OBJECTIVES: The relationship between exercise capacity and right ventricular (RV) components function in repaired tetralogy of Fallot patients with severely dilated right ventricles is poorly understood. The aim of this study was to characterize the exercise capacity and its relationship to RV global and components function in repaired tetralogy of Fallot patients with RV end-diastolic volume index  >150 ml/m2, a currently accepted threshold for pulmonary valve replacement.

METHODS: The medical records and results of cardiac magnetic resonance imaging and cardiopulmonary exercise testing of 25 consecutive eligible patients were reviewed. Twenty age- and gender-matched normal subjects were enrolled as cardiac magnetic resonance control. End-diastolic, end-systolic and stroke volumes, and ejection fraction (EF) were determined for the total RV and its components.

RESULTS: Of the 25 patients, 44% maintained normal exercise capacity. RV outlet EF was higher (P = 0.02) and RV incisions smaller (P = 0.04) in patients with normal exercise capacity than those with subnormal exercise capacity. Predicted peak oxygen consumption correlated better with the RV outflow tract EF than with the EF of other components of the RV or the global EF (r = 0.59; P = 0.002). Multivariate analysis showed the RV outflow tract EF to be the only independent predictor of exercise capacity (ß = 0.442; P = 0.02).

CONCLUSIONS: Exercise capacity is preserved in some tetralogy of Fallot patients with severe RV dilatation. RV outflow tract EF is independently associated with exercise capacity in such patients, and could be a reliable determinant of intrinsic RV performance.

INTRODUCTION

The first two authors contributed equally to this work.

Pulmonary regurgitation is very common late after the repair of tetralogy of Fallot (TOF). Chronic exposure of the right ventricle to a regurgitant load leads to progressive right ventricular (RV) dilatation, with increase in symptoms and in the risk of arrhythmias and sudden death [1].

The optimal timing of pulmonary valve replacement in TOF patients is a point of contention, especially in asymptomatic patients [2]. Severe pulmonary regurgitation with moderate-to-severe RV enlargement has been suggested as an indication for pulmonary valve replacement [3]. In clinical practice, a right ventricular end-diastolic volume index (RVEDVi) of >150 ml/m2 is a commonly used threshold for recommending pulmonary valve replacement in adults with repaired TOF [4]. However, surgical decision making based on volumes alone remains controversial. Additional information on the factors influencing progression of RV dysfunction can greatly aid decision making. Exercise tolerance, objectively assessed by cardiopulmonary exercise testing, provides important prognostic information [5, 6]. A poor exercise capacity indicates increased risk for death or sustained ventricular tachycardia, as well as cardiac-related rehospitalizations [6–8]. The results of cardiopulmonary exercise testing could therefore help the surgeon decide the timing of interventions and reinterventions.

Compared to the left ventricle, the right ventricle has a complex anatomy. From an embryological and anatomic perspective, the right ventricle can best be considered as a tripartite structure—with an inlet, an apical trabecular region, and an outlet [9, 10]—each having different volumetric and functional responses to volume overload in patients with repaired TOF [11, 12]. In asymptomatic patients with severe RV dilatation and pulmonary regurgitation, it is unknown which RV component function is associated, and to what extent, with decreased exercise capacity. A better understanding of this will provide insights into the anatomy of the right ventricle and the relationship of different components with clinical outcomes, and may eventually allow better risk stratification in specific groups of repaired TOF patients.

The present study was designed to (i) clarify how exercise capacity is affected in asymptomatic repaired TOF patients with severe RV dilatation (i.e. RVEDVi >150 ml/m2); and (ii) determine the relationship between different RV components function and exercise capacity in these patients.

MATERIALS AND METHODS

Patients

This single-centre retrospective study was approved by the ethical committee of West China Hospital, Sichuan University. The need for obtaining informed consent was waived because of the retrospective study design.

Between 1 January 2013 and 31 December 2015, a total of 25 patients were recruited into the present study. Patients were eligible for inclusion if they had: (i) undergone TOF repair for average 14.1 ± 3.6 years ago; (ii) completed a standardized cardiac magnetic resonance (CMR) imaging examination protocol and a cardiopulmonary exercise test at our hospital; (iii) undergone detailed clinical evaluation by follow-up in our centre and been reported as asymptomatic; and (iv) severe RV dilatation (RVEDVi >150 ml/m2) and pulmonary regurgitation (regurgitation fraction ≥25%). Patients were excluded if (i) they had undergone previous pulmonary valve replacement; (ii) the image quality of CMR was inadequate for analysis; (iii) they had failed to complete the cardiopulmonary exercise test with best effort; and (iv) residual right ventricular outflow tract (RVOT) obstruction was >15 mmHg at last echocardiographic examination.

In addition, 20- age and gender-matched normal controls received CMR examination were included as controls for CMR parameter comparison. Normal controls were selected from our normal controls study database which was approved by our institutional review board. All normal controls (or their parents if a patient was younger than 18 years old) provided written informed consent before participation.

Patient data, including demographic characteristics, surgical details of the primary repair and follow-up echocardiographic assessment results were retrospectively obtained from our TOF follow-up database. The results of the latest cardiopulmonary exercise tests were recorded. Cardiac rhythm, conduction abnormalities and QRS duration were identified from a 12-lead electrocardiogram. Symptoms of heart failure (New York Heart Association [NYHA] functional class) and 6-min walking distance were documented.

Cardiopulmonary exercise test

The cardiopulmonary exercise test was performed on a bicycle ergometer (EC3000; Customed, Ottobrunn, Germany), with a total 12-min protocol that included four phases, as follows: (i) a 2-min period during which the patient remained stationary on the cycle; this was intended to get the patient accustomed to the equipment; (ii) a 3-min period of cycling against no resistance; this was intended to identify any confounding factors (e.g. anxiety-driven hypertension); (iii) a 5-min period of cycling against continual and ramped resistance to pedalling. Velocity was fixed at 60 cycles per minute, and resistance was increased (taking into consideration the patient’s age, sex and comorbidities). Protocols were designed to produce the maximum predicted work for individuals in the 5 min; and, finally, (iv) a 2-min period of pedalling against no resistance, with continued monitoring of variables. A 12-lead electrocardiogram (Custo Cardio 200, Kontron Bildanalysis, Munich, Germany) and pulse oximetry were recorded continuously throughout the test. Cuff blood pressure was determined every 2 min.

The exercise test included a breath-by-breath gas exchange analysis, with measurement of oxygen uptake (VO2), carbon dioxide output (VCO2) and minute ventilation (VE). The anaerobic threshold was defined as the VO2 at which the respiratory exchange ratio reached 1.0 and there was precipitous and sustained rise in VCO2.

Exercise capacity was defined as the peak oxygen consumption measured by the exercise test. Because of the wide variations in patient age and size, analyses focused on percentage of predicted (rather than absolute) values for peak oxygen consumption. Subnormal exercise capacity was defined as a peak VO2 <85% of predicted [13].

Cardiac magnetic resonance

Data acquisition

CMR scans were performed on a 3.0-T MRI scanner (Magnetom Trio Tim, Siemens Healthcare, Erlangen, Germany) using an eight-channel phased-array body coil. Scanning was performed according to the standard imaging protocol for patients with repaired TOF [13]. The protocol consisted of a stack of consecutive short-axis views covering the left ventricle from base to apex and three long-axis views (two-, three- and four-chamber) by using a steady-state free-precession sequence (time of repetition = 3.4 ms, time of echo = 1.3 ms, flip angle 50, maximal field of view 340 mm, matrix size 256 × 144, slice thickness 8 mm with no gap, temporary resolution 40 ms and in-plane resolution 1.3–1.5 mm); this was preceded by steady-state free-precession cine transverse views, with the same parameters as the short-axis views, covering the whole heart—from the pulmonary bifurcation to just below the diaphragm. All images were acquired by the electrocardiogram-gated breath-hold technique or by the electrocardiogram-gated free-breathing technique, with average of tree signals, typically achieving 25 phases per heart cycle.

Analysis of the right ventricular global function and systolic function of different components

All measurements on the CMR images were made using dedicated CMR post-processing software (Qmass 7.6, Medis Imaging Systems, Leiden, Netherlands). The RV global and component volumes were measured by manually tracing the endocardial border at the end diastole and the end systole on successive axial cine images, with papillary muscle included in the ventricular volume. The inlet component was deemed to extend from the atrio-ventricular junction to the attachments of the tension apparatus of the tricuspid valve; the apical trabecular component from beyond the attachments of the tricuspid valve tension apparatus to the mouth of the outlet part; and the outlet part from this distal border of the apical part to the attachments of the leaflets of the pulmonary valve (Fig. 1) [11]. Effective RV stroke volume (RVSV) was calculated as the difference between the total RV forward flow and the pulmonary regurgitant flow. All the volumetric parameters were indexed to the body surface area.

Figure 1

Three-dimensional reconstruction of the tomographic MR images showing the right ventricular components (front view [left] and back view [right]).

Figure 1

Three-dimensional reconstruction of the tomographic MR images showing the right ventricular components (front view [left] and back view [right]).

Statistical analysis

Continuous variables were expressed as means ± SD or as absolute numbers (with percentage). Comparisons between groups were performed using the Student’s t-test. Pearson correlation coefficient was used to assess the relationship between clinical characteristics or functional parameters derived by MRI and the exercise tolerance. Variables significant at the 0.1 level were entered into multivariate linear regression analysis to evaluate their independent effects on exercise capacity. Multivariate linear regression analysis was further used to evaluate the independent effects of patient-related and procedure-related factors or functional parameters on exercise capacity. Only variables significant at the 0.1 level were retained in the final models. Two-tailed P < 0.05 indicated statistical significance. Statistical analysis was performed with SPSS for Windows, version 21 (IBM Corp., Armonk, NY, USA).

RESULTS

Demographic and clinical characteristics

The characteristics of the 25 patients included in the study are shown in Table 1. All patients had undergone primary correction in their childhood and, at the time of the study, were classified as NYHA class I or II. None had additional congenital cardiac malformations, ischaemic heart disease, or arrhythmias. Transannular patch repair had been performed in 22 (88%) patients, while valvectomy had been conducted in 3 (12%) patients.

Table 1

Demographic and clinical data for all patients and for patients grouped by percentage predicted peak oxygen consumption

 All patients  (n = 25) %Pred O2 <85%  (n = 14) %Pred O2 >85%  (n = 11) P-value VO2 lower versus normala 
Age 17.4   ± 4.3 17.9  ± 3.7 17.0  ± 4.8 0.64 
Males  (%) 17   (68%) 11  (78%) 6  (54.5%) 0.22 
Age at repair  (years) 3.4  ± 2.9 3.5  ± 3.0 3.2  ± 2.8 0.79 
Years since repair  (years) 14.1  ± 3.6 13.6  ± 4.1 14.7  ± 3.0 0.44 
Pulmonary annular Z-value −3.5  ± 1.2 −3.6  ± 0.9 −3.4  ± 1.1 0.52 
Pulmonary valve morphology     
 Bicuspid  (%) 22  (88%) 12  (85.7%) 10  (90.9%)  
 Tricuspid  (%) 3  (12%) 2  (14.3%) 1  (9.1%)  
 Others  
Primary repair  (%) 25  (100%) 14  (56%) 11  (44%)  
Transannular patch  (%) 22  (88%) 12  (85.7%) 10  (90.9%)  
Valvectomy  (%) 3  (12%) 2  (14.3%) 1  (9.1%)  
Length of ventriculotomy  (mm) 8.4  ± 2.3 12.2  ± 3.1 7.2  ± 1.5 0.04 
Tricuspid regurgitation by echocardiography    0.35 
 None  (%) 19  (76%) 11  (78.5%) 8  (81.8%)  
 Mild  (%) 6  (24%) 3  (21.5%) 3  (18.2%)  
 Moderate  
 Severe  
NYHA    0.19 
 I 14  (52%) 7  (50%) 6  (54.5%)  
 II 11  (48%) 7  (50%) 5  (45.5%)  
 III  
 IV  
Arrhythmia  
QRS duration  (ms) 154.7  ± 30.9 145.3  ± 32.2 153.2  ± 31.7 0.83 
6-min walk distance  (m) 489.3  ± 62.6 486.5  ± 77.4 490.2  ± 51.8 0.63 
 All patients  (n = 25) %Pred O2 <85%  (n = 14) %Pred O2 >85%  (n = 11) P-value VO2 lower versus normala 
Age 17.4   ± 4.3 17.9  ± 3.7 17.0  ± 4.8 0.64 
Males  (%) 17   (68%) 11  (78%) 6  (54.5%) 0.22 
Age at repair  (years) 3.4  ± 2.9 3.5  ± 3.0 3.2  ± 2.8 0.79 
Years since repair  (years) 14.1  ± 3.6 13.6  ± 4.1 14.7  ± 3.0 0.44 
Pulmonary annular Z-value −3.5  ± 1.2 −3.6  ± 0.9 −3.4  ± 1.1 0.52 
Pulmonary valve morphology     
 Bicuspid  (%) 22  (88%) 12  (85.7%) 10  (90.9%)  
 Tricuspid  (%) 3  (12%) 2  (14.3%) 1  (9.1%)  
 Others  
Primary repair  (%) 25  (100%) 14  (56%) 11  (44%)  
Transannular patch  (%) 22  (88%) 12  (85.7%) 10  (90.9%)  
Valvectomy  (%) 3  (12%) 2  (14.3%) 1  (9.1%)  
Length of ventriculotomy  (mm) 8.4  ± 2.3 12.2  ± 3.1 7.2  ± 1.5 0.04 
Tricuspid regurgitation by echocardiography    0.35 
 None  (%) 19  (76%) 11  (78.5%) 8  (81.8%)  
 Mild  (%) 6  (24%) 3  (21.5%) 3  (18.2%)  
 Moderate  
 Severe  
NYHA    0.19 
 I 14  (52%) 7  (50%) 6  (54.5%)  
 II 11  (48%) 7  (50%) 5  (45.5%)  
 III  
 IV  
Arrhythmia  
QRS duration  (ms) 154.7  ± 30.9 145.3  ± 32.2 153.2  ± 31.7 0.83 
6-min walk distance  (m) 489.3  ± 62.6 486.5  ± 77.4 490.2  ± 51.8 0.63 
a

The P-value for comparison of subnormal exercise capacity patients with normal exercise capacity patients.

Of the 25 patients, 11 had normal exercise capacity and 14 had subnormal exercise capacity. The incisions on right ventricles were significantly longer in patients with subnormal exercise capacity than in those with normal exercise capacity. There was no significant difference in gender, age at the time of evaluation, age at primary repair, type of repair, Z-value of pulmonary annular diameter, pulmonary valve morphology, degree of tricuspid valve regurgitation, NYHA class, QRS duration and 6-min walking distance between patients with normal and subnormal exercise capacity.

Exercise performance

All patients exercised until exhaustion. The mean respiratory exchange ratio was 1.11 ± 0.08. All patients exercised above their anaerobic thresholds. No arrhythmias were provoked during exercise. The results of exercise testing are given in Table 2.

Table 2

Cardiopulmonary exercise test data for all patients and for patients grouped by percentage predicted peak oxygen consumption

 All patients (n = 25) %Pred O2 <85% (n = 14) Predicted VO2 >85% (n = 11) P-value VO2 lower versus normala 
Peak VO2 (ml/kg/min) 35.8   ± 6.1 36.3  ± 5.4 35.2  ± 7.0 0.65 
Predicted peak VO2 (%) 76.4  ± 8.7 69.8  ± 4.7 89.7  ± 2.9 <0.001 
Peak heart rate (bpm) 168.3  ± 15.5 164.8  ± 18.8 172.8  ± 8.9 0.21 
Predicted peak heart rate (%) 84.7  ± 8.9 81.2  ± 9.8 89.2  ± 4.9 0.02 
VO2 at anaerobic threshold (ml/kg/min) 21.0  ± 4.2 20.8  ± 4.5 21.2  ± 4.0 0.81 
%Pred O2 at anaerobic threshold (%) 58.9  ± 7.9 56.9  ± 9.1 60.8  ± 6.7 0.31 
HR at anaerobic threshold (bpm) 115.2  ± 16.4 117.0  ± 19.3 112.8  ± 12.3 0.54 
Respiratory exchange rate 1.11  ± 0.08 1.09  ± 0.07 1.13  ± 0.09 0.23 
VE/VCO2 at anaerobic threshold 31.2  ± 4.3 32.8  ± 4.9 29.0  ± 2.6 0.04 
VE/VCO2 at peak 29.5  ± 3.8 31.2  ± 4.0 27.5  ± 2.6 0.02 
 All patients (n = 25) %Pred O2 <85% (n = 14) Predicted VO2 >85% (n = 11) P-value VO2 lower versus normala 
Peak VO2 (ml/kg/min) 35.8   ± 6.1 36.3  ± 5.4 35.2  ± 7.0 0.65 
Predicted peak VO2 (%) 76.4  ± 8.7 69.8  ± 4.7 89.7  ± 2.9 <0.001 
Peak heart rate (bpm) 168.3  ± 15.5 164.8  ± 18.8 172.8  ± 8.9 0.21 
Predicted peak heart rate (%) 84.7  ± 8.9 81.2  ± 9.8 89.2  ± 4.9 0.02 
VO2 at anaerobic threshold (ml/kg/min) 21.0  ± 4.2 20.8  ± 4.5 21.2  ± 4.0 0.81 
%Pred O2 at anaerobic threshold (%) 58.9  ± 7.9 56.9  ± 9.1 60.8  ± 6.7 0.31 
HR at anaerobic threshold (bpm) 115.2  ± 16.4 117.0  ± 19.3 112.8  ± 12.3 0.54 
Respiratory exchange rate 1.11  ± 0.08 1.09  ± 0.07 1.13  ± 0.09 0.23 
VE/VCO2 at anaerobic threshold 31.2  ± 4.3 32.8  ± 4.9 29.0  ± 2.6 0.04 
VE/VCO2 at peak 29.5  ± 3.8 31.2  ± 4.0 27.5  ± 2.6 0.02 
a

The P-value for comparison of subnormal exercise capacity patients with normal exercise capacity patients.

The VE/VCO2 showed a consistent change with peak VO2. VE/VCO2 at anaerobic threshold and at peak in subnormal exercise capacity patients was significantly higher than that in normal exercise capacity patients. Consequently, patients with subnormal exercise function had prominent chronotropic incompetence, with lower predicted peak heart rate. The differences in peak heart rate and heart rate at anaerobic threshold were not significant.

Assessment of right ventricular global systolic function

As shown in Table 3, the mean RV end-diastolic volume index (EDVi) and end-systolic volume index (ESVi) were significantly higher in patients than in normal controls. The mean RV ejection fraction (EF) was significantly lower in patients than in normal controls; however, the left ventricular (LV) systolic function was similar in patients and normal controls. There was no significant difference in RVEDVi, RVESVi, RVEF, pulmonary regurgitation fraction, RVSVi and left ventricular ejection fraction (LVEF) between patients with normal exercise capacity and those with subnormal exercise capacity.

Table 3

Parameters of global ventricular function for controls, all patients, normal exercise capacity patients and subnormal exercise capacity patients and by percentage predicted peak oxygen consumption group

 Controls (n = 20) All patients (n = 25) %Pred O2 <85% (n = 14) Predicted VO2 >85% (n = 11) P-value all patients versus controls P-value VO2 subnormal versus normala 
RVEDVi (ml/m274.3 ± 6.2 163.1 ± 63.2 175.3 ± 62.0 147.6 ± 64.1 <0.001 0.28 
RVESVi (ml/m236.5 ± 3.5 90.4 ± 43.9 101.3 ± 46.1 76.6 ± 38.5 <0.001 0.19 
RVEF (%) 50.9 ± 4.0 46.2 ± 7.8 45.9 ± 8.5 47.2 ± 5.9 0.023 0.43 
PRF (%) 0% 29.9 ± 3.4 29.4 ± 3.7 30.5 ± 3.0 – 0.42 
RVSVi (ml/m237.2 ± 4.7 51.3 ± 20.7 52.1 ± 26.9 49.8 ± 20.2 0.34 0.52 
LVEF (%) 60.7 ±5.9 61.0 ± 4.9 60.0 ± 5.1 62.3 ± 4.7 0.87 0.26 
 Controls (n = 20) All patients (n = 25) %Pred O2 <85% (n = 14) Predicted VO2 >85% (n = 11) P-value all patients versus controls P-value VO2 subnormal versus normala 
RVEDVi (ml/m274.3 ± 6.2 163.1 ± 63.2 175.3 ± 62.0 147.6 ± 64.1 <0.001 0.28 
RVESVi (ml/m236.5 ± 3.5 90.4 ± 43.9 101.3 ± 46.1 76.6 ± 38.5 <0.001 0.19 
RVEF (%) 50.9 ± 4.0 46.2 ± 7.8 45.9 ± 8.5 47.2 ± 5.9 0.023 0.43 
PRF (%) 0% 29.9 ± 3.4 29.4 ± 3.7 30.5 ± 3.0 – 0.42 
RVSVi (ml/m237.2 ± 4.7 51.3 ± 20.7 52.1 ± 26.9 49.8 ± 20.2 0.34 0.52 
LVEF (%) 60.7 ±5.9 61.0 ± 4.9 60.0 ± 5.1 62.3 ± 4.7 0.87 0.26 

EDVi: end-diastolic volume indexed to body surface area; ESVi: indexed RV end-systolic volume; LVEF: LV ejection fraction; PRF: pulmonary regurgitation fraction; RVEF: RV ejection fraction; RVSVi: effective right ventricular stroke volume indexed to body surface area.

a

The P-value for comparison of subnormal exercise capacity patients with normal exercise capacity patients.

Assessment of systolic function of the different components of the right ventricle

Parameters of function of the component parts of the right ventricle are presented in Table 4. As expected, the mean EDVi and ESVi of the three RV components were significantly higher in patients than in normal controls. In patients, the apical trabecular portion took up most of the volume load on the right ventricle: with a mean EDVi of 100.4 ± 43.0 ml/m2, the apical trabecular portion accounted for 61.3 ± 20.5% of the total RVEDVi (163.1 ± 63.2 ml/m2). The inlet and outlet EFs were significantly lower in patients than in normal controls. However, the apical trabecular EF was slightly—but significantly—higher in patients than in normal controls, because both end-diastolic and end-systolic volumes of this portion were augmented to similar levels.

Table 4

Functional parameters of the right ventricle component parts for controls, all patients, normal exercise capacity patients and subnormal exercise capacity patients

 Controls (n = 20) All patients (n = 25) %Pred O2 <85% (n = 14) Predicted VO2 >85% (n = 11) P-value (all patients versus controls) P-value VO2 subnormal versus normala 
EDVi (ml/m2
 Inlet 24.9 ± 3.0 28.8 ± 10.7 30.9 ± 9.8 26.1 ± 11.7 0.01 0.27 
 Trabecular 38.6 ± 6.0 100.4 ± 43.0 104.4 ± 37.2 95.2 ± 50.9 <0.001 0.60 
 Outlet 10.8 ± 1.9 33.9 ± 20.9 29.2 ± 5.1 26.4 ± 5.8 <0.001 0.11 
ESVi (ml/m2
 Inlet 9.5 ± 2.9 16.7 ± 8.2 18.6 ± 8.2 14.2 ± 7.9 <0.001 0.12 
 Trabecular 22.0 ± 3.8 54.1 ± 28.7 58.2 ± 28.1 48.8 ± 29.9 <0.001 0.43 
 Outlet 4.9 ± 1.3 19.7 ± 13.1 24.5 ± 15.3 13.6 ± 5.9 <0.001 0.04 
EF (%) 
 Inlet 62.3 ± 11.5 43.8 ±13.9 40.9 ± 14.5 47.4 ± 12.7 <0.001 0.25 
 Trabecular 42.1 ± 8.0 47.9 ± 9.2 46.1 ± 9.8 50.1 ± 8.2 0.03 0.28 
 Outlet 54.8 ± 5.5 41.8 ± 11.9 36.9 ± 12.1 48.1 ± 8.6 <0.001 0.02 
 Controls (n = 20) All patients (n = 25) %Pred O2 <85% (n = 14) Predicted VO2 >85% (n = 11) P-value (all patients versus controls) P-value VO2 subnormal versus normala 
EDVi (ml/m2
 Inlet 24.9 ± 3.0 28.8 ± 10.7 30.9 ± 9.8 26.1 ± 11.7 0.01 0.27 
 Trabecular 38.6 ± 6.0 100.4 ± 43.0 104.4 ± 37.2 95.2 ± 50.9 <0.001 0.60 
 Outlet 10.8 ± 1.9 33.9 ± 20.9 29.2 ± 5.1 26.4 ± 5.8 <0.001 0.11 
ESVi (ml/m2
 Inlet 9.5 ± 2.9 16.7 ± 8.2 18.6 ± 8.2 14.2 ± 7.9 <0.001 0.12 
 Trabecular 22.0 ± 3.8 54.1 ± 28.7 58.2 ± 28.1 48.8 ± 29.9 <0.001 0.43 
 Outlet 4.9 ± 1.3 19.7 ± 13.1 24.5 ± 15.3 13.6 ± 5.9 <0.001 0.04 
EF (%) 
 Inlet 62.3 ± 11.5 43.8 ±13.9 40.9 ± 14.5 47.4 ± 12.7 <0.001 0.25 
 Trabecular 42.1 ± 8.0 47.9 ± 9.2 46.1 ± 9.8 50.1 ± 8.2 0.03 0.28 
 Outlet 54.8 ± 5.5 41.8 ± 11.9 36.9 ± 12.1 48.1 ± 8.6 <0.001 0.02 

EDVi: end-diastolic volume indexed to body surface area; EF: ejection fraction; ESVi: indexed RV end-systolic volume.

a

The P-value for comparison of subnormal exercise capacity patients with normal exercise capacity patients.

Subnormal exercise capacity patients had significantly higher ESVi of the outlet than normal exercise capacity patients. With similar outlet EDVi, outlet EF was significantly higher in normal exercise capacity patients than in subnormal exercise capacity patients. The differences in other right ventricle components parameters between patients with normal and subnormal exercise capacities were not significantly different.

Correlation between clinical and cardiac magnetic resonance parameters and exercise performance

On univariate analysis, the RVOT EF showed the best correlation with percentage of predicted peak VO2. The EF of other components of the right ventricle, including the RV global EF, apical EF and inlet EF showed no significant correlation with percentage of predicted peak VO2 (Table 5). There was no statistically significant correlation between percentage of predicted peak VO2 and age, years after repair, NYHA class, QRS duration and length of ventriculotomy. There was no significant association between percentage of predicted peak VO2 and end-systolic and end-diastolic RV components volumes. There was no relationship between percentage predicted peak VO2 and pulmonary regurgitation. The correlation between exercise capacity and LV function was not significant. On multivariate stepwise linear regression analysis, the RVOT EF was the only significant CMR correlate of percentage of predicted peak VO2 (Table 5).

Table 5

Correlations between clinical and cardiac MR parameters and exercise performance

 %Pred O2, %
 
Univariate analysis
 
Multivariable linear regression
 
r P-value ß P-value 
Age 0.123 0.558   
Years since repair 0.204 0.328   
NYHA −0.135 0.520   
QRS duration 0.270 0.193   
PRF 0.179 0.393   
Length of ventriculotomy 0.149 0.051 −0.562 0.142 
RVEDVi −0.254 0.221   
Inlet EDVi −0.361 0.076   
Apical EDVi −0.160 0.446   
Outflow EDVi −0.255 0.219   
RVESVi −0.315 0.125   
Inlet ESVi −0.370 0.068   
Apical ESVi −0.201 0.336   
Outflow ESVi −0.384 0.058   
LVEF −0.161 0.443   
RVEF 0.394 0.052 0.479 0.852 
Inlet EF 0.230 0.269 0.871 0.775 
Apical EF 0.209 0.317 0.542 0.457 
Outflow EF 0.590 0.002 0.442 0.024 
 %Pred O2, %
 
Univariate analysis
 
Multivariable linear regression
 
r P-value ß P-value 
Age 0.123 0.558   
Years since repair 0.204 0.328   
NYHA −0.135 0.520   
QRS duration 0.270 0.193   
PRF 0.179 0.393   
Length of ventriculotomy 0.149 0.051 −0.562 0.142 
RVEDVi −0.254 0.221   
Inlet EDVi −0.361 0.076   
Apical EDVi −0.160 0.446   
Outflow EDVi −0.255 0.219   
RVESVi −0.315 0.125   
Inlet ESVi −0.370 0.068   
Apical ESVi −0.201 0.336   
Outflow ESVi −0.384 0.058   
LVEF −0.161 0.443   
RVEF 0.394 0.052 0.479 0.852 
Inlet EF 0.230 0.269 0.871 0.775 
Apical EF 0.209 0.317 0.542 0.457 
Outflow EF 0.590 0.002 0.442 0.024 

EDVi: end-diastolic volume indexed to body surface area; ESVi: indexed RV end-systolic volume; LVEF: LV ejection fraction; PRF: pulmonary regurgitation fraction; RVEF: RV ejection fraction.

DISCUSSION

In repaired TOF patients with severe RV dilatation and pulmonary regurgitation, peak oxygen consumption on exercise testing was more strongly correlated with RVOT EF than with systolic function of other RV components or global RV systolic function. By multivariate analysis, RVOT systolic function was shown to be the only independent predictor of exercise capacity in repaired TOF patients.

The benefits of early pulmonary valve replacement should, however, be weighed against the limited durability of pulmonary valve conduits [14]. Replacing the pulmonary valve too early might expose patients to an increased risk for reoperations and interventions, whereas deferring this surgery for as long as possible may reduce the number of surgical procedures needed over a patient’s lifetime [15]. Much emphasis has been placed on the use of RV volumes in the decision-making process and, currently, an RVEDVi of 150 ml/m2 is considered the threshold for recommending pulmonary valve replacement. However, like others [16], we too failed to demonstrate any relationship between RVEDVi and exercise capacity. Furthermore, nearly half (11/25) of the patients in our study had preserved normal peak VO2 despite having severe RV dilatation, suggesting that there was good right ventricle. Dilatation is the initial pathophysiological RV response to severe pulmonary regurgitation. ‘Isolated’ RV volume overload can be tolerated without any symptoms over many years and even decades, and then be followed by RV dysfunction and failure of compensatory mechanisms [17]. Hence, the degree of RV dilatation alone might not truly reflect the RV compensation in response to volume overload.

Previous research has not established any definite relationship between RVEF and peak exercise capacity. Meadows et al. [16] found that RVEF was positively and independently correlated with VO2 peak, VO2 at anaerobic threshold, and oxygen pulse. However, other studies have reported weak or no correlation between RVEF and peak exercise capacity [18–20]. Our results also failed to show any correlation between RVEF and exercise capacity in the setting of severe RV dilatation and pulmonary regurgitation. A severely dilated right ventricle might maintain its output, albeit with mildly or moderately decreased RVEF. The functional parameters of the global right ventricle do not necessarily reflect the true picture of the complex adaptive response of the right ventricle to chronic overload.

Functional analysis of the different components of the right ventricle may provide a better picture of the actual RV compensation. We found that the apical trabecular EF was higher than the EF of the other two components in repaired TOF patients; however, peak oxygen consumption was more strongly correlated with RVOT EF than with the EF of the other components of the right ventricle. On multivariate analysis, RVOT systolic function was the only independent predictor of exercise capacity. By functional quantification of the sinus and outflow part of the right ventricle, Bove et al. [20] recently found that the systolic function of the RV sinus component was a better predictor of exercise performance. However, we think analysis in terms of only the ‘sinus’ and the ‘conus’ does not give the true picture because the inlet component is not taken into consideration. It is important to consider tricuspid valve regurgitation when analysing RV components function. Severe tricuspid valve regurgitation has complex effects on the right ventricle [21], and RV components function might be very different in these patients as compared to that in patients with pulmonary regurgitation alone. Although tricuspid valve regurgitation is a common associated lesion after TOF repair [22], the prevalence was low in our cohort, with mild tricuspid regurgitation seen in only 24% of patients. The reason might be that a transventricular approach is used in our institute to close the VSD in all patients. It has been reported that tricuspid valve regurgitation can directly result from TOF repair, and that the transatrial–transventricular approach places the tricuspid valve at risk for direct damage [23].

Our results highlight the importance of RVOT function in late repaired TOF patients. In these patients, the main determinant of cardiac output, and thereby exercise capacity, is the LV preload [24]. In patients with significant pulmonary regurgitation, LV preload may best be assessed by the effective RV output, since LV preload is determined by the actual net RV forward flow, taking into account the degree of pulmonary regurgitation. Our results failed to show any difference of effective RV stroke volume between the normal and the subnormal exercise capacity groups; however, we think resting MRI may not truly demonstrate the cardiac compensatory mechanisms during exercise. A recent real-time MRI study showed that one of the compensatory mechanisms whereby adult repaired TOF patients with severely dilated RVs augment forward flow is by decreasing the pulmonary regurgitation fraction [25]. The RVOT may play an important role in such compensation. The functional RVOT remains contracted/constricted like a sphincter until late in the diastolic phase, and this is probably essential for reducing pulmonary regurgitation, augmenting forward flow out of the enlarged ventricle, and thus, maintaining normal exercise capacity [26].

The optimal timing of pulmonary valve replacement in TOF patients is a point of contention, particularly in asymptomatic patients. We too could not come to any conclusion on this topic on the basis of the present pilot cross-sectional study; However, our results showed that decisions regarding valve reimplantation should not be based solely on the global CMR parameters such as RVEDVi and RVEF, as has been previously suggested [2], because these factors have poor correlation to exercise capacity in patients with severely dilated right ventricles and significant pulmonary regurgitation We found positive and independent correlation between RVOT function and exercise capacity, which highlights the important contribution of the RVOT towards RV output. Thus, RVOT systolic function can provide important additional information to aid decision making regarding timing of pulmonary valve replacement. The potential role of detailed RV CMR analysis in guiding decisions on optimal surgical timing will be further explored.

More and more centres have begun to monitor exercise capacity during follow-up of patients with repaired TOF. Numerous studies have consistently shown decrease in RV volume and/or increase in RVEF despite relatively unchanged exercise capacities. To date, no CMR parameters have been shown to be reliable predictors of exercise capacity after valve replacement. The systolic function of the RVOT could be a reliable determinant of intrinsic RV performance in repaired TOF patients; therefore, RVOT function could be a promising parameter for deciding timing of pulmonary valve replacement so as to achieve the best possible exercise capacity in repaired TOF patients.

Our findings also support the notion that the surgical strategy in primary repair has important effects on exercise capacity of TOF patients in the long term. Although length of the right ventricle incision was not independently correlated to exercise capacity on multivariate analysis, incision length was significantly longer in patients with subnormal exercise capacity. Ventriculotomy has been shown to reduce regional RV function around the incision site and to impair RVOT function after TOF repair [27]; therefore, ventriculotomy should be minimal and be performed only if there is likely to be benefits such as improved exercise capacity over the long term.

The main limitation of the present study is the small sample size. Although our study shows conclusively that the function of outflow part of the RV is a reliable indicator of overall RV function, especially in repaired TOF patients with severely dilated right ventricles, we acknowledge that our results need to be confirmed in larger populations.

CONCLUSION

This study demonstrated that both RVEDVi and RVEF are not correlated to exercise capacity in the setting of severe RV dilatation and pulmonary regurgitation. Through quantitative analyses of the function of the different anatomical RV components, the systolic function of the RVOT was shown to best predict exercise capacity in individual patients, and thus it appears to be the most reliable determinant of RV performance.

Conflict of interest: none declared.

REFERENCES

1
Weinberg
CR
McElhinney
DB.
Pulmonary valve replacement in tetralogy of Fallot
.
Circulation
 
2014
;
130
:
795
8
.
2
Geva
T.
Indications for pulmonary valve replacement in repaired tetralogy of Fallot: the quest continues
.
Circulation
 
2013
;
128
:
1855
7
.
3
Therrien
J
Provost
Y
Merchant
N
Williams
W
Colman
J
Webb
G.
Optimal timing for pulmonary valve replacement in adults after tetralogy of Fallot repair
.
Am J Cardiol
 
2005
;
95
:
779
82
.
4
Geva
T.
Repaired tetralogy of Fallot: the roles of cardiovascular magnetic resonance in evaluating pathophysiology and for pulmonary valve replacement decision support
.
J Cardiov Magn Reson
 
2011
;
13
:
9
.
5
Giardini
A
Specchia
S
Tacy
TA
Coutsoumbas
G
Gargiulo
G
Donti
A
et al.  .
Usefulness of cardiopulmonary exercise to predict long-term prognosis in adults with repaired tetralogy of Fallot
.
Am J Cardiol
 
2007
;
99
:
1462
7
.
6
Muller
J
Hager
A
Diller
GP
Derrick
G
Buys
R
Dubowy
KO
et al.  .
Peak oxygen uptake, ventilatory efficiency and QRS-duration predict event free survival in patients late after surgical repair of tetralogy of Fallot
.
Int J Cardiol
 
2015
;
196
:
158
64
.
7
Buys
R
Van De Bruaene
A
De Meester
P
Budts
W
Vanhees
L.
Predictors of mid-term event-free survival in adults with corrected tetralogy of Fallot
.
Acta Cardiol
 
2012
;
67
:
415
21
.
8
Arena
R
Sietsema
KE.
Cardiopulmonary exercise testing in the clinical evaluation of patients with heart and lung disease
.
Circulation
 
2011
;
123
:
668
80
.
9
Anderson
RH
Baker
EJ
Redington
AN.
Can we describe structure as well as function when accounting for the arrangement of the ventricular mass?
Cardiol Young
 
2000
;
10
:
247
60
.
10
Sheehan
F
Redington
A.
The right ventricle: anatomy, physiology and clinical imaging
.
Heart
 
2008
;
94
:
1510
5
.
11
Bodhey
NK
Beerbaum
P
Sarikouch
S
Kropf
S
Lange
P
Berger
F
et al.  .
Functional analysis of the components of the right ventricle in the setting of tetralogy of Fallot
.
Circ Cardiovasc Imaging
 
2008
;
1
:
141
7
.
12
van der Hulst
AE
Roest
AA
Holman
ER
de Roos
A
Blom
NA
Bax
JJ
et al.  .
Real-time three-dimensional echocardiography: segmental analysis of the right ventricle in patients with repaired tetralogy of Fallot
.
J Am Soc Echocardiogr
 
2011
;
24
:
1183
90
.
13
Fratz
S
Chung
T
Greil
GF
Samyn
MM
Taylor
AM
Valsangiacomo Buechel
ER
et al.  .
Guidelines and protocols for cardiovascular magnetic resonance in children and adults with congenital heart disease: SCMR expert consensus group on congenital heart disease
.
J Cardiovasc Magn Resonan
 
2013
;
15
:
51
.
14
Villafane
J
Feinstein
JA
Jenkins
KJ
Vincent
RN
Walsh
EP
Dubin
AM
et al.  .
Hot topics in tetralogy of Fallot
.
J Am Coll Cardiol
 
2013
;
62
:
2155
66
.
15
Chen
PC
Sager
MS
Zurakowski
D
Pigula
FA
Baird
CW
Mayer
JE
Jr
et al.  .
Younger age and valve oversizing are predictors of structural valve deterioration after pulmonary valve replacement in patients with tetralogy of Fallot
.
J Thorac Cardiovasc Surg
 
2012
;
143
:
352
60
.
16
Meadows
J
Powell
AJ
Geva
T
Dorfman
A
Gauvreau
K
Rhodes
J.
Cardiac magnetic resonance imaging correlates of exercise capacity in patients with surgically repaired tetralogy of Fallot
.
Am J Cardiol
 
2007
;
100
:
1446
50
.
17
Katz
NM
Blackstone
EH
Kirklin
JW
Pacifico
AD
Bargeron
LM
Jr.
Late survival and symptoms after repair of tetralogy of Fallot
.
Circulation
 
1982
;
65
:
403
10
.
18
Wald
RM
Haber
I
Wald
R
Valente
AM
Powell
AJ
Geva
T.
Effects of regional dysfunction and late gadolinium enhancement on global right ventricular function and exercise capacity in patients with repaired tetralogy of Fallot
.
Circulation
 
2009
;
119
:
1370
7
.
19
Alghamdi
MH
Mertens
L
Lee
W
Yoo
SJ
Grosse-Wortmann
L.
Longitudinal right ventricular function is a better predictor of right ventricular contribution to exercise performance than global or outflow tract ejection fraction in tetralogy of Fallot: a combined echocardiography and magnetic resonance study
.
Eur Heart J Cardiovasc Imaging
 
2013
;
14
:
235
9
.
20
Bove
T
Vandekerckhove
K
Devos
D
Panzer
J
De Groote
K
De Wilde
H
et al.  .
Functional analysis of the anatomical right ventricular components: should assessment of right ventricular function after repair of tetralogy of Fallot be refined?
Eur J Cardiothorac Surg
 
2014
;
45
:
e6
12
.
21
Calafiore
AM
Bartoloni
G
Al Amri
H
Iacò
AL
Abukhudair
W
Lanzaro
BI
et al.  .
Functional tricuspid regurgitation and the right ventricle: what we do not know is more than we know
.
Expert Rev Cardiovasc Ther
 
2012
;
10
:
1351
66
.
22
Cramer
JW
Ginde
S
Hill
GD
Cohen
SB
Bartz
PJ
Tweddell
JS
et al.  .
Tricuspid repair at pulmonary valve replacement does not alter outcomes in tetralogy of Fallot
.
Ann Thorac Surg
 
2015
;
99
:
899
904
.
23
Mahle
WT
Parks
WJ
Fyfe
DA
Sallee
D.
Tricuspid regurgitation in patients with repaired tetralogy of Fallot and its relation to right ventricular dilatation
.
Am J Cardiol
 
2003
;
92
:
643
45
.
24
Yap
J
Tan
JL
Le
TT
Gao
F
Zhong
L
Liew
R
et al.  .
Assessment of left ventricular preload by cardiac magnetic resonance imaging predicts exercise capacity in adult operated tetralogy of Fallot: a retrospective study
.
BMC Cardiovasc Disord
 
2014
;
14
:
122
.
25
O’Meagher
S
Munoz
PA
Muthurangu
V
Robinson
PJ
Malitz
N
Tanous
DJ
et al.  .
Mechanisms of maintained exercise capacity in adults with repaired tetralogy of Fallot
.
Int J Cardiol
 
2014
;
177
:
178
81
.
26
Brock
R.
Control mechanisms in the outflow tract of the right ventricle in health and disease
.
Guy's Hospital Rep
 
1955
;
104
:
356
79
.
27
Takabayashi
S
Shimpo
H
Yokoyama
K
Onoda
K
Mitani
Y.
Reduced regional right ventricular wall motion after transventricular repair of tetralogy of Fallot
.
J Thorac Cardiovasc Surg
 
2007
;
133
:
1656
8
.

Comments

0 Comments