The usefulness of bedside ultrasound in cardiac arrest is clear, giving the clinician instant information on the hemodynamic process resulting in arrest. My arrest sequence is generally done as follows:
Step 1: IVC assessment
Step 2: Subxiphoid cardiac views
Step 3: Lung views if pneumothorax suspected.
Step 4: remaining views if possible (eg abdominal views to find source of bleeding, etc…)
The important part (as per current recommendations) is to have minimal interference on chest compressions. The IVC view, albeit jumpy, can generally be obtained during CPR. The subxiphoid view should be “prepared” during CPR, meaning that the sonographer warns the team member doing CPR not to stop compressions until he is told to do so (unless the team are already used to ultrasound in cardiac arrest), the probe positioned optimally, then instruction given to stop for five seconds while a look +/- loop is acquired. This should be enough to look for pericardial effusion, RV/LV ratio and LV contractility. In fact, experienced sonographers can usually get this while CPR is going on in many cases. Then CPR should be restarted. Hence for now, minimal interruptions in CPR (until the concept of “stutter CPR” really emerges!!!).
Here are a couple of views with active CPR:
In this case there is a clear RV overload with a dynamic but underfilled LV.
From the information obtained in those 5 seconds, one should be able to consider the need for volume (hypovolemia), thrombolytics (pulmonary embolism suspected) or drainage of fluid (tamponade) or air (pneumothorax). The possibility of an acute myocardial infarction must be considered as it is one of the most common causes but is difficult to confirm by ultrasound.
Pulseless Electrical Activity
This may be the most exciting area in which ultrasound will change management. In the absence of ultrasound, all PEA is more or less alike: there is organized electrical activity, but no pulse. Physiologically however, the range of diagnoses is very wide, with on one end, a perfectly good heart that is empty (hypovolemic shock in extremis), and on the other, cardiac standstill despite electrical activity. An astute physician does not need a randomized clinical trial to know that the management and prognoses of those two extremes are very different. Without bedside ultrasound, however, these would appear identical: “PEA.”
The heart rate cannot be relied on since it will largely depend on the phase (both would begin as tachycardic, then eventually bradycardic until asystole occurs).
Notwithstanding guidelines, the information obtained should be considered strongly. If we start by looking at the first end of the spectrum, there would be no physiological rationale for performing chest compressions or an empty and hyperdynamic ventricle: rapid infusers and vasopressors (to recruit venous unstressed volume) should be used instead. At the other end, the heart in standstill definitely needs compressions. Of course, there is then the whole range of varying RV and LV pathologies, tamponade, etc, all of which need to be dealt with individually. It is really a huge grey zone…
CPR or no CPR?
A very important question is whether CPR should or should not be performed in certain cases of PEA. Certainly ACLS protocol dictates so. However, ACLS has not yet truly integrated bedside ultrasound into management, only suggests in a very loose way – understandably since the protocols must be applied by all, and still only few use it regularly.
I have to credit Dr. Sue, an ER doc from Atlanta, who asked me the question about CPR in extreme hypotension, and I had to rewind in my mind the cases in which I had used physiological information to overrule the ACLS protocol in one direction or another and try to formulate an answer.
It is an excellent question and made me realize that there is no clear answer for two reasons:
One: PEA is not a diagnosis but a clinical syndrome. It relies on manual pulse check (unless the arrest occurs in a patient with an arterial line), hence the line between severe hypotension and true PEA is difficult to determine. Technically and physiologically speaking, if the LV contraction is sufficient to open the aortic valve, there is a “pulse.” Now how far along the arterial circuit this pulse travels is not known…unless it is monitored.
Two: The key question then becomes the following: at what level of endogenous blood pressure is the perfusion better than with “good” CPR? We do not yet have that answer. The coronary perfusion pressure (diastolic pressure – wedge pressure) data often quotes a minimal range of 15-25 mmhg, which – if we arbitrarily choose a high-ish wedge – would suggest we need a diastolic pressure in the 40’s (also note that that data is imperfect). Hence the arterial line. Perhaps there could be a role for tissue saturation/near-infrared spectroscopy or other microvascular flow indices in the future…
Now what about the huge spectrum of cases in between? Let us exclude the cases with immediately reversible causes such as tamponade and pneumothorax, where the initial management is clear, and instead focus on differing levels of RV and LV dysfunction resulting in the absence of a palpable pulse.
Predominant RV failure – although PE should be strongly suspected and thrombolysis considered, the question remains about management if you see a hyperdynamic but underfilled LV. CPR would appear reasonable in an effort to try to get some RV to LA flow. Endotracheal milrinone, sometimes used in cardiac anasthesia, can be an option as it provides inotropy and pulmonary vasodilation.
Predominant LV or biventricular failure – if cardiac activity is present, it seems imperative to start an infusion of vasopressors, and traditionally, do CPR until there is a measurable blood pressure. I have used CPR with progressing bradycardic rhythms for a few seconds to circulate the epinephrine, with at least short-term success.
Here is a typical LV “PEA” from the subxiphoid view:
Ideally, an arterial line would be very useful in these patients, and may help to decide on an individual basis when CPR should be used. Remember that CPR on a beating heart will likely worsen cardiac output as asynchronicity and increased mean intrathoracic pressure will impair filling.
Additionally, the arterial line also allows us to notice small trends during resuscitation, such as seeing that a few seconds of CPR may help circulate a bolus of vasopressor and enable it to take effect – progressive BP increase, or that the BP may be trending downwards despite vasopressor infusion – CPR may be useful until enough vasopressor/inotropes have infused.
1. if possible, put in an arterial line
2. bedside ultrasound is mandatory if you don’t want to miss anything reversible
3. if you don’t have a palpable pulse and your diastolic pressure (arterial line) is less than 40, consider some CPR until vasopressors/inotropes have had effect.
4. if you are lucky enough to have ECMO (shout out to Joe and Zack at http://www.edecmo.org) or other mechanical support, it would be the time to consider!
I think this is actually a really interesting area to develop, and I’d really, really like to hear what other sonographer-resuscitationists are doing, or what anyone else might think!
Joe Bellezzo – yes, THE Joe, says:
Phil, I agree with all your points here. Great post! As you know, Shinar, Weingart and I recently published a rant on PEA (over at http://www.edecmo.org/13) and Weingart threw out two possible new monikers: PRE-M (Pulseless Rythm with Echocardiographic – Motion) and PRE-S (Pulseless Rythm with Echocardiographic – Standstill). I don’t disagree with any of those concepts but I think its simpler than that.
PRE-S (standstill) = asystole and you start compressions.
PRE-M (organized cardiac activity) = profound shock. In this setting I like the recent Littmann paper that gives a simplified approach (http://edecmo.org/wp-content/uploads/2014/08/A-Simplified-and-Structured-Teaching-Tool-for-the-Evaluation-and-Management-of-Pulseless-Electrical-Activity.pdf).
As you pointed out above, the real big question is at what point do you start compressions when you have cardiac motion? Your points above are spot on. The problem with compressions on a beating heart is that you don’t know what your end-point is. You lose the ability to do minute-minute diagnostics and doing any procedures with ongoing compressions is tough. And it seems to be a knee-jerk reaction for the RN or pharmacist to have an amp of epi ready to blast away at this point. NO!
I wait. I do stuff first. Step 1 is ECHO. If PRE-M (aka profound shock): EKG and arterial line NOW. Step 2: Stop, think and decide what you think is your top probable etiology of this profound shock and fix that. Step 3: reassess = repeat echo, EKG, and see what your art line pressures are doing. I try to do all that before I start compressions.
Example: “PEA” hits your door. Echo shows a wall motion abnormality and hypokinesis. EKG suggests ischemia but is not obvious STEMI. art line goes in simultaneously. I think this is MI. This heart does NOT NEED epi 1 mg! This is cardiogenic shock and I need to fix some stuff before I start pushing on the chest and blasting superhuman doses of epi! I usually start with a push dose epi (10-20 mics or so) while a pressor drip is prepped. Calcium bolus is given. I likely start dobutamine here (or milrenone if beta blocked). If that fixes your problem, then the pt goes to the cath lab. If it doesn’t, I cannulate and put the pt on VA-ECMO.
And what if I were wrong? what if this were a big PE? massive beta blocker OD? the protocol above is works in those cases too.
An aside, since you already placed a femoral venous line….and you popped in the Art line immediately, you have nice conduits to upsize to ECMO cannulas.
Great post Phil!
Thanks for sharing your approach! As you know and clearly show, a sensical physiological approach is absolutely needed in a day and age when we can (bedside ultrasound) see what’s really going on, and we can (ECMO) give these patients a fighting chance! See you at BMBTL in a couple of weeks!
PS Joe (and Zack and Scott) will be talking about all this and more at CCUS 2015! http://www.ccusinstitute.org to register soon!