Response to: Correspondence on ‘Hypertensive response to exercise in adult patients with repaired aortic coarctation’ by Pavšič et al

The Authors' reply: We thank Pavšič et al for their remarks regarding our recent article. They raise some important points. First, the definition of a hypertensive response to exercise varies between studies in patients with coarctation of the aorta (CoA), which limits their comparability. We used a cutoff value of 210 mm Hg in men and 190 mm Hg in women for systolic blood pressure (SBP) during peak exercise, since these values correspond to the 90th percentile in both sexes in a healthy population. Although this definition is most widely used, we acknowledge that there are potential drawbacks when extrapolating this definition to a relatively young cohort of patients with CoA. As noted by Pavšič et al, contemporary data indicate that peak exercise SBP increases with age in a pattern similar to resting SBP. Since most patients in our cohort were between 18 and 40 years old, we may have detected an even higher prevalence of a hypertensive response to exercise using ageadjusted cutoff values. However, we believe we should be cautious in comparing patients with CoA with apparently healthy individuals. Patients with CoA represent a very distinct group with a high prevalence of hypertension and signs of a generalised arteriopathy, which are often already present from a young age. At this moment, there is insufficient evidence how this generalised arteriopathy progresses over time. Interestingly, our presented data show a trend towards a lower peak exercise SBP with increasing age, even when corrected for workload (table 2 in original article). This may be partly explained by differences in surgical era. Repair techniques have improved over the last decades, which has most benefited the younger patients in our cohort. Consequently, even the patients with the most severe forms of arteriopathy, who presumably have the highest risk of a hypertensive response to exercise, survive into adulthood in reasonable condition. In contrast, patients >50 years whose exercise tolerance is well enough to undergo exercise stress testing may represent a subgroup with a relatively mild arteriopathy. These era differences are likely to introduce some degree of selection bias, which should be taken into account when interpreting exercise stress testing in patients with CoA. Pavšič et al argue that workloadindexed SBP better reflects an abnormal blood pressure response than SBP alone. Workload may indeed confound the relationship between exercise and SBP, which is why we reported workload in metabolic equivalents (METs) and adjusted for this factor in multivariable analysis. It is plausible to adjust for workload, since there is a nearly linear relationship between workload and cardiac output during exercise. However, it has not yet been demonstrated that workloadindexed SBP is superior to SBP alone in predicting adverse cardiovascular events. The importance of preventing cardiovascular events in adult patients with CoA was emphasised by our recent study, showing a substantial burden of cardiovascular morbidity and mortality in this patient cohort. Since the most optimal marker to predict cardiovascular complications is not yet known, we propose to evaluate various exercise measures in upcoming studies. For instance, SBP recovery after exercise may have additional value in predicting future cardiovascular disease. Furthermore, MRIergometry and computational fluid dynamics bear a great potential to characterise flow profiles and pressure gradients during exercise. In conclusion, although it is desirable to have a uniform definition of a hypertensive response to exercise in patients with CoA, more studies are needed to determine which exercise measures are most strongly associated with the development of hypertension, cardiovascular events and mortality in this specific patient population. One should be cautious to extrapolate findings in a healthy population to patients with CoA, since patients with CoA have a distinct pathophysiology of hypertension and there may be a substantial effect of surgical era.