.
Introduction
.
In 2008 and 2013, two prospective RCTs from Greece reported benefits from the combination of vasopressin, steroids, and epinephrine (VSE) for in-hospital cardiac arrest. However, other studies investigating the addition of vasopressin alone to epinephrine have been negative. Consequently, vasopressin has been removed from the AHA/ACC algorithms, with a specific recommendation against the use of vasopressin in combination with epinephrine. Meanwhile, these same guidelines contain a Class IIb recommendation to consider VSE for in-patient cardiac arrest. How should we approach this? (1)
.
.
VSE: Evidence about vasopressin, steroid, and epinephrine
.
Mentzelopoulos 2009
.
This was a single-center prospective double-blind trial which randomized 100 patients with in-hospital arrest to epinephrine vs. epinephrine plus a combination of three interventions: vasopressin 20 IU for up to five cycles of CPR, methylprednisolone 40 mg IV during CPR, and tapered stress-dose hydrocortisone (300 mg/d) for patients with post-arrest shock. Patients treated with VSE had improved return of spontaneous circulation (ROSC; 81% vs. 52%; p=0.003) and survival to hospital discharge (19% vs. 4%; p=0.02). Results were perhaps most dramatic among patients who developed post-resuscitation shock, in whom survival to discharge was 30% with VSE (8/27 patients) versus none in the control group (0/15; p=0.02). Patients receiving VSE had decreased levels of pro-inflammatory cytokines, improved hemodynamics, and less organ failure:
.
.
Mentzelopoulos 2013
.
This was a larger replication of the 2009 study, with attention paid to address the weaknesses of the 2009 study. Rather than a single-center study, this was performed at three centers in Greece. The study was better powered, with an increase from 100 to 268 patients with in-hospital cardiac arrest. Finally, the primary outcome was more patient-centered: discharge with good neurologic function. The intervention (VSE) was exactly the same as in the 2009 study.
.
The results were nearly identical to the 2009 study. Patients in the VSE group had a higher rate of ROSC (84% vs. 66%; p=0.005) and discharge with good neurologic outcome (14% vs. 5%; p=0.02). Compared to patients with post-arrest shock in the control group, patients with post-arrest shock in the VSE group had an improved rate of discharge with good neurologic outcome (21% vs. 8%, p=0.02):
.
.
V-E: Evidence about vasopressin and epinephrine
.
Gueugniaud 2008 performed a double-blind RCT comparing epinephrine vs. epinephrine plus 40 IU vasopressin during the first two cycles of CPR among out-of-hospital cardiac arrest. Despite being very well powered (n = 2,894), there was no evidence of benefit from adding vasopressin (Table below). However, there were also no adverse events observed in patients receiving vasopressin. The average dose of vasopressin administered to subjects in this study (77 IU) was nearly identical to the average dose of vasopressin in the two trials of VSE (73 IU and 70 IU). This study suggests that adding ~80 IU of vasopressin on top of epinephrine has little clinical effect (neither benefit nor harm).
.
.
S-E: Evidence about steroid and epinephrine
.
Theoretical evidence suggests a potential benefit of steroid during cardiac arrest. Correlational studies show higher levels of cortisol in survivors. Animal studies have suggested that intra-arrest steroid improves the return of spontaneous circulation and neurologic outcomes (Smithline 1993, Katz 1989).
.
Tsai 2007 performed a prospective non-randomized trial on the effect of a single intra-arrest dose of 100 mg hydrocortisone for out-of-hospital cardiac arrest. Hydrocortisone was provided whenever it was possible to obtain consent (36/97 patients). Patients treated with hydrocortisone had improved rates of ROSC (61% vs. 39%, p=0.038), but the same in-hospital mortality (92% vs. 90%). The greatest differences were observed when the study drug was given quickly after arrest:
.
.
Despite the potential for confounding factors, this study supports the concept that steroid improves ROSC. However, improvement was only transient, suggesting that ongoing steroid administration may be needed for a sustained benefit.
.
S: Theoretical basis for steroid in post-arrest shock
.
Following ROSC, many patients experience a sepsis-like state characterized by a surge in pro-inflammatory cytokines and vasodilation. Cardiac arrest impairs the function of the adrenal axis, leaving patients especially vulnerable to post-arrest shock (Varvarousi 2014).
.
Although the efficacy of steroid in septic shock remains controversial, it has been demonstrated to reduce the vasopressor requirement. Similar effects could be beneficial in post-arrest patients who may be especially sensitive to hypotension (which could worsen anoxic brain injury) and adverse effects from vasopressors (particularly pro-arrhythmia).
.
Intra-arrest corticosteroid is an attractive concept. One limitation of anti-inflammatory therapies in sepsis is that the inflammatory cascade has already been unleashed by the time the patient is undergoing treatment. Alternatively, providing intra-arrest steroid could modulate the inflammatory response as it begins to unfold.
.
Summary of all evidence
.
VSE is supported by two RCTs involving a total of 368 patients, which is more than the two studies used as the basis for therapeutic hypothermia (Bernard et al. and the HACA trial, which together included 350 patients). VSE trials were more rigorous than the hypothermia trials because treating clinicians were blinded to the intervention (in hypothermia trials, patients being cooled received more attention). The main limitation of the VSE studies is that they were both performed by the same group of investigators.
.
.
Although the efficacy of VSE remains in question, there is considerable evidence that vasopressin and steroid are reasonably safe. The VSE trials noted no adverse events. Larger studies of vasopressin have also found this to be safe in cardiac arrest. Although there is less data about steroid in cardiac arrest, stress-dose steroid has been found to be fairly safe among critically ill patients. There is occasionally concern that steroid could interfere with healing of acute myocardial infarction, but this appears unfounded (Giugliano 2003).
.
Putting evidence into practice
.
From an evidence-based standpoint, it would be ideal to use the full VSE protocol. This was given a Class IIb recommendation by the AHA/ACC for in-hospital cardiac arrest.
.
However, with negative evidence regarding vasopressin, practice is already moving away from the use of vasopressin in cardiac arrest. The removal of vasopressin from the 2015 AHA/ACC algorithm will likely accelerate this trend. Thus, in the near future, ACLS teams may be unprepared to mix and administer vasopressin. This may impair the ability to perform a full VSE protocol.
.
Given that the benefit of VSE seems to derive from the steroid component, it may be reasonable to use steroid plus epinephrine in a situation where VSE cannot be logistically achieved. This combination of steroid and epinephrine is weakly supported by the AHA/ACC guidelines for out-of-hospital cardiac arrest (Class IIb recommendation).
.
Regardless of how ROSC is achieved, for patients with post-arrest shock the use of stress-dose steroid should be considered. No studies have been done specifically to investigate the role of stress-dose steroids for post-arrest shock. However, in both of the Mentzelopoulous studies, patients with post-arrest shock treated with stress-dose steroids had improved survival.
.
In-hospital cardiac arrest (IHCA) vs. out-of-hospital cardiac arrest (OHCA)
.
One point of contention is whether evidence obtained in one type of arrest is applicable in the other type of arrest. Ideally, there would be adequate evidence from both settings, but this is not the case. Evidence about cardiac arrest is so sparse that the adult basic life support guidelines are based partially on studies of baby pigs.
.
.
Until we have more evidence, it is probably safe to assume that there are more similarities between adult OHCA and adult IHCA than between adults and piglets. Furthermore, OHCA and IHCA populations are heterogeneous, so findings derived from either population may be fairly generalizable. For example, the VSE trials included patients with any rhythm located anywhere in the hospital (ICU, ward, emergency department, or operating room).
.
This is reminiscent of debates about using targeted temperature management (TTM) for IHCA. RCTs investigating targeted temperature management have all been done on OHCA patients. So, if one truly believes that IHCA and OHCA are distinct entities, then targeted temperature management shouldn't be used for IHCA. Of course, targeted temperature management is currently recommended for both IHCA and OHCA (2). Thus, there seems to be a double-standard regarding TTM vs. VSE: why is it acceptable to generalize TTM data from OHCA to IHCA, but not to generalize VSE data from IHCA to OHCA?
.
European 2015 Guidelines
.
Although this post focuses on the AHA/ACC guidelines, the European Resuscitation Council has also released fresh guidelines for 2015. They state:
.
.
It is notable that the ERC and AHA/ACC guidelines make conflicting recommendations, although they were released simultaneously and based on identical evidence. Despite attempts to be evidence-based, insufficient evidence exists to reach any definite conclusions.
.
- VSE is the only pharmacotherapy that has ever been shown to improve survival with good neurologic outcome. The AHA/ACC weakly recommends VSE (Class IIb) for inpatient cardiac arrest.
- The addition of vasopressin alone to epinephrine has not improved outcomes, leading the AHA/ACC to recommend against adding vasopressin alone to epinephrine.
- In a context where VSE cannot be implemented, a reasonable approach might be to simply add 40 mg of methylprednisolone during CPR with epinephrine. The AHA/ACC weakly supports this, with a Class IIb recommendation to use steroid for out-of-hospital arrest.
- The occurrence of post-cardiac arrest shock is common, with some similarities to septic shock (i.e. excessive inflammation causing vasodilation). Stress-dose steroid may be considered for these patients.
- More evidence is needed, but in the interim it seems reasonable to utilize therapies where the benefit appears to outweigh the risks.
Related posts from this blog
- Errors of commission vs. omission: Which is worse?
- Steroids in septic shock
More information on Mentzelopoulous 2013 study of VSE:
- Pubmed link to full study
- EMCrit podcast
- The Bottom Line review
- EM Lit of Note
- Buddineni et al. in Critical Care
- Medscape review of the trial
More information about 2015 AHA/ACC & ERC guidelines:
- Full AHA/ACC guidelines in Circulation 18 (Suppl 2)
- Full ERC guidelines here
- Rebel Cast: Five AHA/ACC updates
- EM Cases: ACLS guidelines 2015
Notes
(1) Note that this blog is written from the perspective of a health-care system which uses epinephrine per AHA/ACC guidelines. Whether this is the best approach is a question for another day.
(2) With regards to TTM, this debate has been simplified dramatically by our use of TTM at 36C, which is easier and less risky than cooling to 33C. Thus, if there is ever a question of whether the patient should receive TTM, it is best to err on the side of caution and just use TTM at 36C. More discussion about TTM at 36C here.
Conflicts of Interest: Never.
Image credits:
http://www.cliparthut.com/secretary-cartoons-clipart-7rTvKj.html
https://commons.wikimedia.org/wiki/File:Sow_and_five_piglets.jpg
Latest posts by Josh Farkas (see all)
- Pulmcrit wee: The cutoff razor - April 15, 2024
- PulmCrit Blogitorial – Use of ECGs for management of (sub)massive PE - March 24, 2024
- PulmCrit Wee: Propofol induced eyelid opening apraxia – the struggle is real - March 20, 2024
Thank you for writing him! I'm really glad to see Dr. Mentzelopoulos is on board with the AHA's stance on VSE. I'm not a big believer in VSE (any survival benefit can't possibly be that good… can it?) and also think we've only seen one truly positive study, but knowing the lead author is very aware of the limitations of their work and isn't out there trying to push an unvalidated therapy makes me much more receptive to the concept.
Did any new RCT, etc (excluding meta-analyses, etc) come out since the 2010 ACLS Guidelines were published that would be considered anti-vasopressin? I am not aware of any, so essentially the AHA had stronger evidence for vasopressin and chose to remove it. Seems odd, great point about IHCA/OHCA and TTM. Love it.
Also, I know there are a myriad of issues with this study design (and it was small), but I liked the idea: http://www.ncbi.nlm.nih.gov/pubmed/24871568 (no affiliation with the authors or institution where this came from).
Thanks Matthew, Absolutely agree, there are definitely cases where IHCA is a different animal from OHCA (perhaps most). On the flip side, there are some cases where the borderline between IHCA and OHCA is fuzzy. For example, imagine that paramedics are bringing an unstable cardiac patient to the emergency department. If he codes ten feet outside the hospital then that is an OHCA, whereas if he waits another minute and codes inside the emergency department then that is an IHCA. Similarly, if someone codes in the back of an ambulance with pre-existing IV access and critical care professionals immediately provide… Read more »
Great stuff, thanks for sharing. Enjoyed your podcast and certainly agree with Dr. Mentzelopoulos's e-mail. Unfortunately, according to clinicaltrials.gov, there are no new trials on VSE registered. Thus, it may be several years until we have more data.
Excellent post and review of the evidence!
I wrote the lead author, Dr. Spyros Mentzelopoulos about the VSE protocol not receiving a stronger recommendation. With his consent I published his response at http://pharmacyjoe.com/episode26
Hi Josh,Long time lurker, first time poster! I love your stuff. Regarding the differences between out of hospital and in hospital differences; I agree that there are probably more similarities then differences but I wonder if there are some cases where the differences are significant. For starters, people who are mid-treatment in the hospital may have other interventions already on board which may preclude them to benefit from an alternate intervention. The case which immediately comes to mind is a patient who codes while being already on some form of catecholeamine (epi/nor-epi). In this case vasopressin may provide better vascular… Read more »