Depending on the origin of CS and initial state of the cardiac function, TTE can be used for tracking LVEF evolution. Visual assessment by a simple eyeballing of LVEF is considered to be reliable in Cardiovascular Intensive Care Unit, when used by trained practitioners2 (Videos 1, 2).

Figure 1.

Left ventricle function assessment: left ventricular ejection fraction calculated using Biplan Simpson method (A), and global longitudinal strain based on apical 4-, 2-,and 3-chamber view.

(0.44MB).

Additionally, LVEF can be measured by the Simpson’s Biplane Formula,4 requiring area tracing of left ventricle (LV) cavity and contouring the endocardial border in both the apical four-chamber and two-chamber views in end-diastole and end-systole. LV is considered to have the shape of a cone. Area tracings of the LV cavity divide it into a number of discs (usually 20) and the total of volume of these discs is equal to LV volume. The difference between diastolic and systolic disc volumes divided by the diastolic volume gives LVEF value. In other words:

Other methods for LVEF measurement such as 3D echocardiography is more accurate but not usable in routine.

• •

  Mitral annular plane systolic excursion (MAPSE) (Fig. 2A and B):

  
  

  Figure 2.

  Left ventricular ejection fraction surrogates: mitral annulus plane systolic excursion (MAPSE) (A, B), and Tissue Doppler peak systolic wave at the mitral lateral annulus (S′) (C, D).

  (0.56MB).

It is measured by the use of M-mode echocardiography from four sites of the atrioventricular plane corresponding to the septal, lateral, anterior, and posterior walls using the apical four- and two-chamber views by M-mode echocardiography. The M-mode cursor should always be aligned parallel to the LV walls. The systolic excursion of mitral annulus should be measured from the lowest point at end-diastole to aortic valve closure (end of the T-wave on the electrocardiogram). MAPSE represents the amount of displacement of the mitral annular plane towards the apex and thus assesses the global change in size of the LV cavity (in the long-axis direction). The average normal value of MAPSE derived from previous studies for the four annular regions (septal, anterior, lateral, and posterior) ranged between 12 and 15 mm5,6 and a value of 8 mm was associated with a depressed LVEF (<50%) with a specificity of 82% and a sensitivity of 98%.5 In addition, a mean value for MAPSE of 7 mm could be used to detect an EF < 30% with a sensitivity of 92% and a specificity of 67% in dilated cardiomyopathy patients with severe congestive heart failure.6 It is of note that the association between MAPSE and EF is only valid in case of normal or dilated left ventricles7,8 while the correlation is rather poor in patients with LV hypertrophy.9

• •

  Tissue Doppler peak systolic wave at the mitral lateral annulus (S′) (Fig. 2C and D):

Tissue Doppler imaging enables measurements of atrioventricular annular and regional myocardial velocities, and may be more sensitive than conventional echocardiography in detecting abnormalities of LV systolic and diastolic function.10 Two previous studies showed that there was a close correlation between systolic annular displacement directly measured by M-mode and that indirectly estimated by temporal integration of velocities measured by either pulsed tissue Doppler or colour Doppler in healthy subjects.11,12 Similar results were reported, showing a significant correlation between S′. and MAPSE both at rest and during exercise in heart failure patients with preserved LVEF13 S′ value >10 cm/s is correlated to preserved LVEF, 6–8 cm/s corresponds to altered LVEF between 30 and 45%, and S′ value <6 cm/s is associated with LVEF < 30%.14

Even if systolic ventricular function estimated by LVEF is one of the strongest predictors of total and cardiovascular mortality, assessment of regional wall motion is part of visual echocardiographic examination. Wall motion is assessed in each 17 segments of the LV and LV segments can be akinetic, hypokinetic or dyskinetic, that may be due to a chronic or acute coronary disease.

More precise methods have been developed during the past decades for a better quantification of global and regional myocardial function, as the Strain, Strain Rate and Speckle Tracking (Fig. 1B). These methods can track the motion and the deformation of the myocardium during systole and diastole and point out regional wall motion abnormalities (RWMA) that are not visible on visual echocardiography. LV Global Longitudinal Strain alteration precedes the LVEF one and was demonstrated to be strongly correlated to mortality.4

Measurement of CO remains a corner- stone in the hemodynamic assessment of critically ill patients and in particular in CS patients as decreased CO is often observed in such population. Several methods for determining CO have been described using both two-dimensional and Doppler echocardiography.15,16 Of these methods, the one using the left ventricular outflow tract (LVOT) and aortic valve as the conduit, is probably the most reliable and most commonly used as there is an excellent agreement with the reference CO measured by thermodilution in most situations 15 (Fig. 3B and C). The measurement of stroke volume (SV) is usually made at the LVOT. When using the TTE approach, the operator measures the LVOT diameter from the parasternal long-axis view immediately below the hinge point of the aortic valve leaflets (Fig. 3A). The LVOT area (cm2) is calculated from this diameter measurement using the formula:

Figure 3.

Measurement of the left ventricular outflow tract in parasternal long axis view (A), and velocity time integral (VTO) in left ventricle outflow tract in 5-chamber apical view using pulsed wave Doppler (B, C).

(0.42MB).

Next, the operator places the pulsed wave Doppler (PWD) sample volume in the LVOT to measure the velocity time integral (VTI) of blood flow in the LVOT, using the five-chamber apical view. The SV is calculated as follows:

Leads to right ventricle collapse and decrease of RV output and by consequence LV output. From a subcostal view, we can assess the presence of pericardial effusion, which compromises the functionality of the heart. In a basic analysis of shock, the existence of severe effusion (>2 cm), collapse of the cavities in their respective diastoles, dilation of the Inferior Vena Cava with absence of respiratory variations and in some situations, visualisation on the two-dimensional TTE of “Swinging heart” which is associated to a large pericardial effusion testifies often cardiac tamponade.17

Doppler assessment provides unique information regarding haemodynamic of pericardial tamponade. the following Doppler features are observed during inspiration: in the left heart, there will be a reduction in effective filling gradient (EFG) of the LV (pressure difference between pulmonary capillaries and left ventricle) due to a reduction in pulmonary capillary pressure while left atrium (LA) and LV diastolic pressures are relatively maintained due to reduced transmission of intrathoracic pressure into the heart. Therefore, LV filling will be reduced18 and consequently, the transmittal Doppler early diastolic (E)-wave and in turn LV outflow will be reduced. In the right heart, the opposite is observed; RV filling is increased with increased RV volume as the septum moves to the left (ventricular interdependence), increased tricuspid E-wave and increased RV outflow velocity.

In critically ill patients, however, mechanical ventilation, bronchospasm, significant pleural effusion, respiratory distress, and arrhythmias make the Doppler findings difficult to interpret.

TTE can help to establish a prompt diagnosis of acute pulmonary embolism (PE) and to identify patients with high-risk features. Additionally, when the patient is hemodynamically unstable, TTE may be the only immediately available and appropriate imaging investigation. 19

Using a basic approach, the cause of the shock towards PE when we observe the evidence of hyperechogenic images in the right cavities, in this context, has a high specificity of PE. Additionally, signs of the consequences of acutely increased pulmonary artery/right heart pressures can be observed including dilatation of right heart chambers and more precisely the evolution of an initial abnormal ratio of RV diameter or area to LV diameter or area (Fig. 4, Video 3).

Figure 4.

Indices of right ventricle dilation and/or pulmonary hypertension: parasternal short-axis view at the mid-ventricular level illustrating measurements of left ventricular diameters for calculation of the eccentricity index (A), and assessment of septal motion (C), right and left ventricle basal diameter ratio in apical 4-chamber view (B), peak velocity of tricuspid regurgitation (TR Vmax) obtained in apical 4-chamber view with continuous wave Doppler through tricuspid valve (D).

(0.51MB).

Tricuspid regurgitation is frequent in patients with intermediate-to-high-risk pulmonary embolism. It allows the estimation of RV systolic pressure and thus of pulmonary arterial systolic pressure (PAsP) in the absence of pulmonary valve stenosis. PAsP can be estimated from the peak velocity of the tricuspid regurgitation (TR) jet (V) according to the simplified Bernoulli equation but may underestimate it when tricuspid regurgitation is very severe (Fig. 4D).

In the absence of perceptible TR or inadequate alignment of PWD, Pulmonary Regurgitation (PR) from the parasternal short-axis view is usable to estimate pulmonary artery diastolic and mean pressures (PAdP and PAmP). The measurement of RVOT VTI can estimate RV output which is associated with increased pulmonary embolism-related mortality when it is low.19 Moreover, a decreased Pulmonary artery acceleration time and the presence of a “notch” on the RVOT VTI are valid signs of pulmonary hypertension (60/60 sign: right ventricular ejection acceleration time <60 ms with peak systolic gradient of tricuspid regurgitation <60 mmHg; a mid-systolic notch).20,21

A major pathological contribution to shock in sepsis is peripheral vasoplegia and although this is not measurable with echo, the cardiac findings can be taken into account when estimating it.

For example, in shock a hyperdynamic, after optimal fluid resuscitation, usually a clue to the presence of marked peripheral vasodilatation.

Furthermore, the absence of elevation in LV filling pressure has been reported as a specific characteristic of this hemodynamic profile, not only when evaluated by the E/e’ but also when measured in the past using a pulmonary artery catheter.23,24 It was suggested to be related to an increase in LV compliance due to sepsis.25

It is constant during sepsis and no need for a diagnostic tool at the early phase. Nevertheless, after the initial phase, optimization of fluid therapy using dynamic parameters is mandatory as suggested by the most recent recommendations.22 This will be detailed below.

Severe sepsis is frequently associated with cardiopulmonary dysfunction driven by a cascade of cellular and molecular processes.26 Myocardial dysfunction occurs frequently, early and involves both ventricles.26,27 Parker et al. were the first to describe LV hypokinesis in septic shock.23

They reported that survivors manifested severely depressed LVEF but that adequate LV stroke output was maintained as a result of acute LV dilation.28 LVEF might not be a reliable index of LV systolic function in patients with early septic shock, as this is a state characterized by low systemic vascular resistance that unloads the LV.24 Therefore, normal or supra normal EF in early sepsis might lead clinicians to make the wrong inference about cardiac reserve.

Speckle tracking echocardiography (STE) is a relatively novel and sensitive method for assessing ventricular function and may unmask myocardial dysfunction not appreciated with conventional echocardiography.29 STE may unmask systolic dysfunction not seen with conventional echocardiography. RV dysfunction unmasked by STE, especially when severe, was associated with high mortality in patients with severe sepsis or septic shock.

LVEF improvement after revascularization for acute myocardial infarction or inotropic support introduction for cardiogenic shock for instance should be followed up. However, this need to be interpreted within other hemodynamic parameters in case of shock for an overall assessment of tissue perfusion. Other parameters can be also reassessed after revascularisation as the MAPSE and the S′, in order to evaluate the efficacy of the revascularisation.

Additionally, and in the case of acute coronary disease, it is important to follow up regional wall motion abnormalities (RWMA) evolution after coronary revascularization. The quantification of global and regional myocardial function, as the Strain, Strain Rate and Speckle Tracking can be used also to follow the RWMA after revascularization.

Assessment of CO is important not only to identify the type of shock in particular cardiogenic shock together with other parameters, but also to evaluate the response to medical and surgical interventions, such as administration of inotropic agents for the treatment of right and left heart failure. Indeed, CO is SV multiplied by heart rate. The measurement of SV is usually made at the LVOT.

The VTI, SV or CO can be serially measured noninvasively before and after medical therapies in order to evaluate their effects, all the three variables are interchangeable each can be used as a sole parameter.

In addition, it is to be noticed that Veno-Arterial Extracorporeal Membrane Oxygenation (VA-ECMO) is now taking part in cardiogenic shock resuscitation and treatment, its management may be guided by TTE.32

First, adequate venous canula position in right atrium can be assessed by TTE. Secondly, CO is often diminished after VA-ECMO implantation because of competitive cardiac and assistance flows this may be easily detected at the bedside by TTE. Furthermore, a direct visualisation by two dimensional TTE of the aortic valve may show that it remains closed indicating an urgent LV unloading. The indication of the maintenance of VA-ECMO, the need of further concomitant mechanical circulatory support devices or the detection of complications such as pericardial effusion or intracavitary/valves thrombosis need to be daily reassessed by repeating TTE.

Finally, TTE is now a corner stone of the VA-ECMO weaning protocol, by assessing LVEF, LVOT VTI, S′ and RV function under VA-ECMO and during weaning protocol.33

The main limitations of echocardiographic measurements of SV, CO, and VTI in the LVOT are that all of them, require accurate alignment with the LVOT, and consistent sampling that should occur just beneath the aortic valve. The use of an LVOT diameter adds a second potentially more significant error measurement. It is now recommended to use the stroke distance (i.e., LVOT and RVOT VTI) alone for serial measurements after therapeutic interventions, with the assumption that LVOT and RVOT diameters remain constant.

In clinical practice, only LVOT VTI is measured, considering LVOT to be constant and heart rate to be in stable range. The increase in LVOT VTI reflects CO improvement and myocardial contractile reserve.

In the absence of intracardiac shunt, LV output is equal to RV output. The latter can be estimated by measuring RVOT diameter from the parasternal short-axis view and RVOT VTI. Like LVOT VTI, RVOT VTI is obtained by placing PWD with a correct alignment in the RVOT. Inadequacy between LV output and RV output can be the sign of an atrial septal defect or a ventricular septal defect. In these cases, and without major pulmonary hypertension, left to right shunts lead to an increased RV output and a decreased LV output and by consequence haemodynamic instability.

It has to be noticed that RV output can be measured through the modified subcostal window, enhancing monitoring, especially in mechanically ventilated patients.34

Although invasive methods are considered as the “gold standard” for measuring intracardiac filling pressures, echocardiography is routinely used as a non-invasive alternative.35 This has been achieved using an algorithm based on LVEF status (altered or preserved), Doppler-derived parameters from mitral inflow velocities (E- and A-peak wave velocities, E/A ratio, E velocity DT) and tissue Doppler-derived mitral annular (e’-peak wave and E/e’ ratio).

It has to be underlined that LV filling pressures should not be considered a part of the clinical context, ventilation mode and other echocardiographic data such as LVEF.

Frequently, during cardiogenic shock, filling pressures are high, the improvement of cardiac function is often associated to their decrease.

Leads to right ventricle collapse and decrease of RV output and by consequence LV output. Acute cardiac tamponade with hemodynamic compromise requires urgent pericardiocentesis or surgical removal of pericardial fluid.36

As pericardial tamponade treatment is mostly procedural rather than medical, echocardiography identifies the optimal site for pericardiocentesis by visualizing the location and distribution of pericardial effusion. The para-apical site is the most common entry site for pericardiocentesis and procedural rate is around 95%.17 Some may also recommend injecting agitated saline solution through the pericardiocentesis needle in the pericardial effusion to avoid the puncture of the ventricular cavity.37,38

TTE can help to establish a prompt diagnosis of acute pulmonary embolism and to identify patients with high-risk features. Additionally, when the patient is hemodynamically unstable, TTE may be the only immediately available and appropriate imaging investigation.19,39

Indeed, echocardiography plays a determinant role in making therapeutic decisions in shock patients as it may help to rule out the diagnosis of severe pulmonary embolism in the absence of acute core pumonale. The main findings in acute pulmonary embolism are the consequences of acutely increased pulmonary artery/right heart pressures, these parameters simply assessed are used also to evaluate the efficacy of the treatment in particular thrombolysis.

Rapid decrease in PAsP reflects adequate dissolution of the thrombus. Finally, the improvement of the RV output and the RV function can be helpful to monitor the evolution of the thrombus.

Although during septic shock, patients frequently present with hypovolemia, beyond the very initial phase, an increase in CO after fluid administration is observed in only 50% of the patients.40 Moreover, fluid overload is now widely admitted to be an independent predictor of mortality.41 In this regard, recent guidelines recommend the use of dynamic parameters for the assessment of fluid responsiveness rather than static parameters.22 Preload responsiveness can be assessed by measuring the response of VTI LVOT through dynamic tests as passive leg raising or to a combination of end-expiratory and end-inspiratory occlusions in patients under mechanical ventilation.42

Adequate TTE haemodynamic evaluation should be made regarding ventilation status (spontaneous breathing or mechanical ventilation) and cardiac rhythm (sinusal or not).

Increase in LVOT VTI of >12.5% during passive leg raising predicted the increases in SV in response to intravenous fluids with spontaneously breathing activity.43 In mechanically ventilated patients, recruitment manoeuvres can change cardiac loading conditions and decrease cardiac preload. For instance, the decrease in stroke volume during a recruitment manoeuvre predicted fluid bolus responsiveness in surgical patients during anaesthesia.44,45 Consecutive end-inspiratory occlusion and end-expiratory occlusion change VTI ≥ 13% in total predicted fluid responsiveness more accurately with less inter-observer variability.42

Among various indices, the assessment of respiratory variation of the diameter of the inferior vena cava (IVC min and IVC max) has received growing interest since it can be easily using two dimensional echocardiography in most critically-ill patients.46 However, there are some main concerns for the use of this parameter to predict preload responsiveness in critically ill patients.

Firstly, it has been demonstrated that neither the IVC diameter nor IVC variability accurately predict fluid responsiveness in spontaneously breathing in critically ill patients.47 Secondly, even in mechanically ventilated patients (one of the largest published series of ventilated patients) assessed using advanced critical care echocardiography for any type of acute circulatory failure. IVC variability had a low diagnostic accuracy to predict preload responsiveness with AUC of 0.608.48

Initial assessment of MAPSE, speckle tracking and global longitudinal strain of LV and LVEF in 2D-mode may play a role in septic cardiomyopathy prognostication.29 Feng and al. reported in their analysis of MIMIC-III database that early use of TTE in septic shock had a significant benefit in terms of 28-day mortality, with more fluids, administered during the first day, greater use of dobutamine and a trend to be more quickly weaned from vasopressors.49

Furthermore, and as mentioned above during cardiogenic shock: LVOT VTI and LVEF may be used to evaluate the effectiveness of inotropes in case of cardiac dysfunction.