Article Text

Download PDFPDF

Exercise capacity and stroke volume are preserved late after tetralogy repair, despite severe right ventricular dilatation
  1. Shamus O'Meagher1,2,
  2. Phillip A Munoz1,3,
  3. Jennifer A Alison3,4,
  4. Iven H Young3,
  5. David J Tanous2,5,
  6. David S Celermajer1,2,
  7. Rajesh Puranik1,2
  1. 1Faculty of Medicine, The University of Sydney, Sydney, Australia
  2. 2Department of Cardiology, Royal Prince Alfred Hospital, Sydney, Australia
  3. 3Department of Respiratory and Sleep Medicine, Royal Prince Alfred Hospital, Sydney, Australia
  4. 4The University of Sydney, Discipline of Physiotherapy, Sydney, Australia
  5. 5Department of Cardiology, Westmead Hospital, Sydney, Australia
  1. Correspondence to Dr Rajesh Puranik, Department of Cardiology, Royal Prince Alfred Hospital, Missenden Road, Camperdown, NSW 2050, Australia; raj.puranik{at}


Objectives To assess if exercise capacity and resting stroke volume are different in tetralogy of Fallot (TOF) repair survivors with indexed RV (right ventricle) end-diastolic volume (RVEDVi) more versus less than 150 ml/m2, a currently suggested threshold for pulmonary valve replacement (PVR).

Design Cross-sectional study.

Setting Single-centre adult congenital heart disease unit.

Patients 55 consecutively eligible patients with repaired TOF (age at repair 2.3±1.9 years; age at evaluation 26.2±8.8 years; NYHA Class I or II).

Interventions Cardiovascular MRI (1.5T) and cardiopulmonary exercise test.

Main outcome measures Biventricular volumes and function; exercise capacity.

Results 20 patients had RVEDVi below, and 35 had RVEDVi above 150 ml/m2, at time of referral. In the >150 ml/m2 group, fractional pulmonary regurgitation was higher (41±8 vs 31±8%, p<0.001). Although RV ejection fraction (EF) was lower (47±7 vs 54±6%, p=0.007), indexed RV stroke volume was higher (87±14 vs 64±10 ml/m2, p<0.001) in the >150 ml/m2 group. There were no significant differences in LVEF, indexed LV stroke volume or exercise capacity (% predicted peak work: 90±17 vs 89±11% and; % predicted VO2 peak: 84±17 vs 87±12%).

Conclusions Exercise capacity and stroke volume are maintained with RVEDVi above compared with below a commonly used cut-off for PVR surgery. Optimal timing for PVR, thus, remains unclear.

  • Tetralogy of Fallot
  • right ventricular dilatation
  • exercise capacity
  • pulmonary valve replacement
  • congenital heart disease

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.


Pulmonary regurgitation (PR) is very common late after the repair of tetralogy of Fallot (TOF) in childhood due to the frequent need to open the right ventricular outflow tract across the pulmonary valve as part of the repair.1 Chronic exposure of the right ventricle (RV) to a regurgitant load may lead to progressive RV dilatation, and has been associated with adverse events such as increased symptoms, arrhythmia and death.2–6 Potential causes of RV dysfunction in adults with repaired TOF include abnormalities of the temporal pattern of RV mechanical activation.7 Although pulmonary valve replacement (PVR) is an effective way of reducing the volume load on the RV, there is no proof from prospective studies that it reduces the risk for adverse outcomes, such as sudden cardiac death, sustained ventricular arrhythmias or heart failure.8–13 The timing of PVR, however, is an area of uncertainty. The European Society of Cardiology (ESC), the American College of Cardiology (ACC)/American Heart Association (AHA) and the Canadian Cardiovascular Society (CCS) have currently published guidelines for PVR in adults with repaired TOF.14–16 Severe PR, together with symptoms and/or moderate to severe RV enlargement, are suggested as indicators for PVR. However, there is some uncertainty regarding the definition of moderate to severe RV enlargement. Based on work conducted by Therrien et al (2005), the CCS guidelines suggest an indexed RV end-diastolic volume (RVEDVi) of 170 ml/m2 as an indicator for PVR, as dilatation beyond this is unlikely to result in normalisation of RV dimensions post-PVR.11 Alternatively, the ESC includes progressive RV dilatation in indications for intervention after repair of TOF, and refers to Oosterhof et al (2007) when stating that RV size is unlikely to normalise after PVR when RVEDVi is allowed to dilate beyond 160 ml/m2.10 The ACC/AHA guidelines, however, do not indicate what constitutes moderate to severe RV enlargement. In practice, an RVEDVi of 150ml/m2 is a commonly used RV volume for recommending PVR in adults with repaired TOF.17 As PVR has limited durability,18 deferring this surgery for as long as possible may reduce the number of surgical procedures needed over a patient's lifetime.

In this study, we sought to determine differences in cardiac structure, function and exercise capacity in such patients, above and below an RVEDVi of 150 ml/m2.


Patient population

Between April 2009 and February 2010, a total of 55 consecutive patients with previously diagnosed TOF or pulmonary atresia with ventricular septal defect (PAVSD) were prospectively recruited after being referred to our adult congenital heart service for a cardiac MRI and cardiopulmonary exercise test (CPET) (table 1). Ethics approval was obtained from the Royal Prince Alfred Hospital Human Research Ethics Committee, Sydney, and all participants gave informed consent.

Table 1

Demographic and surgical data for all patients and by RVEDVi group

Cardiovascular MRI protocol:

MRI was performed using 1.5T MR scanner (GE medical system).

Assessment of ventricular volumes and function using cine MRI

Retrospectively gated steady-state free precession (FIESTA) cine MR images of the heart were acquired in the vertical long-axis, 4-chamber view and the short-axis view covering the entirety of both ventricles (9–12 slices). Image parameters: TR=3.2 ms; TE=1.6 ms; flip angle=78°; slice thickness=8 mm; matrix=192×256; field of view=300–380 mm; and temporal resolution=40 ms, acquired during a single breath-hold. Assessment of RV and LV volumes were performed by manual segmentation of short-axis cine images with endocardial outline at end diastole and end systole (performed in Osirix software, version 3.6.1 32bit). End-diastolic and end-systolic volumes were calculated by use of Simpson's rule for each ventricle, and from these volumes the ejection fraction (EF) was calculated.

MR flow quantification

Pulmonary artery (PA) and aortic flow data were acquired by use of a flow-sensitive gradient-echo sequence (TR, <5 ms; TE, <3 ms; flip angle, 15°; slice thickness, 7 mm; field of view=300–380 mm matrix, 256×240, temporal resolution=30 ms) acquired during a breath-hold. Image planes were located at the midpoint of the main PA and ascending aorta (sino-tubular junction). Through-plane flow data was acquired by use of retrospective cardiac gating. Arterial blood flow was calculated from phase contrast images by use of a semiautomatic vessel edge-detection algorithm (Reportcard, GE) with manual operator correction. Net forward flow within the main PA (ml) and PR fraction (%) were calculated as total PA flow minus backward PA flow, and per cent backward PA flow over total PA flow, respectively. These calculations were similarly made for the aortic flow.

Late gadolinium enhancement: scar/fibrosis imaging

Scar imaging within the myocardium was performed with segmented phase-sensitive inversion recovery sequences (Image parameters: TR=2×RR interval; TE=3.4 ms; flip angle=25°; slice thickness=10 mm; matrix=144×256; field of view=300–380 mm, acquired during a single breath-hold) 10 min post-administration of intravenous contrast (0.2 mmol/kg of gadolinium pentatate, Magnevist). Imaging included the entire short-axis and long-axis planes of the LV and RV.

Cardiopulmonary exercise testing

Subjects performed progressive exercise tests on an electronically braked bicycle ergometer (Lode Corival; Lode BV, Groningen, The Netherlands) to maximal volition. A ramped protocol was adjusted for each patient, as described by Jones et al (1985).19 Following maximal exertion, subjects were monitored while cycling at a low resistance and cadence for a period of 2 min, or until it was felt that the subject had adequately recovered. Breath by breath expiratory gas analysis (VMax229; SensorMedics; Yorba Linda, California) and ECG monitoring (Cardiosoft, version 6.51, GE Healthcare, Waukesha, Wisconsin) were performed. Oxygen saturation (Radical, Massimo Corp, Irvine, USA) and periodic manual blood pressure measurements were also obtained. Exercise variables monitored were oxygen consumption, carbon dioxide production, ventilation, work rate, heart rate, blood pressure, oxygen saturation and ECG morphology.

Statistical methods

All data are presented as mean±SD or median and range. The patients were divided into two groups according to a prospectively defined cut-point for RVEDVi, at 150 ml/m2. Secondary analyses were performed using RVEDVi cut-points of 160 ml/m2 and 170 ml/m2. Statistical comparison of parametric data was performed with a 2-tailed unpaired Student t test. Prespecified primary endpoints of interest were (1) for MRI, the calculated LV cardiac output (as the best overall determinant of cardiac performance) and (2) for CPET, the maximum work achieved as a per cent of normal (control) values. All other non-primary comparisons were adjusted by the Hochberg modification of the Bonferroni correction.20 The relationship between dichotomous variables was tested with the χ2 test. A two-tailed p value <0.05 was considered statistically significant. Pearson's correlation coefficient was used to assess relations between cardiac MRI and exercise parameters. Statistical analysis was performed with SPSS V.19 for Windows (SPSS).


Patient population (n=55)

Fifty-five patients (mean age at repair 2.3±1.9 years; age at evaluation 26.2±8.8 years; range 15–49 years; 33 males) were included in this study. Demographic and surgical details are shown in table 1. The primary diagnosis in the cohort was most frequently TOF (n=50; 91%) with a lesser number of PA-VSD (n=5; 9%). A transannular patch (TAP) repair was performed in 46 (84%) patients, while a homograft, valvectomy or valve-sparing repair was conducted in 2 (3.5%), 2 (3.5%) and 2 (3.5%) patients, respectively. Detailed operative data was not available for three patients due to missing medical records. Overall, the patients were assessed 23.6±7.2 years after surgical repair.

Subjects with RVEDVi of >150 ml/m2 (n=35) were later after initial repair (25.4±7.8 vs 20.7±5.8 years; p=0.009) and were older at the time of evaluation (28.4±9.7 vs 22.5±5.5 years; p=0.006). However, there was not a significant difference in primary diagnosis (>150 ml/m2; TOF n=33, PA VSD n=2 vs <150 ml/m2; TOF n=17, PA VSD n=3; p=0.249), age at complete repair (2.5±2.1 years vs 2.0±1.5 years, p=0.344), type of primary repair (TAP n=31, homograft n=0, valvectomy n=1, valve sparing repair n=1 vs TAP n=15, homograft n=2, valvectomy n=1, valve sparing repair n=1; p=0.36), gender (p=0.09).

MRI results (n=55)

For all subjects considered together, RV dilatation (RVEDVi 163±44 ml/m2; RVESVi 84±37 ml/m2) and moderate to severe PR (PR fraction 38±10%; PR volume 53±23 ml/beat) were observed (table 2). Indexed LVEDVi was within the normal range (83±15 ml/m2), however, indexed LVESVi was elevated (36±11 ml/m2). Consistent with the RV dilatation, QRS duration was prolonged (143±29 ms) (table 1), although the effect of LV parameters cannot be excluded. Biventricular indexed stroke volumes (SV) and EFs were maintained (RVSVi 79±17 ml/m2; LVSVi 47±8 ml/m2; RVEF 50±7%; LVEF 58±7%). Fibrosis was rarely detected in the RV body (n=1), and more frequently in the RV outflow (n=38).

Table 2

MRI data for all patients and by RVEDVi group

Cardiopulmonary exercise test results (n=55)

Table 3 shows cardiopulmonary exercise test results. Peak work rate reached 176±51 Watts, which represented 89±15% of that predicted. Peak oxygen consumption reached 85±15% of that predicted at 31±8 ml/kg/min. Peak heart rate achieved was 173±16 beats/min which was 92±8% of that predicted. Oxygen consumption at anaerobic threshold was 21±6 ml/kg/min which represents 68±12% of peak oxygen uptake. Ventilatory response to carbon dioxide (VE/VCO2) was 25±3 at anaerobic threshold and 28±4 at peak exercise. Percentage of predicted peak work achieved significantly correlated with LVEDVi (p=0.048). There were no other significant correlations between the variables shown in table 4.

Table 3

Cardiopulmonary exercise test data for all patients and by RVEDVi group

Table 4

Cardiac MRI predictors of exercise capacity; univariate analyses

RVEDVi of 160 ml/m2 and 170 ml/m2 as cut-points

Secondary analyses were performed using RVEDVi cut-points of 160 ml/m2 and 170 ml/m2. Comparing patients above and below an RVEDVi of 160 ml/m2, LVEDVi (p=0.003) and LVSVi (p=0.008) became significantly higher in the more dilated group. All LV parameters became insignificant between groups when an RVEDVi of 170 ml/m2 was used as a cut-point. There were no significances between group differences in any exercise parameter, when either of these secondary cut-points was used.


In this study, we have shown relatively normal submaximal and peak exercise parameters in a group of adult patients with repaired TOF, prospectively recruited from a single centre, with pulmonary regurgitation and right ventricular dilatation. When the group was divided into those above and below an RVEDVi of 150 ml/m2, there was no significant difference in measured values of maximal exercise capacity. In fact, exercise capacity was at near-normal levels, even in the more dilated group, despite greater PR, RV dilatation and lower normal RVEF. This leaves the crucial decision of PVR timing less certain, in adult patients with repaired TOF.

Over the entire cohort, measures of peak exercise capacity were higher than previously reported in similar groups of adult repaired TOF patients (VO2peak: 21.1–27.6 ml/kg/min; per cent achieved of predicted VO2 peak: 51–78%; peak work: 109–143 Watts; peak heart rate: 146–163 beats/min).9 ,21–24 Measures of ventricular function, age at evaluation, or age at TOF repair surgery do not explain the disparity in exercise capacity, as our results do not meaningfully differ from those of past research with respect to these parameters.9 ,21–24 Relatively preserved heart rate responses, and/or an era effect may contribute to these favourable results.

Chronotropic incompetence contributes to exercise intolerance and is related to long-term outcomes in repaired TOF.25 Our cohort exhibited a substantially increased heart rate at peak exercise compared with previous similar studies,9 ,21–24 and offers in part an explanation for the preserved exercise capacity we observed. It should be noted that none of the patients in our study had evidence of resting or exercise-induced arrhythmia. There is also the possibility of an era effect, with our data being collected on a young group of patients in a later time period (2008–2011), compared with previous studies.9 ,21–24 In this regard, surgical technique, postoperative care and clinical management of patients may have differed from the previously published groups of patients.

Since adults with repaired TOF patients have previously been shown to respond positively to a structured program of physical activity,26 higher levels of physical activity in our patient group may also be a plausible contributor to the high exercise capacity we observed. Previous studies have shown physical activity levels to be low in adult repaired TOF patients, where this data was generated from the northern hemisphere.26 It is possible that our patients (in Australia) were encouraged to participate in physical activity from a young age, and this may have had an impact in preserving and optimising their exercise capacity, however, such data were not routinely collected. More investigation into the opportunities provided, and the impact of shifting attitudes regarding physical activity in this group of patients certainly deserves additional detailed study.

Our study demonstrates a lack of correlation between baseline RV volumes and peak exercise capacity in PR. This result is in keeping with previous studies which have similarly failed to show a relationship between RV dilatation and exercise capacity.24 Moreover, in the context of a dilated but not concentrically hypertrophied RV as a result of PR, LV systolic function is not impaired by ventricular interaction, as may be the case of RV hypertrophy resulting from pulmonary stenosis.27 Since oxygen saturations and LV function are not compromised as the RV dilates, oxygen delivery to working muscles and exercise capacity would not be expected to decrease (unless the RV stroke volume fails). Our findings thus question the presumption that RV dilatation inevitably impairs exercise performance and forms an important piece of evidence as to why repaired TOF patients may not improve their exercise capacity after PVR. Although limited conclusions can be drawn from cross-sectional data, it is worth noting that the more dilated group was significantly older than the less dilated group. There is currently no satisfactory data regarding the temporal patterns of RV dilatation. Our findings suggest that long-term monitoring of RV structure is warranted which would be best addressed by a large prospective study.

Late after TOF repair, PR is known to be associated with RV dilatation and systolic dysfunction,28 which can result in heart failure, exercise intolerance, arrhythmia and death.2–6 The optimal timing of PVR in TOF patients is a point of contention, particularly in asymptomatic patients. The ESC, ACC/AHA and CCS have published guidelines for PVR late after TOF repair.14–16 There is general agreement that the decision for PVR involves consideration of the presence or absence of symptoms, the degree of PR, RV function, exercise capacity, arrhythmia and the degree of RV enlargement. There is, however, some disagreement about the volume to which the RV can be allowed to dilate prior to PVR, whilst still achieving beneficial remodelling post surgery. Threshold RVEDVi measurements of between 150 ml/m2 and 170 ml/m2 have been suggested as points above which normalisation of RV dimension becomes unlikely.8 ,10 ,11 ,15–17 ,29 In order to minimise the number of lifetime procedures in these subjects with PR, developing more specific markers for detecting adverse adaptive mechanisms to PR would be useful in assisting decision making concerning the timing of surgery.


The observed similarities in RV systolic function between the two groups in this study may be explained to some extent, by resting cardiac MRI measurements being compared with exercise variables measured at a point of maximal exertion. Future studies may need to focus on assessment of cardiac function during exercise to elucidate parameters more relevant to the maximally functioning heart. Furthermore, longitudinal data relevant to the chronically dilating RV may be important in understanding the RV-adaptive mechanisms of PR. Also, we have not directly examined two other factors that might influence the timing of PVR for dilated RV; propensity to arrhythmia or ability of the RV to diminish in size postoperatively. These other factors require consideration in clinical decision making.


We found peak exercise capacity to be higher than previously reported, and near-normal in our cohort of adults with dilated RVs late after TOF repair. Contributors may include appropriate peak heart rate, good physical activity levels or an era effect. Furthermore, our patients exhibited near-normal exercise capacity and maintained biventricular systolic function, even in those with RVs dilated well beyond 150 ml/m2. These results question the validity of using this marker of RV dilatation as the major basis for replacing the pulmonary valve in this setting.



  • Funding Dr Rajesh Puranik is a Medical Foundation Fellow, University of Sydney, Australia. Funding for this project was derived from his fellowship.

  • Competing interests None.

  • Patient consent Obtained.

  • Ethics approval Royal Prince Alfred Hospital Human Research Ethics Committee.

  • Provenance and peer review Not commissioned; externally peer reviewed.