• cardiac magnetic resonance (cmr) imaging
• tetralogy of fallot
• pulmonic valve disease

The success story of patients born with tetralogy of Fallot (ToF) is closely linked to significant progress in the treatment of congenital heart disease that resulted in low early mortality and a favourable course during the first two decades of life. Consequently, the majority of patients with repaired ToF survive to adulthood and participate in regular work and social life. The first outcome studies of larger cohorts of patients, however, demonstrated a significant risk of sudden death and malignant arrhythmia in the long-term follow-up.1 As a consequence of surgical relief of right ventricular (RV) outflow tract obstruction, pulmonary regurgitation (PR) resulted as the predominant, residual haemodynamic lesion causing progressive RV enlargement, functional impairment and probably sudden cardiac death. Therefore, pulmonary valve replacement (PVR) was considered necessary in order to abolish valvular insufficiency and to allow reverse remodelling of the dilated RV to normal values. Unfortunately, as RV volumes and function did not recover in all patients and adverse events occurred also frequently following PVR, extensive discussion evolved regarding the optimal time point for PVR taking into account the risk of irreversible RV remodelling versus degeneration of the implanted valve prosthesis over time.

Cardiac MR (CMR) imaging has contributed substantially to the understanding of the pathophysiological consequences of PR due to its ability for reliable quantification of PR and volumetric RV assessment in this population. Consequently, current recommendations for PVR are largely based on CMR studies investigating the effect of PVR on RV properties.

In order to improve management of this patient population, the International Multicenter TOF Registry (INDICATOR) study group aims to collect CMR data in a multicentre registry allowing more appropriate risk stratification. In their Heart manuscript, Bokma and colleagues (on behalf of the INDICATOR group) retrospectively assessed outcomes in a large cohort of patients with ToF either with or without PVR during follow-up who were enrolled in the registry2. After adjustment of baseline differences between the two groups using a propensity-score-adjusted analysis, the study revealed no significant differences in primary (death and sustained ventricular tachycardia) and secondary (heart failure, atrial arrhythmia and non-sustained ventricular tachycardia) outcome variables between patients with and patients without PVR after a mean follow-up of 5.3 years. Furthermore, a significantly higher event rate for secondary outcome variables was observed in the PVR group when consensus criteria for PVR were not met.

At a first look, the reported results clearly do not meet the expectations patients and physicians attribute to PVR. Do these findings suggest the end of the success story in Fallot patients? For now, this question remains unanswered in light of the limitations of a retrospective, observational registry study. As mentioned by the authors, the follow-up time may be too short to see a significant impact on outcome and the overall low event rate (4% for primary outcome) reduced statistical power. Furthermore, clinically relevant data such as exercise performance and echocardiographic indices (ie, tricuspid valve regurgitation and residual RV pressure load) were not included in the analysis. Accordingly, the benefit of PVR on patients’ symptoms and exercise capacity could not be assessed.

The reported ‘mid-term’ outcomes shed a critical light on several aspects of our current theory regarding the time course and underlying mechanisms of RV failure, arrhythmias and sudden cardiac death. If PVR seems not to improve outcome or even increases risk when performed ‘too early’, our hypothesis of an association between RV regression and improved outcome following PVR seems error prone. Although both the severity of PR and RV dilatation have been identified as risk factors in earlier studies, the first INDICATOR publication revealed, among other factors, higher RV pressures and RV hypertrophy as independent predictors for adverse outcomes.3 Furthermore, longitudinal CMR data of patients with ToF showed relatively stable RV volumes and function over time with a low percentage of patients showing real ‘progression’ of chamber dimensions and ventricular dysfunction.4 These findings suggest that the proposed CMR thresholds for PVR may not be appropriate for deciding the optimal time point for RV outflow tract interventions. Interestingly, the ‘proactive’ PVR threshold applied in the study by Bokma et al was set at an end-diastolic volume greater than 160 mL/m², a value above which many centres would certainly refer patients for PVR. As data on RV remodelling and clinical status post-PVR are not reported in the current study, the relationship between insufficient RV recovery following PVR and subsequent outcomes remains unknown.

Furthermore, occurrence of malignant ventricular arrhythmia is the leading cause of major events in patients with repaired ToF. Contrast-enhanced CMR techniques can detect focal areas of myocardial fibrosis (ie, late gadolinium enhancement, LGE) that represent substrates for arrhythmias as demonstrated previously.5 It should be noted that these focal areas of fibrosis are generally considered as irreversible and will probably not respond to PVR. Therefore, we may be looking at the wrong target. The association of LGE with outcomes in the INDICATOR cohort, however, remains unknown as LGE imaging was not assessed in this analysis. On a side note, although frequently detected at points of previous surgical interventions such as the RV outflow tract and the ventricular septal defect patch, myocardial LGE outside these areas has been rarely described (~5%) in current studies, which questions its role for reliable risk stratification.6 7 The potential advantage of more advanced CMR imaging techniques that quantify diffuse myocardial fibrosis (CMR T1 mapping) seems promising in view of the first published reports but needs future investigation regarding its predictive value and its reversibility following PVR.7

An important finding of this study is that patients who underwent PVR who did not meet ‘proactive’ criteria had a significantly higher rate of secondary events while the effect on primary outcome variables was not significant. The authors attributed this finding to the haemodynamic stability and the resulting low risk for adverse events in these patients compared with the risks of major cardiac PVR surgery including cardiopulmonary bypass and damage to the myocardium by additional incisions. Interestingly, the study by Yim and colleagues further supports such an association as they found bypass and cross-clamp time at primary repair to be associated with a higher degree of biventricular fibrotic myocardial changes.6 In this context, the technique of percutaneous pulmonary valve implantation, which is more and more commonly used even in patients with large, native RV outflow tracts, may be a superior alternative to surgical PVR since it avoids cardiopulmonary bypass and surgical scars. Some questions remain regarding the long-term durability of the prosthesis, the need for reinterventions and a potential susceptibility for endocarditis. Reassuringly, current studies on early outcomes did not report a difference between the two approaches, but duration of hospitalisation was shorter using a transcatheter strategy.8 Unfortunately, in the current study, the number of patients treated with percutaneous valves was too small to allow a subgroup analysis within the cohort, but hopefully larger number of data sets will help to clarify this question.

By further expanding the study cohort, continuing to assess longitudinal follow-up data and the potential to include novel imaging parameters, the INDICATOR registry will certainly keep on helping us to address essential questions in the future management of patients with repaired ToF. But until these data are available, the jury regarding the question of PVR in patients with ToF is still out.