Article Text

See original article:

Download PDFPDF
Non-invasive imaging
Intraoperative echocardiography
  1. Donald C Oxorn
  1. Professor Donald Oxorn, University of Washington, 1959 NE Pacific Street, Box 356540, Seattle, WA 98195-6540, USA; oxorn{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Over the past several decades, cardiac surgical practice has changed considerably. As the prominence of simple coronary artery bypass grafting (CABG) has waned, the complexity of cardiac surgical procedures has significantly increased. Patients are returning to the operating room (OR) to have revisions of their original procedures, such as those with dysfunctional prosthetic valves, occluded coronary grafts, and adults with congenital heart disease. Many valves in which prosthetic replacement was previously performed are now being repaired. In addition, the majority of patients present with a variety of comorbidities such as abnormalities of left and right ventricular function, diabetes, renal insufficiency, pulmonary hypertension, and cerebrovascular disease. Surgeons and anaesthetists are therefore requiring more sophisticated, real time assessments of procedural success.

The use of intraoperative transoesophageal echocardiography (TOE) began in the 1980s, with M mode and subsequently two dimensional TOE, and in this initial period was used almost exclusively for the monitoring of left ventricular systolic function before and after CABG. Since that time, TOE has become an integral part of the intraoperative management of patients presenting for a variety of cardiac surgical procedures, while maintaining its pre-eminence as a monitor of ventricular function. In 2003, a committee under the auspices of the American College of Cardiology (ACC), the American Society of Echocardiography (ASE), and the American Heart Association (AHA) presented guidelines for the use of echocardiography,1 including the use of intraoperative TOE (box 1), building on guidelines published by the American Society of Anesthesiologists (ASA), and the Society of Cardiovascular Anesthesiologists (SCA).2

Box 1 Evidence based guidelines for intraoperative transoesophageal echocardiography1

Class I: Conditions for which there is evidence and/or general agreement that a given procedure or treatment is useful and effective
  1. Evaluation of acute, persistent, and life threatening haemodynamic disturbances in which ventricular function and its determinants are uncertain and have not responded to treatment

  2. Surgical repair of valvular lesions, hypertrophic obstructive cardiomyopathy, and aortic dissection with possible aortic valve involvement

  3. Evaluation of complex valve replacements requiring homografts or coronary reimplantation, such as the Ross procedure

  4. Surgical repair of most congenital heart lesions that require cardiopulmonary bypass

  5. Surgical intervention for endocarditis when preoperative testing was inadequate or extension to perivalvular tissue is suspected

  6. Placement of intracardiac devices and monitoring of their position during port-access and other cardiac surgical interventions

  7. Evaluation of pericardial window procedures in patients with posterior or loculated pericardial effusions

Class II: Conditions for which there is conflicting evidence and/or a divergence of opinion about the usefulness/efficacy of a procedure or treatment
Class IIa: Weight of evidence/opinion is in favour of usefulness/efficacy
  1. Surgical procedures in patients at increased risk of myocardial ischaemia, myocardial infarction, or haemodynamic disturbances

  2. Evaluation of valve replacement, aortic atheromatous disease, the Maze procedure, cardiac aneurysm repair, removal of cardiac tumours, intracardiac thrombectomy, and pulmonary embolectomy

  3. Detection of air emboli during cardiotomy, heart transplant operations, and upright neurosurgical procedures

Class IIb: Usefulness/efficacy is less well established by evidence/opinion
  1. Evaluation of suspected cardiac trauma, repair of acute thoracic aortic dissection without valvular involvement, and anastomotic sites during heart and/or lung transplantation

  2. Evaluation of regional myocardial function during and after off-pump coronary artery bypass graft procedures

  3. Evaluation of pericardiectomy, pericardial effusions, and pericardial surgery

  4. Evaluation of myocardial perfusion, coronary anatomy, or graft patency

  5. Dobutamine stress testing to detect inducible demand ischaemia or to predict functional changes after myocardial revascularisation

  6. Assessment of residual duct flow after interruption of patent ductus arteriosus

Class III: Conditions for which there is evidence and/or general agreement that the procedure/treatment is not useful/effective and in some cases may be harmful
  1. Surgical repair of uncomplicated secundum atrial septal defect

The echocardiographer must recognise several factors germane to the intraoperative milieu:

  • Most general anaesthetic agents diminish vascular tone and depress myocardial contractility.

  • Positive pressure ventilation has numerous haemodynamic effects which have the potential to alter echocardiographic findings.

  • Cardiopulmonary bypass, especially when prolonged, has profound effects on vascular tone, and systolic and diastolic function.

This article discusses the use of intraoperative TOE, and its impact on patient outcome.


Pre-bypass assessment

Preoperative studies

The variety and acuity of diagnoses that patients bring to the OR has increased considerably.3 Virtually all patients presenting for valve surgery will have had a transthoracic echocardiogram (TTE) as part of their diagnostic workup, and taken in the context of the patient’s clinical presentation will be crucial in formulating the surgical plan. Notwithstanding, the echocardiographer should perform a complete pre-bypass TOE4 to confirm the diagnosis, to assess valve reparability, to look for other valvular abnormalities which may be related to or independent of the primary lesion, and to assess ventricular function.

The quantitation of valvular regurgitation in the OR differs from the echocardiography laboratory in several respects:

  • Time is constrained, and this may be particularly limiting in unstable patients. Quantitative tests which require complex calculations such as PISA (proximal isovelocity surface area) or regurgitant fraction/orifice area should only be used when severity of regurgitation cannot be confidently assessed.

  • Surgical stimulation and cardiac manipulation result in multiple arrhythmias which may necessitate analysing multiple beats.

  • Positive pressure ventilation and the depth of general anaesthesia have complex effects on loading conditions, which often make the degree of regurgitation appear less.5 This limitation can be circumvented by artificially altering loading conditions (vide infra). That being said, it is crucial that the intraoperative findings are taken in the context of preoperative quantitation of mitral regurgitation (MR), measurements of chamber size, and the patients preoperative clinical status.

Valvular regurgitation

The use of colour Doppler is central to the evaluation of valvular regurgitation, and the direction of the jet may help diagnose the underlying pathophysiology. However, several caveats apply. Numerous technical and physiologic factors such as colour gain, pulse repetition frequency and the driving pressure across the valve6 will affect jet size independently of the regurgitant orifice area. Eccentric jets appear smaller than central jets because they flatten out against the wall of the receiving chamber.6 Electrocautery, ubiquitous in the OR, interferes with the colour Doppler signal.

The vena contracta can be used to assess regurgitation severity if it can be imaged in conjunction with the proximal flow convergence area and the receiving chamber jet, and is measured perpendicular to the commissural line.7 The mid oesophageal long axis plane provides appropriate views of both mitral and aortic valves for this purpose. The vena contracta accurately quantifies both central and eccentric jets.6 8 For MR, its dependence on loading conditions necessitates that the measurement be made at baseline conditions. This most often involves volume loading and the use of a vasoconstrictor such as phenylephrine.5 9 The vena contracta appears less load dependent in aortic regurgitation (AR).8 The presence of multiple jets will require additional tests to quantitate regurgitation severity; the presence of systolic reversal in pulmonary veins indicates severe MR and the presence of pan systolic flow reversal in the descending aorta indicates severe AR.

Valvular stenosis

Because Doppler beam alignment is parallel to mitral inflow in most mid oesophageal views, TOE is ideally suited to the quantification of mitral stenosis with the use of mean gradients and pressure half time measurements. Intraoperative assessment of aortic stenosis involves the measurements of the transvalvular velocity and the left ventricular outflow tract (LVOT) velocity in either the transgastric or deep transgastric long axis imaging planes; with the addition of the LVOT diameter, the aortic valve area (AVA) can be calculated from the continuity equation.10 Care must be taken to ensure that the angle of insonation is appropriate and that loading conditions are as close to baseline as possible.

Valve repair

The frequency with which valve repair is performed varies with the aetiology, the valve or valves involved, surgical skill, and institutional referral patterns. The presence of comorbidities such as ischaemic heart disease and pulmonary hypertension may render the patient unable to withstand a second pump if the repair is unsuccessful.11 12

A comprehensive examination of the mitral valve can accurately guide the surgeon.3 13 It should include information about leaflet length, thickness, mobility, and calcification; whether leaflet motion is excessive, restrictive, or normal; annular diameter and calcification; subvalvular disease (best seen in the transgastric long axis view of the left ventricle (LV)); global and segmental LV function; and the quantity and direction of the regurgitant jet(s).14

The anatomy most conducive to valve repair is isolated posterior leaflet prolapse or flail in the setting of myxomatous MV disease; the most common repair technique is excision of the redundant portions of the leaflet, and the placement of an annuloplasty ring (fig 1A,B). In the presence of excessive posterior leaflet length, a sliding annuloplasty may be used to diminish the incidence of postoperative systolic anterior motion (SAM) with consequent MR and LVOT obstruction (vide infra)15 16 (fig 2).

Figure 1 (A) In panel A, the P2 portion of the posterior leaflet is noted to be flail, and colour Doppler in panel B shows an anteriorly directed jet which hugs the interatrial septum. In panel C, the surgeon has incised the right atrium and interatrial septum to demonstrate the flail segment of P2; on the underside of the leaflet are seen numerous torn chordae. The anterior leaflet is retracted and not seen. Panel D is a representation of the surgeon’s view of the normal mitral valve. AML, anterior mitral leaflet (B) The left hand panel shows the appearance of the valve after excision of the flail segment and closure of the posterior leaflet (double arrows) and placement of a ring annuloplasty. In the right hand panel, the postoperative transoesophageal echocardiogram shows normal leaflet coaptation and only trace mitral regurgitation (arrow).
Figure 2 In this transoesophageal echocardiogram of another patient who had mitral valve repair, panel A demonstrates systolic anterior motion (SAM) of the anterior mitral leaflet (arrow), and panel B shows the resultant mitral regurgitation (MR) and turbulence in the left ventricular outflow tract (LVOT). Discontinuation of inotropes and the administration of fluid and vasoconstrictors led to resolution of the SAM.

Repair of the anterior leaflet is more difficult, because it is in direct continuity with the fibrous skeleton of the heart; therefore techniques have focused on minimising the discrepancy in leaflet height. Chordae may be shortened by burying them more apically in the papillary muscle, transferred from the posterior leaflet, or synthetic chords may be created with polytetrafluoroethylene (PTFE) sutures. At a community hospital, Matsunaga and colleagues found that the use of intraoperative TOE led to an increased rate of successful repair of myxomatous mitral valve disease that often involved the anterior or both leaflets.17

In patients with mitral stenosis, open commissurotomy with or without ring annuloplasty may be attempted in those who are not candidates for percutaneous repair.18

Examination of the aortic leaflets, annulus, sinuses of Valsalva, sinotubular junction and ascending aorta can yield diagnostic information and allow assessment of the likelihood of valve repair.19 In experienced hands, AR secondary to root dilation can be managed with native aortic valve re-suspension.20

Ischaemic MR

The aetiology of ischaemic MR is complex. Frank papillary muscle rupture in the setting of myocardial infarction produces dramatic clinical and echocardiographic findings, but is rare; more often, a combination of leaflet tethering, annular dilation, and ventricular dysfunction are responsible.21 The assessment of ischaemic MR must take into account the loading conditions at the time of measurement, and the use of phenylephrine and volume loading to replicate preoperative conditions is recommended.22 The presence of moderate MR or greater is considered an indication for concurrent MV procedure at some centres, usually ring annuloplasty at the time of CABG; however, this current surgical strategy is suboptimal, as leaflet tethering secondary to LV dysfunction, a critical determinant of ischaemic MR, is rarely addressed. Lesser degrees of MR may improve with revascularisation alone, and this depends in large part on the extent of ventricular remodelling that has taken place.

Prosthetic valve failure

The reasons for prosthetic valve failure depend on the type of prosthesis and the time since implantation. Early failures are usually secondary to technical issues leading to the valve unseating from the annulus. Late failures in bioprosthetic valves are usually from leaflet degeneration, whereas in mechanical valves, the aetiology is most often restricted leaflet movement from either pannus or thrombus.


Patients with bacterial endocarditis presenting for surgery are usually managed with valve replacement, and more aggressive debridement if perivalvular extension is found. Secondary lesions may be repaired if there is minimal interference with valve function.

Associated valvular pathology

The presence of associated valvular pathology must be ascertained. In particular, tricuspid annular dilation with regurgitation secondary to left sided valve disease should be corrected with a remodelling annuloplasty at the time of surgery if the grade is greater than moderate.23 Some secondary valvular problems may improve with correction of the primary valvular lesion—for example, severe AR is often be accompanied by significant MR that improves following aortic valve replacement (AVR).

Post-bypass assessment

If the valve has been repaired, post-bypass TOE is the recognised standard in determining success.24 For mitral valve repair, there are several important considerations:

  • Residual regurgitation predicts an increased risk of reoperation.25

  • SAM with LVOT obstruction and residual MR may be apparent if the patient is hyperdynamic, hypovolaemic, or vasodilated. Administration of volume, increasing vascular tone, and decreasing inotropy will usually lead to resolution.

  • In the above instances, if the suggested manoeuvres do not lead to resolution, reinstitution of cardiopulmonary bypass is required, followed by further reparative techniques or valve replacement.

  • New stenosis following ring annuloplasty should be ascertained.

  • The change in MV area after open commissurotomy is not reliably measured by pressure half time, and other parameters such as mean pressure gradient must be examined.26

  • The adequacy of correction of secondary tricuspid regurgitation (TR) should be ascertained.

Following aortic valve repair, residual AR is often due to uneven suspension of the aortic leaflets within the neoaorta.

If valve replacement is undertaken, it is important to understand that each prosthesis has inherent pressure gradients and patterns of regurgitation. The echocardiographer must check for proper seating of the valve, free movement of the leaflets, and the absence of periprosthetic leaks (fig 3). Small periprosthetic leaks are common, and often resolve after heparin is reversed. The measurement of gradients across prosthetic valves immediately post-bypass is problematic: the haemodynamic milieu is rapidly changing; standard formulae are less reliable; and complex pressure gradients may develop within the prostheses. Despite these limitations, values for the “normal” pressure gradients across prosthetic valves have been published.27

Figure 3 Bileaflet mechanical mitral valve prosthesis. In panel A, both leaflets are fully open in systole (arrows). In panel B, both leaflets are fully closed in diastole. The colour Doppler image in panel C is in systole, and shows the “cleaning” jets (arrows) that are normal as shown. LA, left atrium; LAA, left atrial appendage.

With valve repair or replacement, there is always the potential for injury to surrounding structures. For example, annular stitches placed too deeply during mitral valve replacement may entrap the left or non-coronary cusps of the aortic valve leading to AR. Post-bypass TOE aids in detection and guides treatment in such instances.


Segmental function

Because it reflects myocardium perfused by the three major coronary branches, the transgastric mid papillary short axis view can be used as a starting point for monitoring segmental wall motion (SWM). More complete information is obtained using multiple views to evaluate basal and apical, as well as mid ventricular function, for all segments.4 There is evidence to suggest that deterioration in wall motion after cardiopulmonary bypass is a marker of adverse cardiovascular outcomes.28 Currently, TOE is the most sensitive test for intraoperative myocardial ischaemia though not the most specific; in addition to myocardial ischaemia, changes in SWM may be due to hypovolaemia, intracardiac conduction defects, tethering of non-ischaemic myocardium, or myocardial stunning after cardiopulmonary bypass. Specificity may be improved by the use of Doppler tissue velocity and strain rate imaging.29

Reference to preoperative imaging studies should establish a baseline for comparison.

In the pre-bypass period, acute changes in SWM should prompt therapeutic manoeuvres and, if refractory, an accelerated move towards cardiopulmonary bypass.

Following bypass, there are a number of potential causes for changes seen in SWM. Improvement may result from revascularisation of myocardium that had previously been ischaemic. Segmental deterioration may be the result of pre-bypass ischaemia, air passing into a graft or a native coronary artery (usually the right coronary artery (RCA)), mechanical obstruction of a vein graft, spasm of the internal mammary artery graft, or myocardial stunning from inadequate myocardial protection during the period of aortic cross clamping. TOE can be used to monitor the effects of corrective therapy.

Although numerous studies cite an impact of echocardiographic detection of SWM changes on ongoing management, there is a paucity of data suggesting that TOE during coronary revascularisation is useful in improving outcomes. In Savage’s study from the Cleveland Clinic,30 a group of high risk patients undergoing CABG were monitored using TOE, and compared to a group of historical controls is which intraoperative TOE was not used. Although outcomes were better in the TOE group, the methodology used in this study limits its generalised applicability.

Global ventricular function

TOE affords the echocardiographer the ability to assess the various components of global ventricular function, and may be considered a less invasive modality than pulmonary artery catheterisation.

Abnormalities of diastolic function are common in the setting of cardiac surgery, and can be used to predict difficulty in ventricular filling both before and after cardiopulmonary bypass. Traditional techniques of Doppler interrogation of pulmonary veins and mitral inflow can be supplemented by tissue Doppler imaging of mitral annular motion. Although ventricular volumes can be accurately determined with the use of complex formulae, most practitioners rely on transgastric short axis imaging to assess the patient’s volume status. It must be remembered that the presence of a small end-diastolic or end-systolic area does not always reflect decreased intravascular volume. Small left ventricular volumes can be seen with restrictions to filling as in pericardial disease, decreased right heart function, vascular dilation, and the administration of inotropic agents.

Ejection fraction is practically assessed by visually estimating the fractional area of change at the transgastric mid papillary level. A more formal assessment of ejection fraction using Simpson’s method of disks can be undertaken, but is time consuming, and not suited to the rapidly changing milieu of the OR. More recently, systolic mitral annular velocity as measured by tissue Doppler imaging has been proposed as an index of contractility; however, as with aforementioned methods, its dependence on loading conditions must be taken into account.

Using different anatomic sites, the stroke volume can be calculated by multiplying the cross sectional area of the conduit by the Doppler derived time–velocity integral; multiplying by heart rate yields the cardiac output. The anatomic site chosen is dependent on the quality of image acquisition and the absence of confounding regurgitant lesions.

The assessment of right ventricular function is crucial because dysfunction often occurs secondary to the inadequate delivery of cardioplegia and to the embolisation of intracardiac air down the right coronary artery following separation from cardiopulmonary bypass. Although the methods used for LV analysis can be employed, TOE assessment of intraoperative right ventricular function is best accomplished by visual inspection in both the transgastric and the four chamber views.

When sudden and unexplained hypotension occurs during any surgical procedure, TOE may be extremely helpful in establishing the diagnosis. After intraoperative cardiac arrest, Memtsoudis et al31 found that in 18 of 22 patients, the aetiology was detected by TOE, with subsequent management facilitated.


Screening for atheroma

Embolic stroke is a major source of morbidity and mortality following cardiac surgery. Atheromatous disease of the ascending aorta and arch are markers of increased risk, especially if plaques are >4–5 mm and are mobile. It is generally accepted that manipulation of the aorta in these instances may facilitate embolisation, and can be minimised by a strategy using both TOE and epiaortic scanning before aortic cannulation and cross clamping.32

Aortic disease

Patients presenting with an ascending aortic dissection are optimally managed with surgical repair. Many patients will have had a prior imaging study, but intraoperative TOE is helpful:

  • when the diagnosis is equivocal

  • if there is suspicion of aortic valve involvement

  • if haemodynamic instability suggests the presence of a pericardial effusion or myocardial dysfunction from occlusion of a coronary artery (fig 4).

Figure 4 In panel A, the surgeon has transected the ascending aorta, revealing the true lumen and the right coronary (RCA) ostium. In panel B, reflecting the intimal flap posteriorly exposes the false lumen; the outer aspect of the RCA intima is seen as it enters the medial layer of the aorta. In panel C, mid oesophageal long axis imaging of the ascending aorta shows the dissection flap at the RCA orifice. Intermittent occlusion of the RCA orifice resulted in inferior wall hypokinesis. A small pericardial effusion (PE) is noted anterior to the aorta.

Depending on the involvement of the aortic valve by the dissection, the valve may have to be replaced, or if the AR is secondary to sinotubular effacement, resuspension of the valve may suffice (fig 5).

Figure 5 In panel A, a dissection flap is seen in the ascending aorta. The sinotubular junction (double arrows) is noted to be effaced. In panel B, the leaflets fail to coapt during diastole. In panel C, a large jet of aortic regurgitation (AR) is noted. LMCA, left main coronary artery.


Intraoperative TOE may be used to assess the anastomoses in heart and lung transplantation, confirm the proper placement of intravascular devices, and help define the complex pre- and postoperative anatomy in adult congenital heart disease.

Following myomectomy in patients with hypertrophic obstructive cardiomyopathy, it is important to look for residual SAM and MR, examine the LVOT pressure gradient, and rule out an iatrogenic ventricular septal defect.

On occasion, a patent foramen ovale (PFO) is incidentally discovered during a procedure in which exposure of the inter-atrial septum is not planned (that is, CABG, AVR). If a right to left shunt is anticipated in the postoperative period, the PFO should probably be closed. This decision must take into account the longer bypass time required, and the necessity for bicaval cannulation and a right atriotomy. There is the backup of percutaneous closure if the indications at the time of surgery are equivocal.

If there are contraindications to TOE, a high frequency transducer in a sterile sheath may be placed intraoperatively on the surface of the heart and aorta.


Intraoperative TOE has improved monitoring capability, and allowed the results of complex operations to be assessed while still in the OR. The practitioner must be erudite, not only in his knowledge of echocardiography, but in the appreciation of the unique haemodynamic milieu that is the cardiac OR. Techniques in development include the use of tissue Doppler imaging to aid in the quantification of ventricular function, and “real time” three dimensional TOE to better define the complex anatomy that is often present in modern surgical practice.


  1. 1.
  2. Using evidence based principles, this consensus group outlined clinical applications of echocardiography, and in particular, the use of intraoperative TOE. It builds on guidelines authored by a task force of the American Society of Anesthesiologists and the Society of Cardiovascular Anesthesiologists.

  3. 2.
  4. 3.
  5. 4.

    This seminal paper describes in detail the views necessary for a complete intraoperative examination. Extra attention is paid to the complete assessment of the mitral valve.

  6. 5.
  7. 6.
  8. 7.

    The vena contracta is an important measurement in the assessment of valvular regurgitation, and this paper summarises the important technical aspects of its acquisition and interpretation.

  9. 8.
  10. 9.

    In 25 patients with established MR, the authors were able to demonstrate decreased MR following anaesthetic induction, and an increase in MR following afterload manipulation with phenylephrine. Their conclusion was that caution must be exercised when interpreting MR in the operating room.

  11. 10.
  12. 11.
  13. 12.
  14. 13.
  15. 14.
  16. 15.
  17. 16.
  18. 17.
  19. 18.
  20. 19.
  21. 20.
  22. 21.
  23. 22.
  24. 23.
  25. 24.
  26. 25.
  27. 26.
  28. 27.
  29. 28.

    In a series of 1400 patients, the authors demonstrated that the development of new postoperative wall motion abnormalities was associated with increased risk of long term adverse cardiac morbidity. This would appear to strengthen the case for increased use of TOE during high risk coronary bypass grafting.

  30. 29.
  31. 30.
  32. 31.
  33. 32.
View Abstract


  • Competing interests: In compliance with EBAC/EACCME guidelines, all authors participating in Education in Heart have disclosed potential conflicts of interest that might cause a bias in the article. The author has no competing interests.

Linked Articles

  • Correction
    BMJ Publishing Group Ltd and British Cardiovascular Society