EM Nerd-The Case of the Tell-Tale Heart

We can all agree that the use of bedside ultrasound (US) has changed the practice of Emergency Medicine for the better. But with the addition of such a tool, we are now faced with the question of how best to apply this novel modality into our practice. The emergency management of cardiac arrest seems like the ideal condition in which to utilize the diagnostic powers of bedside US. But a recent article published in Resuscitation by Gaspari et al illustrates that blending theoretical utility with clinical reality is not always as simple as we anticipate (1).

These authors examined 953 patients who experienced either an out of hospital cardiac arrest (OHCA) or arrest in the Emergency Department, found to be in either asystole or PEA. Patients were excluded if they were in ventricular fibrillation or pulseless ventricular tachycardia, experienced ROSC after only a brief resuscitation, or who had do-not-resuscitate (DNR) orders prior to their arrest. The authors examined the ability of cardiac motion visualized on bedside US to predict ROSC, survival to hospital admission, and survival to discharge in this subset of cardiac arrest patients. On first glance, bedside US appears to have underperformed in its predictive capabilities. Of the patients who were found to be in asystole, the absence of cardiac motion predicted non-survival to hospital discharge with a sensitivity of 90% and a specificity of 0%. Among the patients in PEA, bedside US demonstrated a sensitivity and specificity of 47% and 91% respectively. In the entire cohort, independent of rhythm, cardiac motion on bedside US predicted survival to hospital discharge with a sensitivity of 66.7% and a specificity of 67.7% (6).

And while these values are mediocre at best, I think this is more of a statement on our current understanding or misunderstanding of cardiac arrest in the Emergency Department than the performance of bedside US. These authors enrolled patients who experienced OHCA, who had not achieved ROSC at the time of arrival to the Emergency Department. Additionally, they excluded all patients who were still in ventricular fibrillation or pulseless ventricular tachycardia. Of the patients who experienced their arrest in the Emergency Department, the authors excluded all patients who responded to initial resuscitative efforts or who experienced a v-fib/v-tach arrest. Essentially this was a cohort of refractory arrest patients, which are known to have dismal outcomes (2). Survival to hospital discharge among the whole cohort was only 1.3%. And while cardiac motion on US did identify a subset with a slightly higher rate of survival (3.8% vs 0.6%), other historical features (downtime, rhythm on arrival to the ED, and location of arrest) had similar predictive values.

This is a general misconception regarding the meaning of cardiac activity observed on bedside US. The majority of these patients were in arrest for a prolonged period of time. They were either found in PEA or asystole or had transitioned from v-fib/v-tach to PEA or asystole during the early phase of their resuscitation. This means the majority of these patients were far into the metabolic phase of their arrest at the time of enrollment to the study. And while our traditional resuscitative strategies are effective during the early stages following the onset of arrest, these very same strategies become decisively ineffective in the latter stages of arrest (2,3). And so, even if bedside US was capable of predicting which patients were clinically salvageable and which were not, our inability to intervene with any meaningful intervention limits its clinical utility. A real time view of a heart in its final grasps of life will not change the patient’s outcome if we provide no additional effective means of survival.

This post should not be viewed as an argument against the use of bedside US in cardiac arrest, but rather an intellectual examination of how to most effectively utilize it. The vast majority of patients who arrive to the Emergency Department without achieving ROSC are unlikely to be discharged from the hospital neurologically intact. Identifying sonographic proof of these final agonal contractions will not alter their prognoses. As such, our goal should not be to identify every cause of arrest that presents to the Emergency Department, but rather to recognize the rare few cases in which a meaningful intervention can be applied. These cases include: pericardial effusion, tension pneumothorax and pseudo-PEA, with the caveat that the majority of cardiac activity seen in this cohort is the result of an agonal rhythm rather than a patient in severe shock. Gaspari et al reported 34 patients (3.6% of the entire cohort) in arrest in which a pericardial effusion was identified. 15 of these patients had attempts to evacuate this potential cause of arrest. Survival to hospital discharge was observed in 2 (15.4%) of these patients.

As far as the utilization of US to help determine when to cease our resuscitative efforts, while not perfect, the finding of no cardiac activity on bedside US in combination with the universally poor outcomes of patients in refractory arrest, likely stratifies a patient’s chance of survival from extremely unlikely to dismally grim. The contrary cannot be said, cardiac motion on bedside US does not shift the likelihood of survival in any clinically meaningful fashion. I am sure we have all witnessed a code go far beyond the point of futility because the bedside US demonstrates some residual cardiac activity.

The Gaspari et al paper marks the first time a large high quality prospective data set has documented the findings of bedside US in a cohort of patients in refractory cardiac arrest. Perhaps when mechanical support becomes a more viable and commonplace alternative, bedside US will identify a cohort who will have a more favorable outcome. Or maybe it will play little prognostic role, as the major determinants of survival following extracorporeal support are minimizing downtime and early high-quality chest compressions(4,5). But I fear that until we accept the general disutility of our traditional resuscitative methods in patients with refractory cardiac arrest, bedside US will provide little more than a voyeuristic window into the final heaves of a dying heart.

Sources Cited:

Gaspari et al. Emergency Department. Point-of-care Ultrasound in Out-of-Hospital and in-ED Cardiac Arrest. Resuscitation.
Goto Y, Funada A, Goto Y. Relationship Between the Duration of Cardiopulmonary Resuscitation and Favorable Neurological Outcomes After Out-of-Hospital Cardiac Arrest: A Prospective, Nationwide, Population-Based Cohort Study. J Am Heart Assoc.2016;5(3): e002819.
Weisfeldt ML, Becker LB. Resuscitation after cardiac arrest: a 3-phase time-sensitive model. JAMA. 2002;288(23):3035-8.
Kim SJ, Jung JS, Park JH, Park JS, Hong YS, Lee SW. An optimal transition time to extracorporeal cardiopulmonary resuscitation for predicting good neurological outcome in patients with out-of-hospital cardiac arrest: a propensity-matched study. Crit Care. 2014;18(5):535.
Fagnoul D, Combes A, De backer D. Extracorporeal cardiopulmonary resuscitation. Curr Opin Crit Care. 2014;20(3):259-65.
http://caep.ca/sites/caep.ca/files/caep/CAEP2015/Presentations/reason_final_caep_version_ppt_2.pdf