Bedside echocardiography findings in carbon monoxide-poisoned patients

April 28, 2016, 9:34 pm


Incidence and patterns of cardiomyopathy in carbon monoxide-poisoned patients with myocardial injury. Cha YS et al. Clin Toxicol 2016 Apr 11 [Epub ahead of print]


At last year’s Social Media and Critical Care (SMACC) conference in Chicago, I gave a talk remarking on how bedside ultrasound imaging in critically ill toxicology patients is underused and little studied. I suggested that it could provide crucial information in a number of settings. For example, visualizing the inferior vena cava (IVC) in salicylate toxicity to help guide rehydration, or evaluating left ventricular (LV) function to distinguish myocardial dysfunction from vasodilation in the unstable hypotensive patient with calcium channel blocker overdose.

I suggested that a simple, two part protocol looking at the IVC and left ventricle in these patients was an idea whose time had come. Building on the rapid adoption of the FAST and RUSH exams, I suggested that this protocol should be called Toxicologic Ultrasound in SHock — the TUSH exam. (Pro tip: do not search this on Google, especially at work.)

Therefore, I was delighted to see this Korean study and the accompanying editorial. The authors did serial transthoracic echocardiograms on 43 consecutive patients with carbon monoxide poisoning and laboratory evidence of myocardial injury (elevated high-sensitivity troponin I.) The echocardiograms were read by 2 cardiologists.

Eleven of the 43 patients had normal LV systolic function (ejection fraction > 50%), 22 patients had global LV dysfunction and 10 had regional wall motion abnormality not reflecting coronary artery distribution but resembling Takotsubo cardiomyopathy.  The patients with global LV dysfunction recovered normal LV function more rapidly than those with findings suggestive of Takotsubo syndrome.

The authors list a number of limitations inherent in this study, including the small number of patients, the exclusion of patients who had normal high-sensitivity troponin I levels, and the inability of this study design to evaluate if and how the echo findings should impact clinical management.* At this point we really don’t know how to use these findings clinically, although the authors state that they are planning to study that question.

To be sure, this is a very preliminary study. However, the short accompanying commentary “The echoes of intoxication” by Gallic et al is a must-read. The authors emphasize two potential benefits of bedside echocardiography in evaluating critical poisoned patients:

“The first is the recognition that there exists a plurality of clinical manifestations of systemic insult, and that the cardiac manifestations that often contribute to hemodynamic instability are well identified with echocardiography. The second is the demonstration that beyond identifying pathology, echocardiography can be performed at the bedside to characterize the severity of cardiac dysfunction in acutely poisoned patients and directly impact the clinical management of hemodynamically unstable patients. The physician can quickly make rational decisions about intravenous fluid administration and vasopressors with information from echocardiography in addition to obtaining information about the patient’s likely disease progression . . .”

I could not agree more.

*This post originally stated, erroneously, that patients who received hyperbaric oxygen therapy were excluded. In fact, 33 of the 43 patients studied received HBOT.





  1. Willie Eggleston Says:

    Great review and hopefully another indicator that US will continue to permeate the tox world! Just wanted to point out the group didn’t exclude patients who received HBOT and about 3/4 of the patients included in the study group got HBOT (table 2). They did note they did not perform HBOT in patients requiring intubation due to monitoring issues (not sure how many patients this included as the paper doesn’t explicitly state why the 10 patients didn’t receive HBOT) and weren’t able to draw any conclusions on the impact of HBOT on CO-induced MI. While there is still a hefty list of limitations I agree that the findings are promising and look forward to the group’s future work.

  2. Leon Gussow Says:


    Thank you for the comment. You are correct — I misread the data. But this raises some other questions. The authors state in their ‘Limitations’ section:

    ” . . . we did not perform HBOT ins cases where patients required intubation because it was not possible to monitor patients inside the HBOT chamber. Therefore, we could not determine whether HBOT had an impact on CO-induced cardiomyopathy.”

    This doesn’t make much sense to me. 33 or the 43 patients received HBOT. The reason the authors couldn’t determine impact is because of the small numbers and the fact that the study was not designed to evaluate the effect of therapy. Also, unless I missed it, I could not figure out how many of their 43 patients were in fact intubated.