iPhoneIcon_BigA paper in AJEM describes a way to quickly assess left ventricular function that I wasn’t too familiar with. By measuring the distance between the anterior mitral valve and interventricular septum we can roughly assess the heart’s ejection fraction.

Systolic heart failure (SHF) is for our purposes is inadequate emptying of the left ventricle, ie increased preload. With increasing preload in SHF the left ventricle dimensions will increase in both systole and diastole. The heart will appear more spherical in shape with thinner walls. The other hallmark feature of SHF is decreased ejection fraction where stroke volume is still maintained as the heart has compensated by dilating the left ventricle. In reality, SHF echo diagnosis is a lot more complex than this as there is valvular diseases, concomitant diastolic failure, anatomic variation and what not.

UnknownStill, volumetric assessment with very simple parameters is the mainstay diagnostic tool in SHF for most of us.

We estimate preload by diastolic measuring of for example the left ventricular end-diastolic dimension (LVEDD), by measuring the end-diastolic area of the ventricle (EDA) or by letting the US-machine do the maths using the Simpson biplane method. We then move on to measure the corresponding systolic counterparts of these measurements in order to calculate the ejection fraction. If ventricle dimensions are increased, if EF is normal or low and if it patient looks about right then it is SHF.

E-point septal separation
Screen Shot 2014-09-14 at 13.04.11E-point septal separation (EPSS) is the distance from the anterior mitral valve leafleft and the ventricular septum in early diastole.

The measurement is made in m-mode and is simply the closest the mitral valve gets to the septum in the cardiac cycle. In early diastole, the anterior mitral valve should approach or even touch the septum. In SHF, the ballooning heart with increased preload will pull valve away from the septum.

Like the LVEDD, the EPSS is a simple linear m-mode measurement obtained from the parasternal long axis view. As such is easy to obtain, very fast but prone to the same inaccuracies as any single linear measurement, local wall abnormalities and sloppy view angles.

The study
The authors enrolled eighty adult subjects were the treating physician had requested a transthoracic echo, regardless of the indication. Three emergency ultrasound fellows first made a visual estimate of the EF, then measured the EPSS and finally compared their estimates to a gold standard comprehensive cardiac ultrasound performed by cardiac ultrasonographers who estimated EF by using the Teicholz method.

Screen Shot 2014-09-14 at 13.22.42The EPSS measurements are plotted against the gold standard EF in the scattergram. The authors conclude there is a ‘strong correlation’ between increasing EPSS and decreasing EF from comprehensive TTE. The regression formula is in the scattergram.

In contrast, ‘eyeballing’ was not as accurate and interobserver agreement was not as good as EPSS nor as good as reported in previous studies.

Previous literature has recommended a cut-off EPSS of 7 mm to determine severe left ventricular dysfunction. (EF < 30%).  In this study 7 mm would be 100% sensitive for EF <30%.

Take-home message
Could be useful. Quick, easy to learn and relies only on a single parasternal view. Like all volumetric echocardiographic assessments it should be interpreted very cautiously. There’s also a lot more to cardiac failure and cardiogenic shock than EF. As always, the more extreme our measurements are, the more we can trust them. (And the less we need them…)

Paper lives here:
E-point septal separation: a bedside tool for emergency physician assessment of left ventricular ejection fraction. McKaigney CJ, Krantz MJ, La Rocque CL, Hurst ND, Buchanan MS, Kendall JL. Am J Emerg Med. 2014 Jun;32(6):493-7. doi: 10.1016/j.ajem.2014.01.045. Epub 2014 Feb 3.

This entry was posted in Cardiology, Emergency Medicine, Ultrasound. Bookmark the permalink.


  1. Andy Neill says:

    You mean you hadn’t seen the Ultraosund podcast on it



  2. K says:

    Thank you Andy. I missed that. Yes I am ashamed!

  3. Erik says:

    Always interesting with posts about ultrasound!
    I was not familiar with EPSS before reading your post and I took the time to discuss it with our excellent echo technician who teaches ultrasound in the ICU where I work. Unfortunately she was not impressed.
    Basically what EPSS shows is that the heart is big which is common in low EF and you can also see by measuring the size of the heart.
    Quite naturally they would get a correlation between EPSS and EF in the study since they are comparing it to EF derived from Teicholz method which I think most people who do echocardiography consider far from golden standard. It would have been very interesting if they compared it and got a correlation with cardiac MRI for example.

    • K says:

      Hi Erik,
      Thanks for your comment.

      You are absolutely right. EPSS is just a variation of any of the one dimensional measurements of the left ventricle and it will obviously suffer the same inaccuracies.

      I am also not sure using the Teicholz method for a gold standard was the right thing to do. Maybe even a bit dishonest. Like you say its far from the gold standard in volume measurement.

      As far as I can remember the LVEDD is is part of the Teichholz calculation. LVEDD is a measure very similar to EPSS, measured in the same axis in almost the same location. Its like including the EPSS into the Teichholz EF formula.

      Of course there is going to be a correlation between EPSS and a EF based on EPSS.

      I am very tired, coming off shift, and I hope this makes sense.

  4. Anthony paul Raj says:

    Hi guys,
    While it is not much difference of both the teicholz and EPSS method, one must always follow the AHA guidelines if possible.

  5. Pingback: My first shift…

Leave a Reply

Your email address will not be published. Required fields are marked *