Secondary Logo

Oesophageal or transgastric views for estimating mean pulmonary artery pressure with transoesophageal echocardiography: A prospective observational study

Cowie, Brian; Kluger, Roman

European Journal of Anaesthesiology: May 2018 - Volume 35 - Issue 5 - p 349–355
doi: 10.1097/EJA.0000000000000750
Cardiac anaesthesia
Free

BACKGROUND Recent data suggest that in cardiac surgical patients, the pulmonary artery acceleration time (PAT) is useful for estimating mean pulmonary artery pressure (MPAP) noninvasively with transoesophageal echocardiography (TOE). The pulmonary valve can be visualised from multiple echocardiographic windows, but it is unclear which, if any, view correlates best with MPAP.

OBJECTIVE(S) To compare the PAT measured with TOE from oesophageal and transgastric views with MPAP obtained invasively with a pulmonary artery catheter.

DESIGN A prospective observational study.

SETTING St. Vincent's Hospital, Melbourne, a university tertiary referral centre in Australia.

PATIENTS Sixty-three patients having cardiac surgery were included in our study. All patients had insertion of both a TOE probe and pulmonary artery catheter; this is the routine standard of care in our centre.

INTERVENTION(S) Nil.

MAIN OUTCOME MEASURES During a period of haemodynamic stability, the PAT was measured first from an oesophageal view and then immediately after from a transgastric view. The results were then compared with the invasively measured MPAP.

RESULTS Simultaneous measurements of MPAP and PAT were taken in 63 patients. In two patients, these measurements were not possible in the transgastric position due to an inability to visualise the right ventricular outflow tract and pulmonary valve. A Bland–Altman analysis of the PAT measured from the upper oesophageal and transgastric views showed a mean difference of 1 ms and limits of agreement of −18 to 16 ms. The area under the receiver operating curves for predicting pulmonary hypertension with PAT were upper oesophageal view 0.99 [95% confidence interval (CI), 0.98 to 1.00] and transgastric view 0.99 (95% CI, 0.97 to 1.00). The agreement between the results from these two views in the diagnosis of pulmonary hypertension (defined as PAT < 107 ms) was 93.4% with a kappa of 0.85 (95% CI, 0.59 to 1.00). There is an inverse curvilinear relationship between PAT and MPAP. Using a cut-off of 107 ms, the upper oesophageal view predicted pulmonary hypertension (defined as MPAP > 25 mmHg) with a sensitivity of 94.7% and specificity of 97.6%. The transgastric view predicted pulmonary hypertension with a sensitivity of 89.4% and specificity of 95.2%.

CONCLUSION Oesophageal and transgastric measurements of PAT have close agreement and a similar high ability to discriminate between people with and without pulmonary hypertension. The transgastric measurement was unobtainable in a small percentage of patients and required more probe manipulation. We would recommend PAT measurement in the upper oesophageal view.

From the Department of Anaesthesia, St. Vincent's Hospital, Melbourne, Melbourne, Victoria, Australia (BC, RK)

Correspondence to Dr Brian Cowie, Staff Anaesthetist, Department of Anaesthesia, St. Vincent's Hospital, Melbourne, Melbourne, Victoria, Australia E-mail: brian.cowie@svha.org.au

Published online 30 November 2017

Back to Top | Article Outline

Introduction

With the declining use of the pulmonary artery catheter (PAC) in the operating room, noninvasive methods of estimation of pulmonary artery pressure are required. Most commonly this has involved estimation of the right ventricular (RV) systolic pressure (RVSP) using the Tricuspid regurgitation (TR) peak jet velocity measured with echocardiography, and estimation or measurement of central venous pressure (CVP).1 Although the CVP is often measured continuously during cardiac surgery through a central line, this is not the case with many other types of surgery and CVP must be estimated, with consequent inaccurate estimations using the RVSP calculation.1–4 However, the RVSP measurement technique relies on having at least some element of TR and an adequate Doppler signal, conditions which were absent in over 40% of anaesthetised patients with transoesophageal echocardiography (TOE) in a recent series.4,5

The pulmonary artery acceleration time (PAT) is the time from the beginning of the RV ejection until the time to peak flow velocity across the pulmonary valve (Fig. 1). The use of the PAT, measured by TOE, has been described to estimate mean pulmonary artery pressure (MPAP) in cardiac surgical patients.5,6 The pulmonary artery systolic flow waveform changes with increasing pulmonary hypertension and increasing pulmonary vascular resistance.5,7 Using transthoracic echocardiography (TTE) in awake patients with normal pulmonary pressures, this systolic flow waveform has a dome-shaped appearance with a relatively slow upstroke, with a PAT of 134 to 153 ms (±20 to 30 ms).8,9 As pulmonary artery pressures increase, the systolic flow waveform changes to a more triangular appearance, and PAT decreases with increasing MPAP.5 Additional abnormal flow patterns with notching of the pulmonary systolic velocity waveform have been observed with pulmonary hypertension, particularly that caused by thromboembolic disease.10 Flow reversal has also been noted in the pulmonary artery.5,7

Fig. 1

Fig. 1

A study using TOE in cardiac surgical patients showed that patients with a PAT of less than 107 ms were likely to have pulmonary hypertension,6 as defined by a MPAP of more than 25 mmHg.11,12 A large TTE study showed that patients with a PAT of 105 ms or less were likely to have pulmonary hypertension.13 This cut-off of 105 ms is also recommended by the European Society of Cardiology in their recent guidelines.14

In general, the shorter the PAT, the more severe the pulmonary hypertension.15 The PAT has the benefit of being measurable in almost all patients, does not require TR as with RVSP and does not rely on the additional estimate of CVP.6,16 However, it is unclear exactly how to measure PAT. The American Society of Echocardiography guidelines are somewhat imprecise as to what view should be obtained, and where the exact location of the sample volume should be: the description simply states, ‘MPAP may be estimated by using PAT measured by pulsed Doppler of the pulmonary artery in systole’.15

In addition to pulmonary pressure and pulmonary vascular resistance, a number of factors may influence the pulsed Doppler waveform shape and the PAT; these include cursor position,17,18 the angle of alignment of the ultraosind beam to the direction of blood flow,15 extremes of heart rate (HR)13,15 and complex flow and flow reversal patterns.5,7,17,19 Neither is it clear which echocardiographic view should be used. In several TTE studies, the parasternal short axis view was used.16,20 In other studies, the views used are not well described.21 With TOE, the pulmonary valve can be imaged using either an oesophageal or transgastric view.5 A previously published study used either oesophageal or transgastric views to acquire the PAT, depending on which lined up best with the direction of blood flow, but did not compare sequential measurements between the two.6 We are unaware of any data specifically comparing different echocardiographic windows for this measurement. This is relevant as the peak velocities and velocity time integrals vary depending on optimal alignment with the jet: using the Doppler echocardiographic assessment of the aortic valve22 and the tricuspid valve for TR and RVSP15 as examples, the guidelines simply recommend obtaining envelopes from multiple windows.15,22

It has been demonstrated that patients with pulmonary hypertension have complex flow patterns rather than simple homogenous laminar flow and flow reversal can been seen in the pulmonary artery.5,7,17 Hence, the shape of the velocity waveform and time to peak velocity would depend on the exact position as well as the orientation of the sample volume relative to the pulmonary valve. As these vary with imaging window, it is unclear whether oesophageal or transgastric windows are equivalent when measuring PAT and whether one is a better predictor of pulmonary hypertension than the other.

Given the angle dependency of Doppler measurements, the different waveforms and sample volumes in the various views, and the complex flow patterns, it seems possible that PAT may vary systematically in different views. Consequently, we aimed to compare the relationship between PAT and MPAP from oesophageal and transgastric views.

Back to Top | Article Outline

Materials and methods

The study was approved by the Human Research and Ethics committee of St. Vincent's Hospital, Melbourne, Australia (109/15) as a low risk research activity on 4 December 2015 as part of a larger review of intra-operative TOE. Informed consent was waived given that PAC and TOE are both used routinely for all cardiac surgery cases in our institution.

Sixty-three patients undergoing cardiac surgery were included in this prospective study. All patients had insertion of both a TOE probe and PAC, as is routine in our centre. Patients were excluded if for whatever reason, they did not have insertion of both a PAC and TOE probe. Patients having pulmonary valve surgery, or with RV outflow tract (RVOT) obstruction were also excluded. No changes in haemodynamic management or additional invasive procedures were performed on patients as a result of study participation.

As part of a comprehensive TOE study, the RVOT and pulmonary valve were evaluated. The PAT was obtained by placing the pulsed wave Doppler cursor in the middle of the RVOT at the pulmonary annulus, just below the pulmonary valve leaflets.5,10,15,16,20,21 The position of the cursor indicates the position for the sample volume. The time from the onset of ejection to the peak flow velocity across the pulmonary valve was measured.5,16,20

The PAT was measured from oesophageal first then from the transgastric views (Fig. 2).

Fig. 2

Fig. 2

The PAT was measured in the operating room at the time of image acquisition on a representative spectral envelope, using the inbuilt software package on the echocardiography machine. All TOE studies were performed on an iE33 machine with the X7-2t probe (Philips Healthcare, Melbourne, Victoria, Australia). Measurements were repeated randomly on 24 patients to assess intra and interobserver variability. All measurements were performed by experienced cardiac anaesthetists with expertise and qualifications in peri-operative TOE. All measurements were performed at a sweep speed of 75 mm s−1.

Simultaneous measurements of the MPAP were taken from the PAC (Swan-Ganz Ref 83HF75; Edwards Lifesciences, Irvine, California, USA). All measurements were taken at end-expiration according to international guidelines,12 prior to cardiopulmonary bypass and during a period of haemodynamic stability. Patients were ventilated with a tidal volume of 6 to 8 ml kg−1 with a positive end-expiratory pressure of 5 cmH2O. Patients were in the supine position with pressure transducers located at the mid-thoracic position, halfway between the sternum and operating table.

Back to Top | Article Outline

Statistics

For the measurement of PAT from the upper oesophageal and transgastric views, linear regression was used to assess the linear correlation and Bland–Altman analysis was used to assess the limits of agreement. Cohen's kappa was used to assess agreement between the PAT measurements obtained from the two views in the diagnosis of pulmonary hypertension (MPAP ≥ 25 mmHg).

The area under receiver operating characteristic (ROC) curve was used to determine the ability of the two PAT measurements to discriminate between patients with and without pulmonary hypertension. Locally weighted regression (lowess command with bandwidth 0.5 option in Stata) was used to create a line of best fit for the relationship between MPAP and PAT in the two views. Bland–Altman analysis and absolute intraclass correlations coefficients were used to assess interobserver and intra-observer agreement. All statistical analyses were performed using Stata 14.1 (StataCorp, College Station, Texas, USA).

Back to Top | Article Outline

Results

Patient characteristics and surgical procedures are summarised in Table 1.

Table 1

Table 1

Simultaneous measurements of MPAP and PAT were taken in 63 patients. In two patients, these measurements were not possible in the transgastric position due to an inability to visualise the RVOT and pulmonary valve.

Figure 3 shows the relationship between the two measurements and their agreement in the prediction of pulmonary hypertension given a cut-off of 107 ms. The percentage agreement was 93.4% (57 out of 61) with a kappa of 0.85 [95% confidence interval (CI), 0.60 to 1.00].

Fig. 3

Fig. 3

The limits of agreement between PAT measured in the upper oesophageal view and the transgastric view are shown in the Bland–Altman plot (Fig. 4). The mean difference was 1.05 ms with limits of agreement of −18.4 to 16.3 ms.

Fig. 4

Fig. 4

Figure 5 shows the ROC for the ability of PAT, measured in the two views, to predict pulmonary hypertension. The area under the ROC for PAT measured in the upper oesophageal view was 0.99 (95% CI, 0.98 to 1.00) and measured in the transgastric view, 0.99 (95% CI, 0.97 to 1.00). Using a cut-off of 107 ms, the upper oesophageal view predicted pulmonary hypertension (defined as MPAP > 25 mmHg) with a sensitivity of 94.7% and specificity of 97.6%. The transgastric view predicted pulmonary hypertension with a sensitivity of 89.5% and specificity of 97.6%.

Fig. 5

Fig. 5

Figure 6 shows the relationship between the MPAP and the PAT in both oesophageal and transgastric views showing a near identical curvilinear relationship in the two views.

Fig. 6

Fig. 6

The intraclass correlation co-efficient (taking into account absolute differences) for interobserver agreement was 0.95 for upper oesophageal views and 0.98 for transgastric views. A Bland–Altman analysis of the interobserver agreement for the upper oesophageal PAT demonstrated a mean difference of −0.22 ms and limits of agreement of −20.4 to 20.0 ms and for the transgastric PAT demonstrated a mean difference of −1.7 ms and limits of agreement of −19.3 to 15.9 ms.

The intraclass correlation co-efficient (taking into account absolute differences) for intra-observer agreement was 0.96 for upper oesophageal views and 0.99 for transgastric views. A Bland–Altman analysis of the intra-observer agreement for the upper oesophageal PAT demonstrated a mean difference of −0.65 ms and limits of agreement of −12.5 to 11.2 ms and for the transgastric PAT demonstrated a mean difference of −0.48 ms and limits of agreement of −9.8 to 8.9 ms.

Back to Top | Article Outline

Discussion

The current study confirms that oesophageal and transgastric measurements of PAT have close agreement (bias of 1 ms, with narrow limits of agreement of −18 to 16 ms) and a similar high ability to discriminate between patients with and without pulmonary hypertension (areas under ROC curves of 0.96).

The relationship between PAT and MPAP is curvilinear (Fig. 6), with a close to inverse linear relationship between MPAP and PAT below 107 s and fairly flat for PATs greater than 107 ms. These data are similar to the results of previous studies in cardiac surgical patients and supports 107 ms as being a suitable cut-off for defining pulmonary hypertension.6,13 To our knowledge, the largest study using TTE in awake, spontaneously ventilating patients found a cut-off of 105 ms13 for diagnosis of pulmonary hypertension, as is used in European Guidelines.14 Our results compare very well with this number, suggesting that the PAT measured with TOE under anaesthesia and during positive pressure ventilation is an equally valid estimate of pulmonary hypertension.

We believe this is a valuable technique for estimating mean pulmonary arterial pressure in cardiac centres where a PAC is not used routinely and where there is insufficient TR to estimate RVSP on intra-operative echocardiography.

At present, measurement of the PAT is not specifically addressed as part of peri-operative TOE guidelines.23 The PAT is mentioned as part of guidelines for assessment of the right heart with TTE,15 but there is no specific mention of how best to obtain this measurement nor which acoustic window is ideal. This is not surprising, as we are unaware of any comparison data specifically addressing which echocardiographic acoustic window is most appropriate for this measurement.

Our data suggest that there are no additional advantages to be obtained by measuring PAT using both the upper oesophageal and transgastric windows, as they are comparable. The line-up between Doppler beam and flow across the pulmonary valve is not the same in the oesophageal and transgastric views, and, in general, this line up was more challenging to obtain and less optimal in transgastric views. However, despite this, there were no significant differences in PAT.

Transgastric view measurements of the PAT are not possible in a small percentage of patients and do not add additional value over oesophageal views. Thus, in a busy operating room with multiple tasks, our recommendation is that PAT should be measured in all patients using oesophageal views as these were obtained in 100% of patients.

Our study did not aim specifically to evaluate additional abnormal flow patterns such as flow reversal, or notching in the systolic velocity time integral. However, notching has predominantly been observed in patients with pulmonary hypertension from thromboembolic disease and, as far as we are aware, we did not have any patients with this underlying pathology. Similarly, we did not see flow reversal in the pulmonary artery.5,7,17 This is most likely because our cursor was placed in the RVOT at the pulmonary annulus level, just below the pulmonary valve leaflets, where a relatively competent pulmonary valve would prevent significant flow reversal. A study comparing cursor position in both the RVOT and the main pulmonary artery confirmed this, with flow reversal detected commonly in cardiac surgical patients in the main pulmonary artery (96% of patients) but not detected in the RVOT.17

It has long been recognised that cursor location of the pulsed wave Doppler influences the waveform shape and PAT.18 A TOE study in cardiac surgical patients showed significantly different PAT, depending on the cursor location for the pulsed wave Doppler.17 The PAT was significantly greater when the cursor was positioned in the RVOT just below the pulmonary valve leaflets and least when it was in the main pulmonary artery.17 More recent studies have placed the cursor at the pulmonary valve annulus.6,16 Variations in echocardiographic windows, cursor position and the pulsed wave Doppler alignment with flow all have the potential to alter the pulmonary flow measurements and the velocity time integral, and hence PAT.18 Our study confirms that the PAT remains consistent when measured from different echocardiographic windows when the cursor is placed in the middle of the RVOT at the pulmonary annulus, just below the pulmonary valve leaflets

Limitations of our study include the possibility that the pulmonary haemodynamics changed between the times when PAT was measured in the oesophageal and transgastric views. However, as all measurements were performed during stable haemodynamics and done sequentially this seems unlikely.

PAT does vary with extremes of HR, with shorter PAT in the presence of tachycardia. Some authors correct for HR by dividing the PAT by the square root of the RR interval on the ECG,24 but most do not.8,17,18,20,21,25 PAT and MPAP correlate well when the HR is between 60 to 100 bpm.13,26 The American and European guidelines do not recommend routine correction for HR when the HR is less than 100 bpm.15 None of our patients had extremes of HR and all had a HR of less than 100 at the time of PAT measurement. Corrected PAT may be required when the HR is more than 100.15,26 As our study was performed in a population of mixed cardiac surgical patients under general anaesthesia, these results may not apply to different populations under different circumstances.

Our patient cohort had either no (30%) or no more than trace of pulmonary regurgitation (70%). We do not suggest that there is no role for visualisation of the pulmonary valve in the transgastric view as part of a comprehensive TOE study in selected patients, we simply emphasise that there are no additional advantages to be gained by obtaining the PAT from this view, one that is often difficult, and sometimes unable, to be obtained.

Back to Top | Article Outline

Conclusion

Oesophageal and transgastric measurements of PAT have close agreement and a similarly high ability to discriminate between people with and without pulmonary hypertension. The transgastric measurement was unobtainable in a small percentage of patients and required more probe manipulation. When obtainable, we would recommend PAT measurement in the upper oesophageal view.

Back to Top | Article Outline

Acknowledgements relating to this article

Assistance with the study: none.

Financial support and sponsorship: none.

Conflicts of interest: none.

Presentation: none.

Back to Top | Article Outline

References

1. Greiner S, Jud A, Aurich M, et al. Reliability of noninvasive assessment of systolic pulmonary artery pressure by Doppler echocardiography compared to right heart catheterization: analysis in a large patient population. J Am Heart Assoc 2014; 3: http://jaha.ahajournals.org/content/ahaoa/3/4/e001103.full.pdf?download=true
2. Cowie B, Kluger R, Rex S, et al. The utility of transoesophageal echocardiography for estimating right ventricular systolic pressure. Anaesthesia 2015; 70:258–263.
3. Fisher MR, Forfia PR, Chamera E, et al. Accuracy of Doppler echocardiography in the hemodynamic assessment of pulmonary hypertension. Am J Respir Crit Care Med 2009; 179:615–621.
4. Soliman D, Bolliger D, Skarvan K, et al. Intra-operative assessment of pulmonary artery pressure by transoesophageal echocardiography. Anaesthesia 2015; 70:264–271.
5. Tousignant C, Van Orman JR. Pulmonary impedance and pulmonary Doppler trace in the perioperative period. Anesth Analg 2015; 121:601–609.
6. Cowie B, Kluger R, Rex S, et al. The relationship between pulmonary artery acceleration time and mean pulmonary artery pressure in patients undergoing cardiac surgery: an observational study. Eur J Anaesthesiol 2016; 33:28–33.
7. Hu R, Tousignant C, Chen R. Flow velocity patterns in the pulmonary artery and pulmonary hypertension. Can J Anaesth 2012; 59:716–719.
8. Bech-Hanssen O, Lindgren F, Selimovic N, et al. Echocardiography can identify patients with increased pulmonary vascular resistance by assessing pressure reflection in the pulmonary circulation. Circ Cardiovasc Imaging 2010; 3:424–432.
9. Dabestani A, Mahan G, Gardin JM, et al. Evaluation of pulmonary artery pressure and resistance by pulsed Doppler echocardiography. Am J Cardiol 1987; 59:662–668.
10. Hardziyenka M, Reesink HJ, Bouma BJ, et al. A novel echocardiographic predictor of in-hospital mortality and mid-term haemodynamic improvement after pulmonary endarterectomy for chronic thrombo-embolic pulmonary hypertension. Eur Heart J 2007; 28:842–849.
11. Galiè N, Hoeper MM, Humbert M, et al. Guidelines for the diagnosis and treatment of pulmonary hypertension. Eur Respir J 2009; 34:1219–1263.
12. Hoeper MM, Bogaard HJ, Condliffe R, et al. Definitions and diagnosis of pulmonary hypertension. J Am Coll Cardiol 2013; 62:D42–50.
13. Marra AM, Benjamin N, Ferrara F, et al. Reference ranges and determinants of right ventricle outflow tract acceleration time in healthy adults by two-dimensional echocardiography. Int J Cardiovasc Imaging 2017; 33:219–226.
14. Galie N, Humbert M, Vachiery JL, et al. 2015ESC/ERS guidelines for the diagnosis and treatment of pulmonary hypertension: The Joint Task Force for the Diagnosis and Treatment of Pulmonary Hypertension of the European Society of Cardiology (ESC) and the European Respiratory Society (ERS): Endorsed by: Association for European Paediatric and Congenital Cardiology (AEPC), International Society for Heart and Lung Transplantation (ISHLT). Eur Heart J 2016; 37:67–119.
15. Rudski LG, Lai WW, Afilalo J, et al. Guidelines for the echocardiographic assessment of the right heart in adults: a report from the American Society of Echocardiography endorsed by the European Association of Echocardiography, a registered branch of the European Society of Cardiology, and the Canadian Society of Echocardiography. J Am Soc Echocardiogr 2010; 23:685–713.
16. Yared K, Noseworthy P, Weyman AE, et al. Pulmonary artery acceleration time provides an accurate estimate of systolic pulmonary arterial pressure during transthoracic echocardiography. J Am Soc Echocardiogr 2011; 24:687–692.
17. Tousignant C, Van Orman JR. Pulmonary artery acceleration time in cardiac surgical patients. J Cardiothorac Vasc Anesth 2015; 29:1517–1523.
18. Panidis IP, Ross J, Mintz GS. Effect of sampling site on assessment of pulmonary artery blood flow by Doppler echocardiography. Am J Cardiol 1986; 58:1145–1147.
19. Reiter G, Reiter U, Kovacs G, et al. Magnetic resonance-derived 3-dimensional blood flow patterns in the main pulmonary artery as a marker of pulmonary hypertension and a measure of elevated mean pulmonary arterial pressure. Circ Cardiovasc Imaging 2008; 1:23–30.
20. Kitabatake A, Inoue M, Asao M, et al. Noninvasive evaluation of pulmonary hypertension by a pulsed Doppler technique. Circulation 1983; 68:302–309.
21. Granstam SO, Bjorklund E, Wikstrom G, et al. Use of echocardiographic pulmonary acceleration time and estimated vascular resistance for the evaluation of possible pulmonary hypertension. Cardiovasc Ultrasound 2013; 11:7.
22. Baumgartner H, Hung J, Bermejo J, et al. Echocardiographic assessment of valve stenosis: EAE/ASE recommendations for clinical practice. J Am Soc Echocardiogr 2009; 22:1–23.
23. Hahn RT, Abraham T, Adams MS, et al. Guidelines for performing a comprehensive transesophageal echocardiographic examination: recommendations from the American Society of Echocardiography and the Society of Cardiovascular Anesthesiologists. J Am Soc Echocardiogr 2013; 26:921–964.
24. Tossavainen E, Soderberg S, Gronlund C, et al. Pulmonary artery acceleration time in identifying pulmonary hypertension patients with raised pulmonary vascular resistance. Eur Heart J Cardiovasc Imaging 2013; 14:890–897.
25. Sohrabi B, Kazemi B, Mehryar A, et al. Correlation between pulmonary artery pressure measured by echocardiography and right heart catheterization in patients with rheumatic mitral valve stenosis (a prospective study). Echocardiography 2016; 33:7–13.
26. Chan KL, Currie PJ, Seward JB, et al. Comparison of three Doppler ultrasound methods in the prediction of pulmonary artery pressure. J Am Coll Cardiol 1987; 9:549–554.
© 2018 European Society of Anaesthesiology