Share this article on:

Tissue Doppler Measurements Correlate With Central Venous Pressure in Pediatric Patients After Cardiac Surgery

Learn, Christopher P.*†; Yates, Andrew R.*‡; Nicholson, Lisa§; Milton, Keyana*; Lloyd, Eric‡; Cua, Clifford L.*†§

doi: 10.1097/MAT.0b013e3181e5d543
Pediatric Circulatory Support

Tissue Doppler imaging (TDI) measurements have been demonstrated to correlate with central venous pressure (CVP) in adults. We hypothesize that TDI measurements also correlate with CVP in children after cardiac surgery. Patients younger than 10 years with invasive CVP monitoring after cardiac surgery were studied. Annular tissue and atrioventricular inflow velocities were measured daily during CVP monitoring. Tissue Doppler imaging measurements were compared with CVP by univariate analyses. Subgroup analyses were performed for univentricular versus biventricular hearts. p values ≤ 0.05 were considered significant. Fifty studies were performed on 28 subjects; 20 studies were performed on patients with univentricular physiology; and 30 studies were performed on patients with biventricular physiology. For all subjects, CVP correlated significantly with the right ventricular TDI measurements of e′ velocity (r = −0.43), a′ velocity (r = 0.35), e′/a′ ratio (r = −0.42), and E/e′ ratio (r = 0.44). For biventricular patients, e′ velocity (r = −0.35), e′/a′ ratio (r = −0.38), and E/e′ ratio (r = 0.58) correlated significantly. For univentricular patients, E velocity (r = −0.62), e′ velocity (r = −0.50), and a′ (r = 0.61) velocity correlated significantly. Tissue Doppler imaging measurements of the right heart significantly correlated with invasively measured CVP in pediatric subjects. Different measures may correlate more strongly in those with biventricular versus univentricular physiology. With additional study, TDI may prove a valid tool for noninvasively assessing CVP in children with both bi- and univentricular physiology.

From the *The Heart Center, †Division of Cardiology, ‡Section of Critical Care, and §Nationwide Children's Hospital and Research Institute, Columbus, Ohio.

Submitted for consideration November 2009; accepted for publication in revised form March 2010.

Presented at the Pediatric Cardiac Intensive Care Society Meeting, Miami, FL, December 2009.

Reprint Requests: Clifford L. Cua, MD, Heart Center, Nationwide Children's Hospital, 700 Children's Drive, Columbus, OH 43205-2696. Email:

Measurement of central venous pressure (CVP) has significant clinical importance for a variety of disease states.1 According to pediatric consensus guidelines, CVP should be used to guide resuscitation in shock states.2 Central venous pressure has additional importance in patients with congenital heart disease (CHD).3–6 Elevated CVP has been associated with the development of collateral vessels, which causes increasing cyanosis, in single-ventricle patients7 and predicting elevated levels of vasoactive hormones after cardiac surgery.8 Furthermore, CVP is useful in determining responsiveness in patients with CHD being treated for pulmonary hypertension9 or being treated by temporary pacing.10 Traditional Doppler blood flow echocardiographic techniques can estimate diastolic function and indirectly CVP, but they are preload dependent, and pseudonormalization patterns can sometimes make interpretations difficult. Tissue Doppler imaging (TDI) yields additional information about diastolic function by measuring the velocity of specific areas of muscle during the cardiac cycle. Tissue Doppler imaging techniques avoid the issue of a pseudonormalization pattern and also have the advantage of being less preload dependent.11,12

Several studies have demonstrated that mitral annular velocities correlate with hemodynamic measurements in both adults and children with anatomically normal hearts.13–16 Tissue Doppler imaging measurements of mitral valve annular velocities have also been demonstrated to correlate with mean left atrial pressure for children with CHD in the postoperative intensive care setting.16 Tissue Doppler imaging measurements of the tricuspid valve correlate with CVP in adults, but no data exist in the pediatric population cardiac surgery.17 Normative data is available for left ventricular TDI measurements for children without CHD,18,19 but less data are available on right ventricular (RV) annular measurements,20 especially in children with CHD.21,22

The aim of this study was to determine whether RV TDI values correlate with CVP in children after cardiac surgery in the intensive care setting. We hypothesized that TDI measurements would correlate significantly in patients with biventricular and univentricular physiology.

Back to Top | Article Outline

Materials and Methods

This study uses a cross-sectional prospective design at a single center. Following institutional review board approval, patients within the cardiac intensive care unit with indwelling central venous catheters were identified. Patients with right atrial, internal jugular, subclavian, or umbilical catheters were eligible provided the catheter extended to the atrium or caval-atrial junction. Line position was confirmed by chest x-ray. Subjects were excluded if apical four-chamber imaging was severely limited or if subjects were not in a stable sinus or atrial rhythm. To better compare TDI measurements of the tricuspid valve and right ventricle between patients with single ventricle versus biventricular physiology, we excluded subjects with single-ventricle morphology consistent with a left ventricle.

After parental consent, focused echocardiograms (Vivid I system, GE Health care, Waukesha, WI) were performed on postoperative day 0 and daily when CVP monitoring was available to maximize data points. If multiple measurements were made on the same patient, the measurements were separated by at least 24 hours. The CVP was recorded before initiation of the echocardiogram. The right atrial area was traced in the biventricular patients just before ventricular systole. Tricuspid inflow waveforms were recorded at the leaflet tips from the apical four-chamber view. Tissue Doppler from the pulse waveforms were sampled at the tricuspid free wall and the interventricular septum (when present) at the level of the annuli. Settings were optimized to reduce background noise. Off-line measurements were performed on a HeartSuite VERICIS workstation (Emageon, Burmingham, AL) by a single observer blinded to the CVP values. Consistent waveforms were selected for measurement. All measurements were performed in triplicate. Mean values were used for analysis.

Peak early-diastolic blood inflow (E) and peak late-diastolic inflow (A) velocities were measured at the tricuspid valve. Peak annular velocities were recorded during early diastole (e′), late diastole (a′), and systole (s′). Calculated values included the ratio between early- and late-diastolic blood inflow (E/A), the ratio between early-diastolic blood inflow and early-diastolic annular velocity (E/e′) and the ratio between early- and late-annular velocity (e′/a′). The myocardial performance index was calculated from TDI time intervals.23,24

Back to Top | Article Outline

Statistical Analysis

Linear regression analysis was performed with Microsoft Excel. Pearson′s r correlations were performed to determine the strength of associations between variables. Significance levels were tested in STATA 11 SE. Analyses were initially performed for all subjects. Subanalysis was then performed separating those with biventricular versus univentricular physiology. p values <0.05 were considered significant.

Back to Top | Article Outline


Thirty patients were enrolled. Two patients with biventricular physiology were excluded because of poor echocardiographic windows; therefore, this study consisted of 50 studies on 28 subjects. Thirty studies were performed on 18 patients with biventricular physiology. Twenty studies were performed on 10 patients with univentricular physiology. Demographics are outlined in Table 1, and anatomic diagnoses are summarized in Table 2. All results are reported as means with standard deviations unless otherwise stated. Table 3 outlines data for physiologic parameters and RV measurements. There were no statistical differences between groups unless outlined.

For all subjects, the following RV TDI measurements had weak to moderate significant correlations with CVP: e′ velocity (r = −0.43), a′ velocity (r = 0.35), e′/a′ ratio (r = −0.42) (Figure 1), and E/e′ ratio (r = 0.44) (Figure 2). For biventricular patients, e′ velocity (r = −0.35), e′/a′ ratio(r = −0.38) (Figure 3), and E/e′ ratio (r = 0.58) (Figure 4) correlated significantly. Right ventricular annular a′ velocity did not correlate significantly (r = 0.30). For univentricular patients, moderate significant correlations were found in E velocity (r = −0.62), e′ velocity (r = −0.50), and a′ (r = 0.61) (Table 4). The e′/a′ ratio approached significance with a p value of 0.058 (r = −0.48) (Figure 5). The E/e′ ratio did not reach significance (Figure 6). The right atrial area indexed to body surface area did not significantly correlate with any of the TDI values measured (Table 5).

Back to Top | Article Outline


Adult data clearly demonstrate that TDI measurements of the left and right ventricles correlate with left and right atrial mean pressures, respectively.15,25–27 Data in the pediatric population have shown similar correlations of left ventricular TDI measurements and left atrial mean pressures,13,14,16 but scarce information is available for the RV17,21 in a biventricular physiology, and no data are available for patients with single-ventricle physiology. This study is the first to our knowledge that demonstrates that there are significant correlations between TDI measurements of the RV and CVP in pediatric patients with biventricular and univentricular physiology undergoing cardiac surgery.

When analyzing all patients′ measurements, correlations were found in the RV e′, a′, e′/a′, and E/e′ parameters with CVP. This is consistent with adult data15,25,26 where increasing CVP affected these measurements. Furthermore, when evaluating those patients with biventricular physiology, the e′, E/e′, and e/a continued to have significant correlations with the E/e′ ratio having the strongest correlation of 0.58. Similar findings correlating RV TDI and CVP have been demonstrated in the biventricular pediatric population with CHD, but this was under a more controlled situation in the cardiac catheterization laboratory.21 The fact that these correlations still exist in the postoperative CHD patient where multiple factors may influence the hemodynamic parameters is promising. Despite these significant correlations, there were still variability in the TDI measurements, and these measurements could not predict a specific CVP. There was no correlation with the biventricular patients' right atrial volumes and the TDI values measured. This would imply that TDI is a more sensitive measurement for CVP. This is consistent with a recent adult study that demonstrated that TDI better correlates with ventricular filling pressures versus atrial volumes, albeit on the left side.28

For patients with univentricular physiology, only e′ and a′ demonstrated significant correlations with CVP. The more common e′/a′ and E/e′ ratios used in the adult population did not significantly correlate with CVP although there was a trend with the e′/a′ ratio (p = 0.058). This could be due to differences between left and RV contraction patterns. It has been shown that over time the systemic RV gradually adopts a configuration more typical of a left ventricle.29 These complex changes in ventricular configuration could affect TDI measurements annular velocity. The transitional state of the single-systemic RV physiology compared with the RV in a biventricular physiology is still not well understood and deserve further investigation. Despite these shortcomings, there still were correlations in the TDI with values with the RV a′ velocity having the strongest correlation (r = 0.61). This is consistent with previous findings showing a′ velocities increasing with increasing CVP.21,25,27 Because TDI techniques are applied to patients with CHD, it will be important to determine which TDI parameters are more accurate in approximating CVP in the respective populations.

Although correlations were found with TDI and CVP in patients with both the univentricular and biventricular CHD postoperatively with the strongest correlations being a′ and E/e′, respectively, there was still variability in these measurements, such that a specific TDI value could not predict a specific CVP. It may be that TDI values may be more useful as trend following changes within a patient versus comparing measurements between patients. Tissue Doppler imaging is not a continuous monitoring device, and it does not yet correlate strongly enough with CVP to replace invasive monitoring in the pediatric intensive care unit. However, with further studies, it may be used adjunctively or in patients where invasive monitoring is technically challenging or relatively contraindicated. A noninvasive method to monitor CVP in this complex population would be useful and as such TDI deserves further study.

Limitations of this study include its small sample size and the heterogeneity of the CHD present. Retrospective regression-based power analysis suggested that a sample size of 27 is required to yield a power of 80% in testing our hypothesis and including two covariates. There is enough power to examine the full sample of patients and the biventicular group of patients. Although there is not enough power to examine the univentricular patients on their own, results from the full sample and the biventricular sample may be able to shed light on the applicability of the overall findings to the univentricular group. For example, the correlations and significance of the RV e′, a′, and TV E values are stronger in the univentricular group than the biventricular group and all patients group, suggesting that these results may be more applicable to the univentricular patients. Patients with single left ventricular morphology and those not in a stable sinus or atrial rhythm were excluded, so no comment can be made on these patient populations. Several dynamic factors common in the intensive care unit, such as inotrope usage and ventilator management, were not controlled for in this study; however, to be clinically useful, a noninvasive measurement would need to accurate in the face of these variables. Despite this heterogeneity, there were still significant relationships between TDI measurements and CVP. Finally, all wave forms were measured by a single observer, so the potential for interobserver variability was not assessed.

Back to Top | Article Outline


Tissue Doppler imaging parameters correlated with CVP in pediatric cardiac patients undergoing cardiac surgery. Significant TDI correlations differed between biventricular versus univentricular patients, thus larger studies are still needed to determine whether TDI can be used consistently to estimate CVP and to determine which parameters are appropriate for the different physiologies present.

Back to Top | Article Outline


1. Gelman S: Venous function and central venous pressure: A physiologic story. Anesthesiology 108: 735–748, 2008.
2. Brierley J, Choong K, Cornell T, et al: Clinical practice parameters for hemodynamic support of pediatric and neonatal septic shock: 2007 update from the American College of Critical Care Medicine. Crit Care Med 37: 666–688, 2009.
3. Amoozgar H, Behniafard N, Borzoee M, Ajami GH: Correlation between peripheral and central venous pressures in children with congenital heart disease. Pediatr Cardiol 29: 281–284, 2008.
4. Burnell RH: Venous pressure in congestive heart failure in infancy. Arch Dis Child 45: 360–362, 1970.
5. Buchhorn R, Bartmus D, Buhre W, Bursch J: Pathogenetic mechanisms of venous congestion after the Fontan procedure. Cardiol Young 11: 161–168, 2001.
6. Hayashi Y, Uchida O, Takaki O, et al: Internal jugular vein catheterization in infants undergoing cardiovascular surgery: An analysis of the factors influencing successful catheterization. Anesth Analg 74: 688–693, 1992.
7. Heinemann M, Breuer J, Steger V, et al: Incidence and impact of systemic venous collateral development after Glenn and Fontan procedures. Thorac Cardiovasc Surg 49: 172–178, 2001.
8. Alvarez Kindelan A, Perez Navero JL, Ibarra de la Rosa I, et al: Relationship between hemodynamic changes and blood hormone concentrations after cardiac surgery in children with congenital heart disease. Crit Care Med 22: 1754–1761, 1994.
9. Morris K, Beghetti M, Petros A, et al: Comparison of hyperventilation and inhaled nitric oxide for pulmonary hypertension after repair of congenital heart disease. Crit Care Med 28: 2974–2978, 2000.
10. Janousek J, Vojtovic P, Chaloupecky V, et al: Hemodynamically optimized temporary cardiac pacing after surgery for congenital heart defects. Pacing Clin Electrophysiol 23: 1250–1259, 2000.
11. Pela G, Regolisti G, Coghi P, et al: Effects of the reduction of preload on left and right ventricular myocardial velocities analyzed by Doppler tissue echocardiography in healthy subjects. Eur J Echocardiogr 5: 262–271, 2004.
12. Sohn DW, Chai IH, Lee DJ, et al: Assessment of mitral annulus velocity by Doppler tissue imaging in the evaluation of left ventricular diastolic function. J Am Coll Cardiol 30: 474–480, 1997.
13. Nagueh SF, Middleton KJ, Kopelen HA, et al: Doppler tissue imaging: A noninvasive technique for evaluation of left ventricular relaxation and estimation of filling pressures. J Am Coll Cardiol 30: 1527–1533, 1997.
14. Eidem BW, McMahon CJ, Ayres NA, et al: Impact of chronic left ventricular preload and afterload on Doppler tissue imaging velocities: A study in congenital heart disease. J Am Soc Echocardiogr 18: 830–838, 2005.
15. Sundereswaran L, Nagueh SF, Vardan S, et al: Estimation of left and right ventricular filling pressures after heart transplantation by tissue Doppler imaging. Am J Cardiol 82: 352–357, 1998.
16. Larrazet F, Bouabdallah K, Le Bret E, et al: Tissue Doppler echocardiographic and color M-mode estimation of left atrial pressure in infants. Pediatr Crit Care Med 6: 448–453, 2005.
17. Nageh MF, Kopelen HA, Zoghbi WA, et al: Estimation of mean right atrial pressure using tissue Doppler imaging. Am J Cardiol 1999;84:1448–1451, A8.
18. Roberson DA, Cui W, Chen Z, et al: Annular and septal Doppler tissue imaging in children: Normal z-score tables and effects of age, heart rate, and body surface area. J Am Soc Echocardiogr 20: 1276–1284, 2007.
19. Eidem BW, McMahon CJ, Cohen RR, et al: Impact of cardiac growth on Doppler tissue imaging velocities: A study in healthy children. J Am Soc Echocardiogr 17: 212–221, 2004.
20. Roberson DA, Cui W: Right ventricular Tei index in children: Effect of method, age, body surface area, and heart rate. J Am Soc Echocardiogr 20: 764–770, 2007.
21. Watanabe M, Ono S, Tomomasa T, et al: Measurement of tricuspid annular diastolic velocities by Doppler tissue imaging to assess right ventricular function in patients with congenital heart disease. Pediatr Cardiol 24: 463–467, 2003.
22. Harada K, Tamura M, Toyono M, Yasuoka K: Comparison of the right ventricular Tei index by tissue Doppler imaging to that obtained by pulsed Doppler in children without heart disease. Am J Cardiol 90: 566–569, 2002.
23. Tei C, Nishimura RA, Seward JB, Tajik AJ: Noninvasive Doppler-derived myocardial performance index: Correlation with simultaneous measurements of cardiac catheterization measurements. J Am Soc Echocardiogr 10: 169–178, 1997.
24. Tei C, Ling LH, Hodge DO, et al: New index of combined systolic and diastolic myocardial performance: A simple and reproducible measure of cardiac function—A study in normals and dilated cardiomyopathy. J Cardiol 26: 357–366, 1995.
25. Nagueh SF, Lakkis NM, Middleton KJ, et al: Doppler estimation of left ventricular filling pressures in patients with hypertrophic cardiomyopathy. Circulation 99: 254–261, 1999.
26. Nagueh SF, Mikati I, Kopelen HA, et al: Doppler estimation of left ventricular filling pressure in sinus tachycardia. A new application of tissue doppler imaging. Circulation 98: 1644–1650, 1998.
27. Ommen SR, Nishimura RA, Appleton CP, et al: Clinical utility of Doppler echocardiography and tissue Doppler imaging in the estimation of left ventricular filling pressures: A comparative simultaneous Doppler-catheterization study. Circulation 102: 1788–1794, 2000.
28. Dokainish H, Nguyen JS, Sengupta R, et al: Do additional echocardiographic variables increase the accuracy of E/e' for predicting left ventricular filling pressure in normal ejection fraction? An Echocardiographic and Invasive Hemodynamic Study. J Am Soc Echocardiogr 23: 156–161, 2010.
29. Pettersen E, Helle-Valle T, Edvardsen T, et al: Contraction pattern of the systemic right ventricle shift from longitudinal to circumferential shortening and absent global ventricular torsion. J Am Coll Cardiol 49: 2450–2456, 2007.
Copyright © 2010 by the American Society for Artificial Internal Organs