Author + information
- Received January 23, 2003
- Revision received June 8, 2003
- Accepted June 23, 2003
- Published online January 7, 2004.
- Michael Vogel, MD, PhD*,
- Graham Derrick, MB, BS†,
- Paul A White, PhD†,
- Seamus Cullen, MB, CH*,
- Heidi Aichner, MD*,
- John Deanfield, MB, BS* and
- Andrew N Redington, MD, FRCP‡,* ()
- ↵*Reprint requests and correspondence:
Dr. Andrew N. Redington, Division of Cardiology, The Hospital for Sick Children, 555 University Avenue, Toronto M5G 1X8, Canada.
Objectives The aim of this study was to assess the utility of tissue Doppler echocardiography in the setting of repaired transposition of the great arteries when the right ventricle (RV) functions as the systemic ventricle.
Background Myocardial acceleration during isovolumic contraction, “isovolumic myocardial acceleration” (IVA), has been validated as a sensitive non-invasive method of assessing RV contractility. Although traditional indexes may be less valid for the abnormal RV, the relative insensitivity of IVA to an abnormal load makes it a potentially powerful clinical tool for the assessment of RV disease.
Methods We examined 55 controls and 80 patients (mean age 22 years) with transposition, who had undergone atrial repair at age 8 (0.3 to 72) months. A subgroup of 12 underwent cardiac catheterization. The RV systolic function was derived by analysis of pressure-volume relationships and IVA both at rest and during dobutamine stress. In all 80, myocardial velocities were sampled in the RV free wall.
Results During dobutamine (10 μg/kg/min for 10 min), the increase of IVA mirrored the increase in end-systolic elastance (r = 0.69, p < 0.02). In the group as a whole, IVA was reduced compared with the subpulmonary RV and the systemic left ventricle of controls. There was abnormal wall motion in 44 patients, which was associated with reduced IVA. Diastolic myocardial velocities were also abnormal but unrelated to the presence of wall motion abnormalities.
Conclusions The IVA can accurately assess changes in RV contractile function in patients with an RV as the systemic ventricle. Global long-axis RV function is reduced in patients with transposition, and this is associated with abnormal regional function.
The introduction in the late 1950s (1)and early 1960s (2)of the atrial switch operation markedly improved the prognosis of children with transposition of the great arteries (TGA). Although a 25-year actuarial survival of 75% has been reported (3), the very long-term ability of the right ventricle (RV) to support the systemic circulation remains unknown. A multitude of studies have described “abnormalities” of RV function; but longitudinal deterioration is difficult to demonstrate, and the assessment of drug therapies and the possible need for surgical intervention (e.g., conversion to arterial switch, or transplantation) (4–6)have not been possible using “traditional” non-invasive indexes of RV performance. This is because volume-based indexes are exquisitely dependent on the loading conditions of the RV. In patients with TGA, the RA myocardium is in the subaortic position, and the ventricular preload may be abnormal in patients with latent (e.g., exposed by exercise or dobutamine stress) (7)or manifest inflow obstruction and/or tricuspid valve incompetence. Our recently validated tissue Doppler echocardiography (TDE)-based index of RV contractile function, isovolumic myocardial acceleration (IVA), has been shown experimentally not only to measure contractile function accurately but also to be relatively independent of acute changes in ventricular preload and afterload (8). However, we have recently shown that abnormalities of diastolic function may be equally, or more important, determinants of functional performance in these patients (9). Tissue Doppler echocardiography can also describe diastolic myocardial events and may, therefore, offer potential new insights into the relationship between diastolic function and both regional and global systolic performance.
Thus, the purpose of this clinical study was to compare assessment of systemic RV function by IVA to invasive pressure-volume relationships, to assess the relationship between regional and global systolic RV performance, and finally, to assess TDE-derived diastolic performance in adults after Mustard and Senning procedures.
We studied 80 patients with complete TGA (situs solitus, d-loop, and d-transposition) and intact ventricular septum who had undergone atrial redirection by either the Mustard (n = 39) or Senning (n = 41) technique. These patients constituted all the patients seen during a one-year period (the duration of the research grant of G.D.) as in- or out-patients who were in sinus rhythm. The patients were studied at age 21.5 ± 13.8 (7.5 to 41.2) years. The age at time of the Senning operation was significantly younger than that at the time of a Mustard type of atrial redirection (6.2 ± 3.6 vs. 15.5 ± 11.8 months). The study protocol had been approved by the local institutional review board. Of the 16 patients who were asked to participate in the cardiac catheterization part of the protocol, four declined to participate on personal grounds.
Tissue Doppler echocardiography
Transthoracic imaging of the heart was performed using the GE Vingmed System V (GE Vingmed, Horten, Norway) with a frame rate between 98 and 178 Hz. The RV (RV and left ventricular [LV] free wall in controls) was separately imaged from an apical position, and color-coded myocardial velocities were recorded at the base immediately below the insertion of the atrioventricular valve leaflets. Recordings were made during apnea in the catheter laboratory; and, during held expiration in the ambulatory patients, tissue Doppler data were acquired with simultaneous electrocardiogram and phonocardiogram (GE Vingmed, Horten, Norway). A cine loop of at least three consecutive heartbeats was stored digitally for off-line analysis. Echopac software (GE Vingmed, Horten, Norway) was used to analyze the stored myocardial Doppler data. The peak myocardial velocities during isovolumic contraction, systole (S wave), early diastole (E wave), late diastole (A wave), and IVA (Fig. 1) were measured. Isovolumic relaxation time was measured from the onset of the second heart sound to the beginning of the myocardial E wave. Measurements of myocardial acceleration and velocities were performed on three consecutive heartbeats, and the average of the three measurements was calculated. Regional RV and LV function were also assessed by TDE using our previously described methods (10). A regional wall motion abnormality was defined by the presence of complete reversal of the myocardial velocity in systole and/or diastole (10). In addition to the direction of the velocity vectors, we assessed tissue tracking. This method, which is based on integration of the velocities in the longitudinal direction of the RV and LV wall, allows the assessment of the regional displacement of the myocardium from base to apex. In normal controls, the entire myocardium of the RV and LV free wall moves from base to apex in systole (Fig. 2). The distance of myocardial displacement is color-coded to facilitate recognition of regional wall motion abnormalities. Data in patients were compared to 55 age-matched controls studied in identical fashion.
Study preparation and protocol for invasive study
In a subset of 12 patients, we also evaluated contractile reserve during dobutamine infusion by analysis of pressure-volume relationships. All patients underwent conventional cardiac catheterization (Table 1). The conductance catheter study was performed after confirmation of acceptable hemodynamics or after a successful interventional procedure (7). Cardiac catheterization was performed under general anaesthesia. A custom-made 8 polar 6F conductance catheter (Millar Instruments, Houston, Texas) was placed into the apex of the RV via the aorta. The conductance electrodes were connected to a signal processing unit (Sigma 5DF, Cardiodynamics Corp., Leiden, the Netherlands). A 20-mm Latex balloon catheter (Boston Scientific, Boston, Massachusetts) was placed in the junction of the inferior vena cava and the systemic venous atrium and prepared for inflation to modify the preload. A standard 7F thermodilution catheter (Baxter Healthcare, Boston, Massachusetts) was placed in the pulmonary artery and connected to a dedicated cardiac output processing computer (Com2, Baxter, Edwards). The RA pressure-volume relationships and cardiac output measurements were obtained using our previously described methods (11). Measurements of pressure-volume relationships and cardiac output were performed at baseline and at the end of a 10-min infusion of 10 μg/kg/min of dobutamine, together with simultaneous TDE. End-systolic elastance was calculated off-line using our previously described technique by one examiner (P.W.) who was blinded to the data obtained by TDE, which were analyzed by a different observer (M.V.).
Changes in hemodynamic parameters between rest and dobutamine stress were compared using a paired Student ttest. Linear regression analysis was used to assess the relationship between changes in end-systolic elastance and changes in IVA. The null hypothesis was rejected when p < 0.05.
Comparison of IVA and conductance-derived contractile function
At rest, there was no significant correlation between IVA and end-systolic elastance, but during dobutamine infusion, a significant increase in IVA and end-systolic elastance was observed in all patients. There was a significant correlation (r = 0.69, p < 0.02) between the increase in IVA and the elastance during dobutamine infusion (Table 1, Fig. 3).
Systolic ventricular performance
Systemic RV IVA was reduced in TGA patients when comparing both to the LV and the RV of age-matched controls (Table 2). Regional wall motion abnormalities were found in 44 patients (Fig. 4). They were restricted to the apical third of the RV in 13 and extended to the middle segment of the RV lateral wall in the remainder (Table 3). No patient had basal incoordination. The age of patients with wall motion abnormalities was greater and their IVA was significantly lower than that of patients with “normal” wall motion (Table 3). There was no difference in the incidence of wall motion abnormalities when the Mustard and Senning patients were compared.
Diastolic ventricular performance
Compared with normals, the systemic RV in TGA had reduced early (p < 0.0001) and late (p < 0.0001) diastolic myocardial velocities (Table 2). The isovolumic relaxation time was prolonged in comparison with a normal RV (p < 0.007) but not in comparison with a normal LV. Furthermore, 20 of the patients (25%) had no detectable myocardial velocities related to atrial contraction. The A wave myocardial velocities were absent in 17 of 39 Mustard and three of 41 Senning patients (p < 0.02). The IVA in patients with absent A waves (1 ± 0.5 m/s2) was not different from the IVA in patients with A waves (1 ± 0.4 m/s2) on the myocardial Doppler trace (p < 0.49).
This study demonstrated that IVA can be used, in a way similar to invasive measurements of load-independent indexes, to describe the response of RV contractility to dobutamine stress in patients with a systemic RV. Furthermore, IVA was significantly lower at rest than either the normal RV or LV, irrespective of the presence of regional wall motion abnormalities. However, when present, abnormal wall motion was associated with significantly lower basal IVA levels. Abnormal diastolic function was unrelated to these systolic abnormalities and primarily manifest as markedly reduced early and late filling-phase myocardial velocities, the latter of which were particularly common after the Mustard procedure.
The TDE data in our current study confirm reduced systolic contractile function in the systemic RV of Mustard and Senning patients. The ejection phase indexes, S wave velocity and acceleration, were both reduced. This is in keeping with previous studies of ejection-phase indexes such as tricuspid annular motion assessed by echocardiography (12)or ejection fraction measured by echocardiography (13), radionuclide methods (14), angiography (15), or magnetic resonance imaging (16). All these measurements are subject to modification as a result of the subaortic position of the (systemic) RV and abnormalities of ventricularpreload (7). This makes interpretation of their relevance and of any longitudinal change difficult. However, IVA was also markedly abnormal. Although we have not validated the load independence of this index in a model of subaortic RV, we have shown its relative resistance to acute changes in ventricularafterload and preload in the normal RV (8)and LV (17).
Furthermore, in our subset of 12 patients with simultaneous conductance catheter measurements, IVA correlated with the change in elastance associated with dobutamine stress. It was also clear that additional wall motion abnormalities are associated with a further reduction in contractile performance, but interestingly, not in diastolic performance. The cause of wall motion abnormalities is unknown, but we (15)and others (18)have previously demonstrated their presence pre-operatively, presumably resulting from pre-operative myocardial damage (19). The relationship between RV wall motion and the frequently observed perfusion abnormalities in the systemic RV (20)is also poorly understood, although two previous studies have shown a relationship between resting and induced RV perfusion defects and global function (20,21). Tissue Doppler echocardiography evaluation of regional function may facilitate an examination of the relationship between non-invasively assessed perfusion and function in these hearts. A perhaps more important role of TDE will be in the longitudinal assessment of systolic performance. It has been difficult to demonstrate a temporal decline in RV performance in these patients, and one study of ejection fraction suggested that it remains unchanged over a four-year follow-up (4). Nonetheless, there remain significant concerns that RV failure will ensue in the long-term. Traditional indexes of RV function derived from angiography (15), echocardiography (13), and magnetic resonance imaging (16)have historically failed to provide the sensitivity required to track RV functional decline, if it is indeed occurring in these patients. This is almost certainly because much of the “dysfunction” detected by these techniques is adaptive. In our recent conductance catheter study (9), we showed appropriate responses to dobutamine in a variety of systolic indexes, and none of them, either load-dependent or load-independent, predicted functional performance. It is clearly impractical to perform invasive studies longitudinally, but basal IVA and its response to dobutamine stress may in the future allow monitoring of intrinsic myocardial functional reserve in a far more robust fashion.
Our TDE data are also relevant to our previous observations (9). We showed that abnormal stroke volume responses during exercise and dobutamine stress were not explained by reduced chronotropic or contractile responses but were associated with fixed ventricular filling rates. We speculated that non-physiologic atrial pathways were responsible for this abnormal atrioventricular coupling (9). The current study reinforces the potential role of abnormal diastolic function in these patients (22). Prolonged isovolumic relaxation time and reduced myocardial lengthening during early rapid filling may occur in any ventricle in which there are wall motion abnormalities. Furthermore, a possibledisruption of normal atrial function is made more likely by the finding of reduced A wave velocities. Despite electrical activity in all, mechanical atrial activity manifest as A wave myocardial lengthening was absent in 25%.
We have examined RV performance only in its long axis. Although it may have been interesting to make an additional assessment of radial function, this is technically impractical using TDE, because of the difficulty in defining a true short-axis section to these RVs, and is perhaps more appropriate to magnetic resonance tagging protocols (23). Although this index may provide a tool for the longitudinal assessment of RV systolic function and its long-term viability, we cannot provide repeated, long-term measurements with robust outcome data in this study. We do not have any data on the effect of dobutamine stress on tricuspid valve regurgitation. It is conceivable that its degree changed during the course of the study. The use of indexes of ventricular function, which are less load-dependent than ejection phase indexes, makes this less of an issue than in most studies, however. Finally, because of the relatively small number of patients undergoing cardiac catheterization, the hemodynamic responses described here may not be applicable to all patients after these operations. This should not detract from the validity of the comparative data between invasive indexes and IVA, however.
Tissue Doppler echocardiography provides additional insights into global and regional RV performance. Abnormal systolic performance is characterized by reduced IVA, which is further attenuated by the presence of regional wall motion abnormalities. Abnormalities of early and late diastolic events may be explained by the presence of abnormal atrial pathways with reduced atrial mechanical activity.
- isovolumic myocardial acceleration
- left ventricle/ventricular
- right ventricle/ventricular
- tissue Doppler echocardiography
- transposition of the great arteries
- Received January 23, 2003.
- Revision received June 8, 2003.
- Accepted June 23, 2003.
- American College of Cardiology Foundation
- Wong K.Y.,
- Venables A.W.,
- Kelly M.J.,
- Kalff V.
- Vogel M.,
- Schmidt M.R.,
- Kristiansen S.,
- et al.
- Vogel M.,
- Sponring J.,
- Cullen S.,
- et al.
- Derrick G.P.,
- Josen M.,
- Vogel M.,
- Henein M.Y.,
- Shinebourne E.A.,
- Redington A.N.
- Vogel M.,
- Gutberlet M.,
- Dittrich S.,
- et al.
- Redington A.N.,
- Rigby M.L.,
- Oldershaw P.,
- Gibson D.G.,
- Shinebourne E.A.
- Rees S.,
- Somerville J.,
- Warnes C.,
- et al.
- Vogel M.,
- Cheung M.M.,
- Li J.,
- et al.
- Trowitzsch E.,
- Colan S.D.,
- Sanders S.P.
- Lubiszewska B.,
- Gosiewska E.,
- Hoffmann P.,
- et al.
- Millane T.,
- Bernard E.J.,
- Jaeggi E.,
- et al.
- Reich O.,
- Voriskova M.,
- Ruth C.,
- et al.