Printer Friendly

Does left ventricular tissue Doppler peak systolic velocity (Sm) reflect cardiac output in the critically ill?


Cardiac output (CO) is dependent on a number of factors, in particular, the systolic function of the heart. Tissue Doppler (TD) is a modality in echocardiography that measures myocardial velocity and is related to contractility. TD can therefore be used to measure the systolic function of the heart. This study sought to establish whether the systolic component of TD can be used to estimate CO in critically ill patients.

Retrospective data was obtained from a total of 80 patients: 29 patients with a normal echocardiogram, and 51 intensive care unit patients; 28 septic and 23 with heart failure.

The mean TD peak systolic velocity (Sm) was significantly lower in the heart failure patients (P <0.05) compared to both normal and septic group. The mean CO was significantly higher in septic patients when compared to heart failure patients. A mild to moderate positive correlation was found between Sm and CO in the heart failure group and with all patients combined ([r.sup.2] = 0.19, P <0.001). Subsequent analysis of Sm versus stroke volume again showed a mild positive correlation in the heart failure group and combined results ([r.sup.2] =0.18, P <0.001). Sm was weakly correlated to heart rate only in the normal group but not in the combined cohort.

Our data confirms a weak to moderate correlation between Sm and CO, probably resulting from a positive correlation of Sm and stroke volume. This correlation is not strong enough to support the use of an individual's Sm to estimate CO in intensive care patients.

Key Words: echocardiography, tissue Doppler imaging (TD|), systolic myocardial velocity (Sm), cardiac output


Cardiac output (CO) is an important determinant of tissue perfusion in the critically ill and plays an important role in guiding treatment. Measurement of CO traditionally involves invasive procedures such as the pulmonary artery catheter and pulse-induced contour CO thermodilution methods. Non-invasive CO determination using transthoracic echocardiography is a reliable method, but is skill-demanding. CO by echocardiography can be measured in two ways: 1) by Simpson's method and 2) by Doppler. In the former, the end-systolic and end-diastolic volumes are first estimated by delineating the endocardial borders in two projections. The stroke volume (SV) is calculated by the difference between the end-systolic and end-diastolic volumes. CO is finally obtained by multiplying the SV by heart rate (HR). The Doppler method calculates SV from the cross-sectional area (CSA) of the left ventricular outflow tract (LVOT) and the velocity time integral (VTI) in the LVOT; SV = LVOT CSA x VTI. CO is again calculated as SV x HR. Reliable results depend on optimal ultrasound windows and operator experience. For example, a 10% error in measuring the LVOT diameter results in a 20% error in calculating CO'. Further, a small Doppler angle error may result in underestimation of the VTI, hence CO.

Myocardial peak systolic velocity measured by tissue Doppler imaging at the mitral annulus reflects left ventricular (LV) contractility, is less dependent on image quality and is easy to obtain. It has been used as a surrogate of LV systolic function (2) and has been shown to correlate with LV global dP/dt and left ventricular ejection fraction (LVEF) (3-6). The association between mean tissue Doppler peak systolic velocity (Sin) and CO however, is not as well established. Defining a relationship between Sm and CO would potentially allow the clinician to estimate CO rapidly and non-invasively with minimal measurement errors. This study examined the association between Sm and CO in non-critically ill patients with normal cardiac function as well as critically ill patients with sepsis and heart failure, respectively.



Data was examined from a total of 80 studies conducted at the Nepean Hospital intensive care unit echocardiography laboratory over a six month period. We excluded all patients with significant valvular disease. The patients were divided into three groups: normal, septic or heart failure. Thirty consecutive outpatients with normal cardiac function who were not septic were included in the normal group, with one being excluded due to insufficient data. Thirty-seven consecutive patients admitted to Nepean intensive care unit for sepsis were included in the septic group, with eight excluded for not being in sinus rhythm and one being excluded for poor image quality. Sepsis treatment included vasopressors and mechanical ventilation in most cases. Twenty-four consecutive patients admitted to Nepean intensive care unit with documented heart failure were included in the heart failure (HF) group, with one being excluded due to poor image quality. All HF patients had either a history of congestive heart failure or ischaemic heart disease, presented with LVEF <40%, and were treated with vasoactive/inotropic drugs or positive pressure ventilation for HE No patient had isolated right ventricular failure or diastolic heart failure with normal LVEE

Demographic details for each group are shown in Table 1.


All transthoracic echocardiography was performed by qualified sonographers using ultrasound systems equipped with tissue Doppler option (Vivid-I or Vivid-7, GE Medical, Norway). CO was obtained using the Doppler method and was calculated from SV x HR, where SV is the product of LVOT CSA and LVOT VTI. LVOT diameter, obtained from the parasternal long axis view, was used to calculate LVOT CSA. LVOT VTI, which represents the summation of velocities per heart beat, was obtained in the apical five chamber view using pulsed-wave Doppler placed in the LVOT. Tissue Doppler imaging-derived Sm measures were obtained by placing the sample gate at the level of mitral annulus of the inter-ventricular septum (SmM) and lateral wall (SmL) in the apical four chamber view. SmM and SmL were averaged for each study as recommended by Rivas-Gotz et al (7).


An average of two or more cardiac cycles was used in all measurements. Statistical calculations and comparisons were performed using the average values. All data was analysed using NCSS 2007 (NCSS, LLC, Kaysville, UT, USA) statistical software and plotted using SigmaPlot 10.0 (Systat Software, Inc, San Jose, CA, USA). Group comparisons were performed using Student's t-test. Multiple group comparisons were performed using one-way analysis of variance and, if statistically significant, followed by post-hoc pairwise comparison using Student's t-test after Bonferroni adjustment for multiple tests. Correlations were tested by linear regression and significance test was carried out using Pearson's correlation. Level of significance was taken as P [less than or equal to] 0.05.


Study population

The mean Sm and CO for the three groups are depicted in Figure 1 and Table 1. The Sm of the HF group (6.1 [+ or -] 1.8 cm/second) was significantly lower (P <0.001) than the normal (9.2 [+ or -] 2.0 cm/second) and septic groups (8.9 [+ or -] 3.0 cm/second). The normal and septic groups did not differ from each other statistically. Analysis of CO in each of the three groups showed a significant difference between the heart failure group (4.3 [+ or -] 1.4 1/minute) and the septic group (5.8 [+ or -] 2.3 1/minute), (P=0.023).

Sm vs CO

The correlation in the HF patients was significant (r=0.48; P=0.021) (Figure 2b). Neither the normal group (r=0.36; P=0.054) nor the septic group demonstrated a significant correlation (r=0.34; P=0.078) (Figures 2a and 2c). On the other hand, the combined data did reveal a significant but weak correlation with r=0.43 and P <0.001 (Figure 2d).

Sm vs SV

When analysed separately, there was only a significant relationship between Sm and SV for the HF group with r=0.47 and P=0.025 (Figures 3a to 3c). With all groups combined a significant but weak correlation was found (r=0.42, P <0.001) (Figure 3d).

Sm vs HR

Increasing HR was associated with increasing Sm in the normal groups (r=0.44, P=0.018) (Figure 4a). A similar but insignificant trend was noted in the septic group (r=0.33, P=0.09) (Figure 4c). There was no apparent relationship between HR and Sm in the HF group (r=0.13, P=0.549) (Figure 4b). When combined, the significant association seen in the normal group was lost (r <0.001, P=0.993) (Figure 4d).


The results of the present study showed that LV myocardial contractility as measured by Sm had weak but significant correlations with CO and SV when all three patient groups were combined. As both CO and Sm were reduced in HF patients the correlation was more robust. Only patients in the normal group displayed significant but weak correlation between Sm and HR. As expected septic patients had significantly higher CO then HF patients but no correlation was found with Sm.

Sm, which measures the myocardial peak systolic velocity at the level of mitral annulus, has been shown to correlate strongly with the LV dP/dt, a measure of global LV contractility (3,4). According to pressure-volume loop analysis, an increase in contractility (end-systolic LV elastance) results in an increase in SV, hence CO. Our data confirm a weak to moderate correlation between Sm and CO. The lack of a strong correlation, in particular in the septic group, may be due to a number of confounding factors. Sm is known to be a load-sensitive variable with relationship to LV systolic pressure (3). Almuntaser et al found that reducing blood pressure in hypertensive patients significantly increased Sm over three months (8). Huez et al found that dobutamine administration significantly increased CO and Sm, most probably due to dobutamine's inotropic and vasodilatory properties. (9). Thus low vascular resistance in sepsis would reduce the work of emptying the left ventricle and hence potentially increase Sm. On the other hand, sepsis-induced cardiomyopathy may lead to reduced Sm and the contribution of both is likely variable.

Since CO is a function of SV and HR, we sought to find if the observed correlation between Sm and CO was due to SV or HR. Duzenli et al demonstrated that the mean Sm correlated with LVEF in HF patients but not in a normal population (10). Van Melle et al found that a decrease in Sm can be used to define a LVEF <40% post myocardial infarction (11). In the present study, only the HF group demonstrated significant correlation between Sm and SV. The lack of correlation in the normal group was somewhat surprising but in line with the findings of Duzenli. The reason for this is unclear but indicates that the SV in HF patients is more dependent on LV contractility (the 'limiting' factor), whereas the SV in normal and septic subjects is likely to be dependent on other factors such as HR, preload, afterload, increased LV wall thickness and the use of vasoactive medication.

Despite an overall lack of statistically significant correlation between Sm and CO, the normal group showed a significant but weak correlation between Sm and HR. The effect of HR on Sm is difficult to demonstrate mainly because most chronotropic agents or manoeuvres are also inotropic. It would appear that in order to support a higher HR the normal myocardium must increase the speed at which it contracts. At higher HRs the Treppe effect leads to increased force of contraction due to the inability to expel the excessive intracellular [Ca.sup.2+] influx by the [Na.sup.+]-[Ca.sup.2+] exchanger during the shortened diastolic phase. The correlation between HR and Sm in normal subjects was noted by Quintana et al (12). They found that Sm increased with HR during exercise. However, they could not exclude the inotropic effect during exercise which might exert an overwhelming effect on Sm. Burns et al, on the other hand, by pacing the HR at two different levels (~67 and ~80 bpm) found that HR had no discernible effect on Sm (13). In this study, we observed a mild correlation between Sm and HR in the normal (significant) and septic (non significant) groups but not in the HF group. The narrow range of HR in the normal group made the interpretation difficult. However, pooled results clearly indicated that Sm was not correlated to HR.


All echocardiography was performed by a small group of qualified sonographers; however, interobserver variation cannot not be completely ruled out. Patient habitus and position during the study might also contribute to the variability observed. Doppler angle error can contribute to measurement errors of both LVOT VTI and Sm resulting in underestimation of measurements. Finally, as the HF group did demonstrate a statistically significant correlation between Sm and CO it is plausible that the lack of significant correlation in the normal and septic groups could be genuinely physiological or that our sample size was too small to detect it.


This study confirms a weak correlation of Sm and CO, mainly in intensive care patients with compromised LV systolic function. However, the relationship is not strong enough to reliably estimate CO from Sm in individual patients. Further studies with a larger sample size and controlling for other confounding factors such as vasoactive and inotropic drug use, mechanical ventilation and systemic vascular resistance may be useful.

Caption: Figure 1: Group comparison using one-way analysis of variance for Sm (average Sm of medial and lateral walls) and CO (cardiac output). * P <0.05 between HF and sepsis. # P <0.05 between HF and Normal. ([dagger]) P <0.05 between HF and sepsis.

Caption: Figure 2: Scatter plot of CO (cardiac output) vs Avg Sm (average Sm of medial and lateral walls) with regression line for normal (2a), heart failure (2b), and sepsis (2c) groups, with combined data shown in 2d.

Caption: Figure 3: Scatter plot of SV (stroke volume) vs Avg Sm (average Sm of medial and lateral walls) with regression line for normal (3a), heart failure (3b), and sepsis (3c) groups, with combined data shown in 3d.

Caption: Figure 4: Scatter plot of HR (heart rate) vs Avg Sm (average Sm of medial and lateral walls) with regression line for normal (4a), heart failure (4b), and sepsis (4c) groups, with combined data shown in 4d.


(1.) Huang S, Maclean A. The appreciation of the strengths and weaknesses of transthoracic echocardiography in hemodynamic assessments. Submitted for publication.

(2.) Witte KKA, Nikitin NP, De Silva R, Cleland JGE Clark AL. Exercise capacity and cardiac function assessed by tissue Doppler imaging in chronic heart failure. Heart 2004; 90:1144-1150.

(3.) Yu WC, Lee WS, Huang WE Wu CC, Lin YE Cben CH. Evaluation of cardiac function by tissue Doppler echocardiography: hemodynamic determinants and clinical application. Ultrasound Med Biol 2005; 31:23-30.

(4.) Seo JS, Kim DH, Kim W J, Song JM, Kang DH, Song JK. Peak systolic velocity of mitral annular longitudinal movement measured by pulsed tissue Doppler imaging as an index of global left ventricular contractility. Am J Physiol Heart Circ Physiol 2010; 298:H1608-1615.

(5.) Sanderson JE, Wang M, Yu CM. Tissue Doppler imaging for predicting outcome in patients with cardiovascular disease. Curr Opin Cardiol 2004; 19:458-463.

(6.) Sallach JA, Klein AL. Tissue Doppler imaging in the evaluation of patients with cardiac amyloidosis. Curr Opin Cardiol 2004; 19:464-471.

(7.) Rivas-Gotz C, Manolios M, Thohan V, Nagueh SE Impact of left ventricular ejection fraction on estimation of left ventricular filling pressures using tissue Doppler and flow propagation velocity. Am J Cardiol 2003; 91:780-784.

(8.) Almuntaser I, Mahmud A, Brown A, Murphy R, King G, Crean P et al. Blood pressure control determines improvement in diastolic dysfunction in early hypertension. Am J Hypertens 2009; 22:1227-1231.

(9.) Huez S, Retailleau K, Unger P, Pavelescu A, Vachiery JL, Derumeaux G et al. Right and left ventricular adaptation to hypoxia: a tissue Doppler imaging study. Am J Physiol Heart Circ Physiol 2005; 289:H1391-1398.

(10.) Duzenli MA, Ozdemir K, Aygul N, Altunkeser BB, Zengin K, Sizer M. Relationship between systolic myocardial velocity obtained by tissue Doppler imaging and left ventricular ejection fraction: systolic myocardial velocity predicts the degree of left ventricular dysfunction in heart failure. Echocardiography 2008; 25:856-863.

(11.) van Melle JE van der Vleuten PA, Hummel YM, Nijveldt R, Tio RA, Voors AA et al. Predictive value of tissue Doppler imaging for left ventricular ejection fraction, remodelling, and infarct size after percutaneous coronary intervention for acute myocardial infarction. Eur J Echocardiogr 2010; 11:596-601.

(12.) Quintana M, Gustafsson T, Sundblad P, Langanger J. The effects of heart rate on myocardial velocity and atrio-ventricular displacement during exercise with and without beta-blockade: a tissue Doppler echocardiographic study. Eur J Echocardiogr 2005; 6:127-133.

(13.) Burns AT, Connelly KA, La Gerche A, Mooney D J, Chan J, Maclsaac Al et al. Effect of heart rate on tissue Doppler measures of diastolic function. Echocardiography 2007; 24:697-7(11.

A. R. GOLOWENKO *, M. NALOS ([dagger]), S. J. HUANG ([double dagger])

Intensive Care Unit, Nepean Hospital, Sydney, New South Wales, Australia

* MB, BS, BInfoTech, DipEd, Honors Medical Student, Nepean Clinical School, University of Sydney

([dagger]) MB, BS, PhD, Staff Specialist, Department of Intensive Care Medicine, Nepean Hospital and Senior Lecturer, Nepean Clinical School, University of Sydney

([double dagger]) BSc, MSc, PhD, Cert Ed, DiplLaw, GradDipLegalPrac, CRFS, AMS, Chief Scientist, Department of Intensive Care Medicine, Nepean Hospital and Nepean Clinical School, University of Sydney

Address for correspondence: Dr M. Nalos. Email:

Accepted for publication on May 10, 2013

Table 1
Patient characteristics

       Normal                 Sepsis

Male   59% (17/29)            75% (21/28)
Age    58.3 ([+ or -] 12.0)   63.1 ([+ or -] 17.1)
CO     5.0 ([+ or -] 1.0)     5.8 ([+ or -] 2.3)
Sm     9.2 ([+ or -] 2.0)     8.9 ([+ or -] 3.0)

       Heart failure          Combined

Male   87% (20/23)            73% (58/80)
Age    68.1 ([+ or -] 12.1)   61.6 ([+ or -] 14.1)
CO     4.3 ([+ or -] 1.4)     5.1 ([+ or -] 1.7)
Sm     6.1 ([+ or -] 1.8)     8.2 ([+ or -] 2.7)

Group and combined patient characteristics showing percentage
males and number out of total group size, age as mean with
standard deviation, CO (cardiac output) as mean with standard
deviation, and Sm (average Sm of medial and lateral walls, cm/
second) as mean with standard deviation.
COPYRIGHT 2013 Australian Society of Anaesthetists
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2013 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Original Papers
Author:Golowenko, A.R.; Nalos, M.; Huang, S.J.
Publication:Anaesthesia and Intensive Care
Article Type:Clinical report
Date:Jul 1, 2013
Previous Article:Is comorbid status the best predictor of one-year mortality in patients with severe sepsis and sepsis with shock?
Next Article:Chronic pain after caesarean delivery: an Australian cohort.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters