Aortic coarctation (AC) was first described by Morgagni in 1760 and was characterized clinically in the early 20th century.1,2 It is a common malformation, found in 5–8% of children with congenital heart defects.3,4 It consists of a severe narrowing of the proximal descending aorta, creating a significant gradient in the descending aorta. If left untreated it is associated with poor prognosis and high mortality.5 Development of surgical techniques and more recently percutaneous aortic angioplasty with stent implantation has significantly improved the prognosis of these patients.6–8 However, AC cannot be considered a simple and easily corrected lesion. Even after successful repair, 20–40% of patients continue to have hypertension (HTN) at 10–20 years of follow-up and 20–35% have an abnormal blood pressure (BP) response to exercise despite being normotensive at rest.9–12 The pathophysiology of this abnormal BP behavior is unclear and different mechanisms have been proposed. Exercise-induced HTN has been attributed to abnormalities in aortic arch geometry, primary baroreceptor alterations and persistent structural and functional abnormalities of the vascular wall of the aorta and large arteries proximal to the former coarctation site.9 Moreover, life expectancy, although better than in untreated patients, is still less than in healthy subjects.13

There is ongoing controversy regarding the significance of abnormal exercise BP after AC repair.14 In this study we aimed to assess the prevalence of exercise-induced HTN in adult patients with repaired AC.

We included 65 patients, 40 (61.5%) male, mean age 30±8 years (19–57). Ten (15.4%) had dyslipidemia and only one (1.5%) was diabetic. Fifty-eight (89%) patients were under antihypertensive therapy (54 [83.1%] with beta-blockers and 15 [23.1%] with angiotensin-converting enzyme inhibitors [ACEI]). Eleven (17%) patients were treated with more than one antihypertensive drug. The majority of patients had controlled BP at rest and only nine (18%) were still under antihypertensive medication. Mean systolic BP at rest was 128±14 mmHg and mean diastolic BP at rest was 73±9 mmHg.

An additional congenital heart defect was identified in 27 (42%) patients: bicuspid aortic valve (n=18), ventricular septal defect (n=4), patent ductus arteriosus (n=3), mitral valve prolapse (n=1) and ostium secundum atrial septal defect (n=1).

Twenty-four (37%) patients had undergone AC repair within the first year of life and 40 (62%) later on (29 [45%] between one and 17 years old and 11 [17%] at 18 years or older). Mean follow-up was 20±7 years.

The majority of patients (32%) had undergone subclavian flap aortoplasty, 28% prosthetic patch aortoplasty and 19% had an end-to-end aortic anastomosis. Only 6% of patients had undergone balloon angioplasty. In 10 (15%) cases the type of surgery was not available.

Among these patients, eight (12.3%) were reoperated during the study follow-up and all patients were in New York Heart Association functional class I.

According to the TTE evaluation at follow-up 64 (98.5%) patients had preserved LV systolic function. LV hypertrophy was identified in five (7.7%) patients and mean peak instantaneous gradient in the descending aorta was 23.1±12.2 mmHg.

The results of the symptom-limited treadmill exercise test, according to the Bruce protocol, were available in 49 (75%) patients (33 [67%] male). The mean duration of exercise stress testing was 10.7±3.1 minutes and the mean peak heart rate achieved at peak workload was 166±18 beats per minute. Mean peak systolic BP was 181±37 mmHg. Eleven (22%) patients had a hypertensive response to the exercise stress test. Of these, only three (33%) had uncontrolled BP at rest. There were no differences between the groups (exercise-induced HTN patients versus non-exercise-induced HTN patients) in gender, age at surgery, time since surgery or prevalence of uncontrolled BP at rest. There were also no differences in the presence of LV hypertrophy on TTE, exercise duration, peak heart rate at exercise, current treatment with beta-blockers or need for reoperation. However, current treatment with ACEI (OR 4.0 [95% CI 1.9–18.1]) and a higher peak instantaneous gradient in the descending aorta on TTE (OR 8.2 [95% CI 1.8–37.0]) were associated with exercise-induced HTN in our study population (Table 1).

Our results show that in long-term follow-up, HTN at rest and exercise-induced HTN are prevalent in a significant percentage of adult patients even after successful AC repair, which is in agreement with previous studies.9–12 “Repair” may not be synonymous with “cure” in the case of cardiovascular abnormalities. Systemic HTN can occur late after AC repair and may be due to residual or recurrent coarctation, but even patients with a successful repair may develop systemic HTN due to anatomical and functional changes in the arterial wall.9 Moreover, physiological abnormalities of large and small arteries in the precoarctation vascular bed may be an important contributor to exercise-related HTN and late morbidity or mortality.16

An abnormally high systolic BP on exercise was found in patients with adequate BP control at rest and in those under antihypertensive therapy with ACEI. The prognostic significance of exercise-induced HTN in patients with AC is still unknown, but there is ongoing interest in its potential predictive value for late systemic HTN. In this context, the exercise-induced HTN observed in these patients could hypothetically identify higher-risk patients who could benefit from a different target for BP control.

Several other studies have demonstrated an exaggerated exercise systolic BP response in patients with repaired AC and questions have been raised over its relation to outcome and target organ damage. Some studies have shown that peak exercise systolic BP in patients with AC is strongly associated with mean daytime systolic BP on ambulatory monitoring.17 A more recent prospective study showed that peak exercise systolic BP was a predictor of chronic HTN in patients with AC.18 Furthermore, exercise-induced HTN is associated with increased intima-media thickness of the common carotid artery, which is a validated and reproducible surrogate endpoint for atherosclerosis.19,20 Several investigators have suggested that patients with exercise-induced HTN might be amenable to antihypertensive therapy or limitation of strenuous physical activity.21 In fact, exercise-induced HTN might become a therapeutic target if an unequivocal link with adverse outcome is demonstrated in this population.

In this study patients under beta-blocker therapy more frequently had a normal BP response to exercise, suggesting that these patients could be under more effective BP control. Moltzer et al. showed that in adult hypertensive patients after AC repair, metoprolol had a greater antihypertensive effect than candesartan. In their study the neurohormonal outcome did not support a significant role for the renin-angiotensin system in the causative mechanism of HTN after AC repair.22 These findings may explain the association of ACEI and the exercise-induced HTN found in our study population. However, additional studies are needed to define the optimal antihypertensive treatment in patients after AC repair.

In contrast to other published studies,12,23 we found no correlation between age at surgery and exercise-induced HTN. In our study the mean age at surgery was similar in both groups, with and without exercise-induced HTN, and exercise-induced HTN was not more prevalent in patients operated after the first year of life. However, the small sample size may explain this finding.

Similar to previous reports,17,24 we found no significant relationship between LV hypertrophy and exercise BP variables.

By contrast, the correlation between exercise-induced HTN and a higher peak instantaneous gradient in the descending aorta by TTE was significant in our cohort. Exercise testing may be an efficient method for follow-up monitoring of these patients, identifying patients in need of further investigation, and for medical therapy optimization. In our population, patients with higher descending aorta gradients underwent additional investigation with cardiac magnetic resonance or cardiac catheterization. Those with confirmed recoarctation underwent surgical or interventional treatment. However, we found no significant correlation between reoperation and exercise-induced HTN. The small number of reinterventions in this population may explain this finding.

These results highlight the complexity of the adult AC population and the fact that, even with a good result after repair, several patients remain at high cardiovascular risk in the long term. Recent findings showing the predictive value of peak exercise systolic BP for the development of chronic HTN in these patients18 focus attention on this parameter. Further studies will be needed to clarify the long-term consequences of exercise-induced HTN and the potential role of early antihypertensive treatment in this population.

A considerable proportion of repaired AC adult patients have exercise-induced HTN. In our population, patients with a higher TTE peak instantaneous gradient in the descending aorta and patients under antihypertensive therapy with ACEI are at greater risk of exercise-induced HTN. Treadmill exercise testing can detect an abnormal BP response with exercise early during follow-up after AC repair and can be an efficient method of identifying patients at high cardiovascular risk, providing a screening tool for further investigation or therapy optimization.

The clinical significance of exercise-induced HTN in patients late after AC repair is still a matter of debate. Further research is required in order to clearly define its additional cardiovascular risk and the potential protective role of antihypertensive treatment in these patients.

The authors have no conflicts of interest to declare.