April 2024

In this VETgirl online veterinary continuing education blog, Dr. Christopher Kennedy, DACVECC, DECVECC discusses objective assessment of left ventricular systolic function on Focused Cardiac Ultrasound (FCU) in the dog and cat.

By Dr. Christopher Kennedy, DACVECC, DECVECC

Objective Assessment of Left Ventricular Systolic Function on Focused Cardiac Ultrasound

The goals of this blog are:

1. Image the left ventricle from multiple angles.
2. Identify normal and reduced systolic function.
3. Measure fractional shortening and fractional area change.
4. Understand ejection fraction.

The views:

  • PLAX4
  • PSAX-pap

Objectively assessing the left ventricle (LV)
Last time we discussed LV systolic function. Recall, we are seeing the LV contraction, which is dependent on loading conditions (preload, afterload), intrinsic contractility (unmeasurable) and heart rate. We are not assessing cardiac performance nor contractility.

Objective measurements of the LV can be performed using ultrasound machines. These parameters have all been investigated in cardiology clinics, typically in relatively stable patients without the minute-to-minute changes we may see in critical care. As such, applying these parameters to critically ill patients should be done cautiously.

Fractional Shortening (FS)
FS is the proportional difference between a linear dimension of the LV in diastole and systole. It is calculated using the equation:

FS = LVIDd – LVIDs / LVIDd x 100

Where LVIDd and LVIDs are the LV internal diameters in diastole and systole. It may be measured in 2d (B-mode) or M-mode via the PLAX4 or the PSAX-pap views (See Figures 1 & 2). Normal canine values are often between 25 – 45%, though values between 20 – 50% may also be acceptable.(1,2) Ideally, breed-specific references would be consulted, as specific breeds – e.g., Whippets, Greyhounds, Dobermans, Great Danes – may have lower reference ranges.(2)  The lower limit measured in PLAX4 may also be lower (19%).(1)  Athletic dogs may have reduced FS and often their LV might appear mildly dilated.(2) For cats, the lower limit is around 30% and commonly FS > 40% is recorded.(2)

Figure 1a: PLAX4 view in B-mode showing measurement of the left ventricular internal diameter in diastole (LVIDd) and in systole (LVIDs). Photo courtesy of Dr. Christopher Kennedy, DACVECC, DECVECC


Figure 1b: In M-mode, the LVIDd is traditionally measured at the peak of the Q-wave on the ECG; however, in this example (or when an ECG is not available), the LVIDd is measured at the widest point and LVIDs at the narrowest point. The timing of the LVIDd measurement should be specified in the medical record. Photo courtesy of Dr. Christopher Kennedy, DACVECC, DECVECC


Figure 2a: PSAX-pap view in B-mode showing measurement of the left ventricular internal diameter in diastole (LVIDd) and in systole (LVIDs). Photo courtesy of Dr. Christopher Kennedy, DACVECC, DECVECC


Figure 2b: PSAX-pap view in M-mode showing measurement of the left ventricular internal diameter in diastole (LVIDd) and in systole (LVIDs). Photo courtesy of Dr. Christopher Kennedy, DACVECC, DECVECC

Unpublished data from our lab investigated FS in dogs.† In PLAX4 and PSAX-pap, in both B- and M-mode, FS performed by two non-cardiologists had relatively poor agreement with the cardiologist. Our more experienced non-cardiologist (me) had less variability when measuring the FS twice in the same patient, though in PLAX4 B-mode that variability was clinically unacceptable; our less experienced non-cardiologist had unacceptable variability for all FS measurements. When performed by two cardiologists, the inter-operator variability was high and the authors cautioned against over-interpreting differences in FS < 18%! (3)

FS suffers from several other limitations. It requires appropriate image acquisition, including alignment of the LV and image quality. It also requires appropriate positioning of the calliper or M-mode sample line, which, if not correctly placed can lead to huge under or over-estimation (See Figure 3). Given there are two measurements (LVIDd and LVIDs), there is double the chance of measurement error, which can be compounded by the equation. In patients with arrhythmias, bundle branch blocks and regional wall abnormalities, the FS may not be an accurate measure of systolic function. Finally, as the FS considers a single linear slice of the LV, it provides very limited information on the overall function.

Figure 3: PLAX4 view showing the potential to over or underestimate the left ventricular internal diameter. The pink line is mostly correct (it has been shifted slightly apically to increase visibility but is typically measured further toward the base). The red line terminates early, as the papillary muscle has contaminated the view, resulting in under-estimation. The green line is at an incorrect angle, resulting in over-estimation. Photo courtesy of Dr. Christopher Kennedy, DACVECC, DECVECC

Fractional area change (FAC)
Instead of a single linear measurement, for FAC you outline the inner surface of the LV at the blood-endothelial interface in PSAX-pap (See Figure 4). The papillary muscles should be excluded. This gives you the LV shortening in multiple minor dimensions, though remains over only a single slice of the heart. It may be advantages when the M-mode alignment isn’t optimal or when there is dyssynchrony between the ventricular septum and the LV free wall.(2)  Normal values are approximately > 35% in dogs. (2)

Figure 4: PSAX-pap with the area measured in both diastole and systole. The papillary muscles have been excluded. Notice that, particularly in systole, the blood-endothelial interface can be unclear, which can result in errors. Importantly, the FAC is > 60%, which is well in excess of 35%, so for FCU we can conclude that there is no apparent systolic dysfunction. Photo courtesy of Dr. Christopher Kennedy, DACVECC, DECVECC

Ejection fraction (EF)
Ejection fraction is typically measured using the method of discs (See Figure 5). (2) In FCU, it is measured via the PLAX4 ensuring complete inclusion of the LV. The LV cavity at the blood-endothelial interface is outlined in diastole and then systole for the same heartbeat. The machine divides the area into a series of discs, calculates the estimated volumes, and adds them together. The ejection fraction is obtained by the formula:


Figure 5a: PLAX4 showing the measurement of left ventricular end diastolic and systolic volumes. Notice how the blood-endothelial interface can be unclear, particularly in systole, which can lead to errors. This is common with images obtained by non-cardiologists. However, the EF is > 60%, so it is unlikely that there is systolic dysfunction present. Photo courtesy of Dr. Christopher Kennedy, DACVECC, DECVECC

Figure 5b: PLAX4 showing the same measurements but there is a papillary muscle contaminating the view (outlined in green). This is a common problem: if we try to exclude the papillary muscle, it can result in over-estimation of the volume (slight over-estimation is seen here in systole); if we include the muscle it results in marked under-estimation of the volume. Photo courtesy of Dr. Christopher Kennedy, DACVECC, DECVECC

Where LVEDV and LVESV are the end-diastolic and end-systolic volumes. Normal canine values are > 45%. (1)  There are many benefits to this methodology. Although it technically considers only a single plane – in human medicine they use a bi-plane method – it covers a more relevant section of the heart, considering shortening along its length, rather than just in one horizontal slice. It also provides volumes, which overcomes one of the limitations of other methods of LV systolic function assessment in FCU, as we can derive stroke volume (SV = LVEDV – LVESV). With SV, we can derive CO. We can also consider the volumes individually: as a rough guide, LVEDV should be < 3 ml/Kg and LVESV < 1.5 ml/Kg, and values greater than this suggest systolic dysfunction and/or LV volume overload. (1)

So, using the method of discs approach, we can consider LV systolic function (EF), LV volume status (LVEDV, LVESV) and cardiac output ([LVEDV – LVESV] x HR).

The most important limitations are acquisition error and measurement error. Obtaining the LV perfectly is difficult in critical patients. Off-angle images, images invaded by the papillary muscles and foreshortening (under-displaying the length of the LV) are common. Often the LV apex is un-imaged, the blood-endothelial interface is hazy, or the papillary muscles confuse the boundaries. Try sliding the probe ventrally, work on maximizing the LV length, remove the papillary muscles from the sector and be sure you can see the opening mitral valve. Try turning out the lights, turn down the overall gain and perform at least three measurements. Sometimes, you just cannot appropriately image or measure the LV, in which case the EF and LV volumes should not be used. When in doubt, ask a cardiologist for help.

Worth noting, LV dilation may be associated with improved survival in sepsis and EF is not associated with survival in human sepsis.(4,5)  This underscores the need for clinical integration: it is not sufficient to perform FCU alone.

References and further reading:
1. Visser LC, Ciccozzi MM, Sintov DJ, Sharpe AN. Echocardiographic quantitation of left heart size and function in 122 healthy dogs: A prospective study proposing reference intervals and assessing repeatability. J Vet Intern Med. 2019 Sep;33(5):1909-1920. doi: 10.1111/jvim.15562.
2. Echocardiography. In: Cardiovascular Disease in Companion Animals Dog, Cat, and Horse. Ware WA and Bonagura JD (editors), 2nd edition. Boca Raton, Florida, USA; pp. 95-106.
3. Dukes-McEwan J, French AT, Corcoran BM. Doppler echocardiography in the dog: measurement variability and reproducibility. Vet Radiol Ultrasound. 2002 Mar-Apr;43(2):144-52. doi: 10.1111/j.1740-8261.2002.tb01662.x.
4. Zanotti Cavazzoni SL, Guglielmi M, Parrillo JE, Walker T, Dellinger RP, Hollenberg SM. Ventricular dilation is associated with improved cardiovascular performance and survival in sepsis. Chest. 2010 Oct;138(4):848-55. doi: 10.1378/chest.09-1086.
5. Boissier F, Aissaoui N. Septic cardiomyopathy: Diagnosis and management. J Intensive Med. 2021 Dec 27;2(1):8-16. doi: 10.1016/j.jointm.2021.11.004.

† Unpublished data from my on-going PhD.

Only VETgirl members can leave comments. Sign In or Join VETgirl now!