The Meaning of Hemodynamic Monitoring in Patients with Shock: Role of Echocardiography

with Shock: Role of Echocardiography

A. Vieillard-Baron

Introduction

The development of noninvasive devices to manage hemodynamics in patients with shock is directly prompted by the results of recent studies in the intensive care unit (ICU), and during the perioperative period, which demonstrated the inability of an invasive approach, based on right heart catheterization, to improve prognosis [1].

Some authors have suggested that these results were largely due to inaccurate use of right heart catheterization, without clear goals or protocol [2]. However, previous studies have demonstrated that optimization of cardiac output and mixed venous oxygen saturation (SvO

) with clear endpoints also fails to improve prognosis [3].

So, the lack of efficacy is inherent in the device. In 1985, Eugene Robin suggested that using right heart catheterization in patients with shock led physicians to give fluids plus diuretics whatever the wedge pressure [4]. In 2003, Fran¸cois Jardin claimed that we were going to move from the “age of oil lamps” to the “age of elec- tricity” [5]. In fact, we are going to change our practices in the management of shock, from an invasive and quantitative approach of hemodynamics to a non-inva- sive one, more functional and especially qualitative, mainly thanks to the use of echocardiography. This leads us to think about the meaning of hemodynamic moni- toring.

What is Monitoring?

Monitoring should be a diagnostic aid and guide treatment. So, in hemodynamics,

monitoring should help us to determine the cause of shock, i.e., hypovolemia, left or

right ventricular failure, vasoplegia, pericardial effusion, and so guide the treatment,

i.e., infusion of fluids, inotropic drugs, vasoconstrictive drugs, or fluid removal. This

is especially true in septic shock, where most of these causes can be associated,

making it essential to have a monitoring device capable of assessing all of them

independently. Martin Tobin also emphasizes that ‘good’ monitoring should measure

relevant variables, provide interpretable data, be easy to implement, and not cause

harm (M. Tobin, post-graduate course on ICU monitoring, Congress of the Ameri-

can Thoracic Society, San Diego, 2006). Once again, in hemodynamics, this does not

seem to correspond to an invasive approach, but rather to a non-invasive one using

echocardiography.

sion pressure (PAOP) as high as 18 and 17 mmHg, respectively, could be associated with a fluid-responsive status in shock patients. Cardiac index (CI), a parameter commonly measured by right heart catheterization, is frequently in a normal range at baseline before volume expansion in true hypovolemic patients receiving mechan- ical ventilation [9], and frequently not significantly different between patients who do and do not respond to fluids [8]. However, it is true that coupling CI and PAOP variations following a fluid challenge could be used to assess fluid requirement. An increase in CI, associated with a slight increase in PAOP, suggests the presence of hypovolemia, whereas the absence of a significant increase in CI, associated with a marked increase in PAOP, demonstrates the uselessness of volume expansion. But, in this situation, physicians risk the deleterious effects of a useless volume expansion several times a day, such as cardiac overload, pulmonary edema, and impairment in oxygenation.

Recently, echocardiography has been reported to accurately identify patients who need fluids, providing the errors made using an invasive approach are not repeated.

This means not evaluating cardiac filling pressures, as previously proposed [10], but using alterations in some cardiac function parameters induced by tidal ventilation [11]. The best is probably to examine the venae cavae and their respiratory diameter variations [12]. Whereas the superior vena cava, visualized by a transesophageal approach, partially or totally collapses at each insufflation in the case of hypovole- mia [8], the inferior vena cava, visualized by a subcostal approach, significantly increases in diameter (Fig. 2) [13]. From the concept of ‘fluid challenge’, we now pass to the concept of ‘fluid responsiveness’, which is totally adapted to the use of echocardiography.

Fig. 1. Maximum and minimum

values of central venous pressure

(CVP) and pulmonary artery occlu-

sion pressure (PAOP) in patients

who responded to fluids, as reported

in three recent studies in critically

ill patients [6 – 8].

a

b

Fig. 2. Panel a represents a cyclic collapse of the superior vena cava (SVC) at each insufflation in a hypovo- lemic patient. Panel b represents significant increase in inferior vena cava (IVC) diameter at each insuffla- tion in another hypovolemic patient. TP: tracheal pressure.

How to ‘Monitor’ Left Ventricular Failure?

Using right heart catheterization, left ventricular (LV) systolic dysfunction is also

classically assessed by the comparison between PAOP and CI. LV failure is diagnosed

when a high PAOP is associated with a low CI. Whereas this is true in very simple

clinical situations, such as pure cardiological situations, it is not relevant in more

complex ones, as in the ICU. For example, LV failure is common in septic shock, and

may frequently require infusion of an inotropic drug [14]. But this failure is associ-

tic shock. In most of these patients, this relation- ship suggested hypovolemia (open circle) at base- line, although most of them were fluid non- responders, as demonstrated by the first step (first closed circle) and the second step (second closed circle) of volume expansion. Finally, dobu- tamine infusion restored a normal relationship (closed square), suggesting the presence of severe systolic LV dysfunction.

ated with a low or normal PAOP [15], even after resuscitation. Right heart catheteri- zation is, therefore, inaccurate for this diagnosis, as illustrated in Figure 3, and this probably explains why LV systolic dysfunction has long been markedly underesti- mated in this situation.

Echocardiography does not require an algorithm combining measurements of cardiac filling pressures and CI, but directly visualizes segmental and global LV con- tractility. Whereas measurement of LV volumes, to calculate LV ejection fraction (LVEF), is classically recommended in assessment of LV systolic function, we recently demonstrated by a transesophageal route the accuracy of a qualitative approach, which permitted non-echocardiographers to separate patients with severe and moderate LV failure from those without such failure [16].

How to ‘Monitor’ Right Ventricular Failure?

Clearly, most parameters used with right heart catheterization to assess right ven-

tricular (RV) function are actually indicators of the status of the pulmonary circula-

tion, such as pulmonary vascular resistance (PVR), and pulmonary artery pressure

(PAP). The assumption is that impairment of these parameters, reflecting damage to

the pulmonary circulation, may suggest that the right ventricle tolerates poorly these

effects. However, since the famous paper of Zapol and Snider [17], it is well recog-

nized that, in mechanically ventilated patients, PVR strongly depends on flow: An

increase in CI induces a decrease in PVR, and a decrease in CI induces an increase

in PVR. This follows the recruitment and derecruitment of pulmonary capillaries

crushed by a positive alveolar pressure. We have also demonstrated that, in patients

with severe acute respiratory distress syndrome (ARDS), an elevated systolic PAP

does not predict RV tolerance [18]. In some cases, a slight increase in PAP may be

enough to induce RV failure, whereas in other situations the right ventricle is able to

adapt to a marked increase in PAP. Previous studies proposed assessing RV function

directly by measuring RVEF, using fast thermistance catheters [19]. Once again, this

suggestion turned out to be inaccurate, especially in mechanically ventilated patients

[20], and in patients with significant pulmonary hypertension [21]. Finally, we dem-

onstrated that RV fractional area contraction, a surrogate for RVEF, measured by

echocardiography, did not differ significantly among patients with and without

acute cor pulmonale (Fig. 4) [22].

Fig. 4. Right ventricular fractional area contraction (RVFAC) in three groups, healthy volunteers, and patients with acute respiratory dis- tress syndrome (ARDS) with and without acute cor pulmonale (ACP).

Note the large overlap of values.

a

b

Fig. 5. Acute cor pulmonale in a shock patient ventilated for acute respiratory distress syndrome related to varicella pneumonia. The right ventricle (RV) was severely dilated (panel a), whereas paradoxical sep- tal motion (arrow) was present (panel b). LV: left ventricle.

Echocardiography may directly visualize the right ventricle, and so quickly assess its

function. Because of its properties, i.e., a ‘passive conduit’ that in a normal situation

ejects blood into a low-pressure circulation, a failing right ventricle dilates markedly,

which is very easy to assess with echocardiography. Moreover, echocardiography is

also able to detect acute cor pulmonale, a situation not so rare in critically ill

One of the major criticisms of skeptics regarding the use of echocardiography for hemodynamic monitoring in critically ill patients is that it cannot be done continu- ously. This is true, but then which kind of elaborate continuous monitoring has pre- viously had an impact on monitoring in the ICU? None, and surely not right heart catheterization. This is illustrated in Figure 6, which shows the results of a study by Gattinoni et al. [24]. When compared to a control group, the mortality was exactly the same in a group where CI was continuously monitored and optimized, and in another group in which SvO

was also continuously monitored and optimized [24].

The second message of this study was that, after initial resuscitation, CI and SvO

were already within the normal range of values [24].

One of the objectives of hemodynamic monitoring, perhaps the most important, is to evaluate the risk for organ hypoperfusion, and then to correct it. In the litera- ture, low blood pressure and metabolic acidosis in critically ill patients seem to eval- uate this risk accurately. Low blood pressure is easily detected because most of our seriously ill patients have an arterial catheter, and so blood pressure is continuously recorded. In a recent study, Varpula et al. demonstrated that a strong independent factor of mortality was the time during which a patient had a mean arterial pressure (MAP) less than 60 mmHg [25]. Metabolic acidosis can be diagnosed by repeated measurements of arterial base excess. Estenssoro et al. reported a significantly lower base excess in non-survivors in the ICU, and, more interestingly, a lack of improve- ment during the first days of treatment [26]. Finally, by coupling continuous moni- toring of blood pressure and ‘semi-continuous’ monitoring of base excess, physi- cians may estimate the risk of organ hypoperfusion. These tests can be used for screening, and echocardiography, which is quickly performed if the screening test is positive, may then be used to detect hypovolemia, RV failure, or LV failure.

Fig. 6. Schematic representation of the study by Gattinoni et al. [24]. The mortality rate did not differ

according to the group, or to the optimization of cardiac index (CI) or mixed venous oxygen saturation

(SvO

). After initial resuscitation, CI and SvO

were within normal ranges in the control group.

Conclusion

Development of a new type of more functional and qualitative monitoring, mainly based on the use of echocardiography, calls for a complete change in the way we think about hemodynamic diagnosis. Use of algorithms, which couple CI and filling pressure, should be abandoned, and physicians need only describe what they see on the screen of the echocardiograph to give adequate treatment, providing they have performed echocardiography in high-risk situations for organ hypoperfusion, such as low blood pressure or persistent metabolic acidosis.

References

1. Shah M, Hasselblad V, Stevenson L, et al (2005) Impact of the pulmonary artery catheter in critically ill patients: meta-analysis of randomized clinical trials. JAMA 294:1664 – 1670 2. Pinsky M, Vincent JL (2005) Let us use the pulmonary artery catheter correctly and only

when we need it. Crit Care Med 33:1119 – 1122

3. Boldt J (2002) Clinical review: hemodynamic monitoring in the intensive care unit. Crit Care 6:52 – 59

4. Robin E (1985) The cult of the Swan-Ganz catheter. Overuse and abuse of pulmonary flow catheters. Ann Intern Med 103:445 – 449

5. Jardin F (2003) Ventricular interdependance: how does it impact on hemodynamic evaluation in clinical practice ? Intensive Care Med 29:361 – 363

6. Toussignant C, Walsh F, Mazer C (2000) The use of transesophageal echocardiography for preload assessment in critically ill patients. Anesth Analg 90:351 – 355

7. De Backer D, Heenen S, Piagnerelli M, Koch M, Vincent JL (2005) Pulse pressure variations to predict fluid responsiveness : influence of tidal volume. Intensive Care Med 31:517 – 523 8. Vieillard-Baron A, Chergui K, Rabiller A, et al (2004) Superior vena caval collapsibility as a

gauge of volume status in ventilated septic patients. Intensive Care Med 30:1734 – 1739 9. Tavernier B, Makhotine O, Lebuffe G, Dupont J, Scherpereel P (1998) Systolic pressure varia-

tion as a guide to fluid therapy in patients with sepsis-induced hypotension. Anesthesiology 89:1309 – 1310

10. Combes A, Arnoult F, Trouillet JL (2004) Tissue Doppler imaging estimation of pulmonary artery occlusion pressure in ICU patients. Intensive Care Med 30:75 – 81

11. Charron C, Caille V, Jardin F, Vieillard-Baron A (2006) Echocardiographic measurement of fluid responsiveness. Curr Opin Crit Care 12:249 – 254

12. Jardin F, Vieillard-Baron A (2006) Ultrasonographic examination of the venae cavae. Inten- sive Care Med 32:203 – 206

13. Barbier C, Loubieres Y, Schmit C, et al (2004) Respiratory changes in inferior vena cava diameter are helpful in predicting fluid responsiveness in ventilated septic patients. Intensive Care Med 30:1740 – 1746

14. Vieillard-Baron A, Prin S, Chergui K, Dubourg O, Jardin F (2003) Hemodynamic instability in sepsis: bedside assessment by Doppler echocardiography. Am J Respir Crit Care Med 168:

1270 – 1276

15. Jardin F, Valtier B, Beauchet A, Dubourg O, Bourdarias JP (1994) Invasive monitoring com- bined with two-dimensional echocardiographic study in septic shock. Intensive Care Med 20:550 – 554

16. Vieillard-Baron A, Charron C, Chergui K, Peyrouset O, Jardin F (2006) Bedside echocardio- graphic evaluation of hemodynamics in sepsis: is a qualitative evaluation sufficient? Intensive Care Med 32:1547 – 1552

17. Zapol W, Snider M (1977) Pulmonary hypertension in severe acute respiratory failure. N Engl J Med 296:476 – 480

18. Vieillard-Baron A, Schmitt JM, Augarde R, et al (2001) Acute cor pulmonale in acute respira- tory distress syndrome submitted to protective ventilation : incidence, clinical implications, and prognosis. Crit Care Med 29:1551 – 1555

19. Dhainaut JF, Pinsky M, Nouria S, Slomka F, Brunet F (1997) Right ventricular function in

human sepsis : a thermodilution study. Chest 112:1043 – 1049

tion of acute cor pulmonale at the bedside in the medical intensive care unit. Am J Respir Crit Care Med 166:1310 – 1319

23. Jardin F, Dubourg O, Bourdarias JP (1997) Echocardiographic pattern of acute cor pulmo- nale. Chest 111:209 – 217

24. Gattinoni L, Brazzi L, Pelosi P, et al (1995) A trial of goal-oriented hemodynamic therapy in critically ill patients. SvO2 Collaborative Group. N Engl J Med 333:1025 – 1032

25. Varpula M, Tallgren M, Saukkonen K, Voipio-Pulkki L, Pettila V (2005) Hemodynamic vari- ables related to outcome in septic shock. Intensive Care Med 31:1066 – 1071

26. Estenssoro E, Dubin A, Laffaire E, et al (2002) Incidence, clinical course, and outcome in 217

patients with acute respiratory distress syndrome. Crit Care Med 30:2450 – 2456