studies that excessive compression pauses are associated with worse outcomes. This paper did not report on the proportion of patients who experienced compression interruptions in either group. This would be a helpful piece of information. Fortunately, some of the systems in this trial used monitors that recorded compression quality. This data is being analyzed in a secondary analysis and might shed additional light on this question.
My bottom line on this very important and wonderfully done study is that insertion of an LT in cardiac arrest results in better survival than a poorly inserted ETI. I don’t know, however, what the impact would be in agencies with better ETI performance.
I took a deeper dive into this study and described it for the FOAMFrat blog here.
Like the PART study, AIRWAYS-2 is a large pragmatic prospective randomized trial of airway management in out-of-hospital cardiac arrest. The lead author was Dr. Jonathan Benger, and it was also published in JAMA in 2018.2 This study was conducted by four U.K. ambulance services covering 21 million citizens, more than 40% of the U.K. population. It compared endotracheal intubation (performed exclusively with direct laryngoscopy) to the i-gel SGA.
Unlike the PART study, which randomized EMS agencies, this trial randomized paramedics to either intubation or SGA. Personally I prefer this methodology. Both ETI and SGA require specific skills that benefit from maintenance. I worry that switching back and forth between the different approaches could cause some skill decay, something more likely with intubation than SGA. The AIRWAYS-2 trial recruited paramedics who volunteered to be randomized into two groups: one group would use the i-gel as the default and initial airway on all cardiac arrests they worked, while the other would do the same with ETI. In this fashion each medic would presumably have more experience with the device they were using and maximize their skill with it.
There were 1,523 paramedics included in the trial, and they treated 9,296 patients (4,886 with the i-gel and 4,410 with ETI). The primary outcome was neurologically intact survival at 30 days as defined by a modified Rankin score of 0–3. This is the most meaningful outcome we can look at in cardiac arrest trials. There was no significant difference in meaningful survival between the two groups (6.4% with i-gel and 6.8% with ETI, adjusted difference -0.6%, 95% CI, [-1.6%]–0.4%).
Intubation success seems to have been higher in this study, although FPS wasn’t reported. Success rates for the first two attempts were 87.4% with i-gel and 79.0% with ETI. It isn’t clear if the improved ETI performance accounts for the lack of difference in outcomes compared with the worse ETI outcomes in the PART trial.
There were more patients in the i-gel group who actually had it attempted than there were patients who received intubation in the ETI group. In other words, more patients in the ETI group had either no airway attempted or an SGA placed in violation of the protocol. Benger suspects this may have been because paramedics felt they were more likely to succeed with the SGA. A sensitivity analysis was also done in this trial based on the device patients actually received (as opposed to the intention-to-treat approach of the primary analysis). This sensitivity analysis looked only at those who had an airway attempted. It found more survival in the i-gel group (3.9%) than in the ETI group (2.6%), a significant difference.
AIRWAYS-2 had a small number of patients who had CPR performance data available. When compression quality was included in the analysis, there was still no difference in survival between i-gel and ETI. This suggests, in this trial at least, compressions were not being interrupted to intubate.
There were several interesting ancillary findings to this study. There was more unintentional tube dislodgement and regurgitation after airway placement seen with the i-gel than the ETI. There was no difference in the groups’ rates of aspiration pneumonitis, however.
My bottom line on this study is that a strategy of either SGA or ETI as the initial airway in cardiac arrest is reasonable. In fact, when taken together with the PART study, I think that is the main takeaway. If your system doesn’t or can’t field an oversight, training, and QI program to assure high-performance airway management with ETI, your cardiac arrest patients will most likely do better if you adopt an SGA strategy. I also wrote a deeper dive on this trial for the FOAMfrat blog.
Epinephrine in Cardiac Arrest
The PARAMEDIC trial (for Prehospital Assessment of the Role of Adrenaline: Measuring the Effectiveness of Drug Administration In Cardiac Arrest) is a wonderful study for two main reasons. First, that’s an awesome acronym, and it was probably worth doing the study just for that. Next, it was a well-done study that aimed to answer a very important question that has eluded definitive answer for years: Does epinephrine improve outcomes in cardiac arrest?3
Though epinephrine given every 3–5 minutes is one of the most fundamental tenets of ACLS, there is little evidence to support it. Whether it is given at 1 mg, 3 mg, or 5 mg, we have yet to show improved neurologically intact survival in a prospective study. We have, however, pretty conclusively shown it increases rates of return of spontaneous circulation (ROSC) and survival to hospital admission. While this may seem like a good thing, if we are only getting pulses back on patients who never wake up or live with devastating disability, we are not really doing those patients any good.
With this dearth of information in mind, the U.K. National Health Service funded this large study. It was a multicenter study of five NHS ambulance systems that, prospectively and in a double-blinded fashion, randomized adult patients with all causes of out-of-hospital cardiac arrest to get either epinephrine 1 mg or normal saline every 3–5 minutes. The actual method used for blinding was brilliant: Researchers prepared special study boxes that contained 10 prefilled 10 ml syringes. These syringes contained 1 mg of either epinephrine or normal saline. Patients would be randomized to a group when the medic opened the box. Because epinephrine is known to be effective in anaphylaxis and asthma, patients who arrested with these conditions were excluded from the trial.
There were 8,014 patients enrolled (4,015 getting epi, 3,999 placebo). The primary outcome was survival at 30 days. There was improved survival with epinephrine versus placebo (3.2% vs. 2.4%, OR 1.39, 1.06–1.82). Predefined secondary outcomes included ROSC (36.3% with epi vs. 11.7% with placebo) and, most important, neurologically intact survival at 30 days (2.2% with epi vs. 1.9% with placebo, OR 1.18, 0.86–1.61, not a significant difference). Most telling, they looked at those who were alive at 30 days and determined the proportion with a devastating neurologic disability. This group was much higher if given epinephrine (31.0%) than if given placebo (17.8%). This means that more patients survived with epinephrine but did so with worse neurologic function.
There are several theoretical reasons why epinephrine should both be helpful and harmful. It stimulates alpha receptors, leading to increased diastolic aortic constriction, which improves coronary artery perfusion. Unfortunately it also stimulates beta receptors, leading to increased myocardial irritability and demand that can worsen ischemia, as well as promote inflammatory changes that may be associated with worse neurologic outcomes.
My bottom line on this important paper is that it conclusively tells us giving 1 mg of epinephrine every 3–5 minutes improves ROSC and survival but at the cost of horrible neurologic outcomes, something most people would consider to be a contraindication to its use. In other words, we now know what doesn’t work. We don’t, however, know what does work. It is possible epinephrine may still play a positive role for some patients, at some interval, at some dose. While this may frustrate those who want the answer, it is tremendously exciting because, in true scientific fashion, it gets us one step closer to the truth.
Plasma in Severe Trauma
The Prehospital Air Medical Plasma trial (PAMPer) is another multicenter prospective randomized pragmatic trial.4 It randomized adult patients with severe trauma and evidence of massive bleeding to get either standard therapy or two units of thawed group AB plasma. The inclusion criteria were injured patients with either profound hypotension (SBP less than 70) or hypotension (SBP less than 90) and tachycardia (HR greater than 108). The plasma was given prior to any crystalloids. Based on local protocol and availability, some patients in both the plasma and control groups also received packed red blood cells (PRBCs).
Patients were transported by aircraft from 27 bases to nine trauma centers, and 564 were included, with 230 getting plasma and 271 getting standard therapy. Most of these patients were male (73%), had blunt trauma (82%), and got an operation within the first 24 hours (58%). More than half (51%) had prehospital intubation, 35% got PRBCs, and the median injury severity score was 22. The groups were well matched among these factors. The primary outcome was 30-day mortality.
Those patients receiving plasma had a mortality rate of 23.2% vs. 33% in the standard care group, a difference of -9.8% (95% CI, [-18.6%]–[-1.0%]). The odds ratio for death with plasma vs. standard therapy was 0.61 (0.40–0.91), meaning the odds were 39% lower if given plasma. My calculations show this 10% decrease in mortality resulting in a number needed to treat to save one life of 10. That is a very impressive number. For context, the NNT to save one life with aspirin when experiencing a STEMI is 42.
There were no differences in transfusion-related lung injury or allergic reactions. The main factor limiting this therapy in the field is logistics. The shelf-life on thawed plasma is five days. It is expensive and requires specific maintenance. These logistic restrictions make the use of thawed plasma, even with its impressive mortality benefit, impractical. The good news, however, is that a much more stable plasma product has been developed and successfully deployed. The Israel Defense Forces uses freeze-dried plasma for severe trauma. A U.S. version of this product is under review, with hopes of approval in 2020. The goal would be a product that is stable for more than a year at room temperature and needs only reconstitution with 200 ml of saline.
If no problems arise with this and it gets approval, plasma could become a feasible option for ground EMS. A 10% decrease in mortality would be a very good thing.
In order to summarize the findings of these important studies, the initial use of a supraglottic airway in out-of-hospital cardiac arrest is a safe and reasonable approach. It is easier to maintain competence with SGAs than endotracheal intubation, and that’s likely the approach most U.S.-based EMS agencies should adopt. We don’t yet know the effect of competently performed intubation on cardiac arrest.
Epinephrine in cardiac arrest, as we currently use it, does not benefit patients. Epinephrine increases ROSC and short-term survival, but it comes with the cost of increased neurologic devastation. We have to stop this.
Unfortunately, while we now know what doesn’t work, we don’t yet know what does. We’ll likely see an ongoing role for epi in cardiac arrest, but only for a select group of patients and likely at lower doses spread further apart.
The use of plasma products in severely injured, hypotensive trauma patients can decrease mortality by up to 10% with a number needed to treat to save one life of 10. Logistical difficulties currently limit its adoption, but a freeze-dried plasma under regulatory review could be feasible for ground-based EMS.
1. Wang HE, Schmicker RH, Daya MR, et al. Effect of a strategy of initial laryngeal tube insertion vs. endotracheal intubation on 72-hour survival in adults with out-of-hospital cardiac arrest: A randomized clinical trial. JAMA, 2018 Aug 28; 320(8): 769–78.
2. Benger JR, Kirby K, Black S, et al. Effect of a strategy of a supraglottic airway device vs. tracheal intubation during out-of-hospital cardiac arrest on functional outcome: The AIRWAYS-2 randomized clinical trial. JAMA, 2018 Aug 28; 320(8): 779–91.
3. Perkins GD, Ji C, Deakin CD, et al.; PARAMEDIC2 collaborators. A randomized trial of epinephrine in out-of-hospital cardiac arrest. N Engl J Med, 2018 Aug 23; 379(8): 711–21.
4. Sperry JL, Guyette FX, Brown JB, et al.; PAMPer study group. Prehospital plasma during air medical transport in trauma patients at risk for hemorrhagic shock. N Engl J Med, 2018 Jul 26; 379(4): 315–26.
Jeffrey L. Jarvis, MD, MS, EMT-P, FACEP, FAEMS, is EMS medical director for the Williamson County EMS system and Marble Falls Area EMS and an emergency physician at Baylor Scott & White Hospital in Round Rock, Tex. He is board-certified in emergency medicine and EMS.