#Both authors have equally contributed to this work
Coarctation of the aorta (CoA) is the sixth most common cardiovascular malformation, accounting for 5–8% of all congenital heart diseases, and is mostly diagnosed during infancy [
We included 212 patients with native or recurrent CoA corrected at our institution, the Bambino Gesù Children’s Hospital, from 1983 to 2018 (with first stent positioning in 2002). We excluded patients with genetic syndromes (e.g., Turner syndrome) or associated significant congenital heart defects, except for bicuspid aortic valve, which was present in 82% of the study population and had normal function or only trivial regurgitation. We also excluded patients with aortic arch hypoplasia (any aortic segment with Z-score lower than −2 SD [
We divided patients into 3 groups depending on CoA treatment type: The surgical group (CoA-S) included 139 patients who underwent one-time surgical aortic repair via patch aortoplasty, subclavian flap repair or end-to-end anastomosis; The percutaneous stenting group (CoA-PS) included 18 patients with CoA repair by means of one-time endovascular aortic stent positioning; The hybrid group (CoA-H) included 55 patients who underwent multiple aortic procedures (with or without balloon angioplasty) because of reCoA occurrence.
We collected demographic and clinical data including height, weight, body mass index (BMI) and anti-hypertensive (HTN) drugs at baseline and at the latest follow-up. All patients treated with aortic stenting performed the same percutaneous procedure with implantation of balloon-expandable Cheatham-Platinum (CP) stent (NuMED, Inc., Hopkinton, NY, USA) at the site of CoA. When appropriate, we also collected surgical data regarding age at first repair, correction repair technique and subsequent reinterventions and/or complications. All patients underwent clinical examination including blood pressure (BP) measurement using an OMRON oscillometric device placed on the right arm in the seated position. BP values were measured at rest according to pediatric guidelines recommendations [
All patients performed ABPM (Spacelab 90207, Spacelab Inc., Redmond, Wash). We collected 24-hour mean systolic blood pressure (SBP) and pulse pressure (PP) values, as well as mean day-time and night-time SBP and percentage of dipping. Monitoring was considered reliable if at least 75% of readings were successful. Patients were defined as hypertensive if they had elevated BP values at rest according to hypertension pediatric guidelines or European Society of Hypertension guidelines for adults [
TTE was performed in all patients by an expert sonographer according to standardized technical methods and reviewed off-line with computerized review stations by two independent echocardiographists (Intellispace, Philips, Andover, USA). We recorded aortic gradient at the site of intervention(s) as well as left ventricular internal dimensions and septal and posterior wall thickness measured at end-diastole and end-systole using American Society of Echocardiography recommendations on 3 cardiac cycles [
Primary end-point was to identify which aortic repair technique (CoA-S versus CoA-PS versus CoA-H) was predictive of re-coarctation during a long-term follow-up. Secondary end-point was to evaluate the incidence of late arterial hypertension at ABPM after different types of aortic repair.
All continuous variables were assessed for normality with the Shapiro-Wilk test and by examination of their histogram. Variables with normal distribution were expressed as means and standard deviations and tested for differences using ANOVA with post-hoc Bonferroni correction and Student-T test, as appropriate. Non-parametric variables were expressed as median and interquartile range and differences tested using Kruskall-Wallis test and Mann-Whitney test, as appropriate. Categorical variables were expressed as percentages and analyzed by chi-squared test. Survival curves were estimated using the Kaplan-Meier product-limit estimator and compared using the log-rank test. Cox proportional hazards analysis was used to calculate the adjusted hazard ratios for each clinical variable. The final multivariable Cox regression model was selected via a stepwise approach based on minimization of Akaike Information Criterion. Only
The study population included 212 CoA patients (male, 72%) with a median follow-up of 17 years (IQ range 11–24 years). Mean age at data collection was 19 ± 8.7 years with 47% patients in the pediatric age (<18 years). Baseline characteristics of the overall study population are shown in
Patients, n | 212 |
---|---|
Male, n (%) | 152 (72) |
Follow-up, years (median, IQ range) | 17 (11–24) |
Age at follow-up, years (median, IQ range) | 19 (12–26) |
BMI, Kg/m2 (mean ± SD) | 22 ± 4 |
Total Surgical procedures, n (%) | 153 (72) |
End-to-end anastomosis, n (%) | 85 (56) |
Patch aortoplasty, n (%) | 37 (24) |
Subclavian Flap, n (%) | 31 (20) |
Total aortic stenting procedures, n (%) | 39 (18) |
HTN medications, n (%) | 115 (54) |
ACE-I/ARBs, n (%) | 53 (46) |
BBs, n (%) | 12 (10) |
Multiple therapy |
50 (44) |
139 (66) | |
End-to-end anastomosis, n (%) | 83 (60) |
Patch aortoplasty, n (%) | 25 (18) |
Subclavian Flap, n (%) | 31 (22) |
18 (8) | |
55 (26) | |
Multiple surgical procedures, n (%) | 4 (7) |
Multiple percutaneous procedures, n (%) | 5 (9) |
Surgical and percutaneous procedures, n (%) | 46 (84) |
Note: ACE-I: Angiotensin Converting Enzyme Inhibitors. ARBs: Angiotensin II Receptor Blockers. BBs: Beta Blockers. BMI: Body Mass Index. CAs: Calcium Antagonists. HTN = anti-hypertensive. reCoA: reCoarctation of the Aorta.
At last follow-up visit, only 5 patients (2%) had arm-to-leg gradient ≥20 mmHg (median 25 mmHg, IQ range 22–34 mmHg), 4 of which were children with previous aortic stenting. Conversely, at TTE 44.8% of the total cohort had an aortic gradient ≥20 mmHg; among those, a significant higher proportion of children had an ongoing HTN treatment compared to those without any medication (54%
Overall, 54% of the total population had anti-hypertensive therapy. Medical treatment was performed with single therapy in the majority of patients, using angiotensin converting enzyme inhibitors, angiotensin II receptor blockers or beta blockers. However, 44% of hypertensive patients needed multiple therapies with different drug combinations involving the above-mentioned ACE-I, ARBs and BBs with diuretics or calcium antagonists.
Age at first repair was higher in the CoA-PS group compared to both other 2 groups. Similarly, BMI was significantly higher in the CoA-PS group compared to the CoA-S group. Differences in the need of HTN therapy and type of medications are shown in
CoA-S group (n = 139) | CoA-PS group (n = 18) | CoA-H group (n = 55) | p S |
p PS |
p S |
|
---|---|---|---|---|---|---|
Age at repair, days/years (mean ± SD) | 39 days (18 d–2.6 y) | 12 years (9 y–16 y) | 23 days (11 d–2 y) | 0.07a | ||
Male, n (%) | 95 (68) | 10 (55) | 47 (85) | nsb | ||
BMI, Kg/m2 (mean ± SD) | 21 ± 4 | 24 ± 3 | 23 ± 4 | nsc | nsc | |
Need of HTN therapy, n (%) | 64 (46) | 15 (83) | 36 (65) | nsb | ||
HTN therapy with ACE-inhibitors/ARBs, n (%) | 34 (64) | 5 (9) | 14 (26) | |||
HTN therapy with BBs, n (%) | 7 (41) | 5 (29) | 5 (29) | nsb | nsb | nsb |
Multiple HTN therapy, n (%) | 23 (51) | 5 (11) | 17 (38) | nsb |
Note: ACE: Angiotensin Converting Enzyme. ARBs: Angiotensin II Receptor Blockers. BBs: Beta Blockers. BMI: Body Mass Index. HTN = anti-hypertensive. aMann-Whitney test, bChi square, cANOVA with Bonferroni correction.
We observed 9% of the whole population with elevated 24-hour SBP values, 7.5% with elevated daily SBP values and 9% with elevated nocturnal SBP values. CoA-PS patients had a higher proportion of mean day-time SBP values exceeding the normal value compared to CoA-S patients (22%
CoA-S group (n = 139) | CoA-PS group (n = 18) | CoA-H group (n = 55) | p S |
p PS |
p S |
|
---|---|---|---|---|---|---|
Mean 24hSBP, mmHg (mean ± SD) | 116 ± 10 | 121 ± 8 | 118 ± 10 | nsb | nsb | nsb |
24h Pulse Pressure, mmHg (mean ± SD) | 52 ± 11 | 58 ± 8 | 57 ± 11 | 0.085 | ns | 0.020 |
Day-time SBP, mmHg (mean ± SD) | 120 ± 10 | 125 ± 8 | 121 ± 10 | nsb | nsb | nsb |
Number of mean daytime SBP values > 95th centile, n (%) | 9 (6) | 4 (22) | 3 (5) | nsa | nsa | |
Night-time SBP, mmHg (mean ± SD) | 106 ± 10 | 112 ± 11 | 107 ± 10 | 0.075b | nsb | nsb |
Nocturnal dipping, mmHg (mean ± SD) | 11 ± 5 | 10 ± 7 | 11 ± 4 | nsb | nsb | nsb |
Note: PP: Pulse Pressure. SBP: Systolic Blood Pressure. aChi square, bANOVA with Bonferroni correction.
Echocardiography showed that the number of patients with significant aortic gradient was higher in CoA-PS (50%) and CoA-H (73%) groups compared to CoA-S (33%) group (
CoA-S group (n = 139) | CoA-PS group (n = 18) | CoA-H group (n = 55) | ||
---|---|---|---|---|
IVSd, mm (mean ± SD) | 7.9 ± 1.7 | 8.5 ± 1.2 | 8 ± 1.4 | 0.396 |
LVIDd, mm (mean ± SD) | 48 ± 7 | 49 ± 6 | 49 ± 7 | 0.407 |
LVPWd, mm (mean ± SD) | 7 ± 1.6 | 8 ± 1.2 | 8 ± 1.7 | 0.069 |
RWT, mm median (IQR) | 0.30 (0.27–0.34) | 0.35 (0.29–0.38) | 0.32 (0.27–0.36) | 0.065 |
LVMi (mean ± SD) | 43 ± 11 | 47 ± 12 | 44 ± 12 | 0.190 |
LVM OBG (mean ± SD) | 41.5 ± 14 | 42 ± 10 | 41 ± 12 | 0.948 |
Note: IVSd: end diastolic interventricular septum. LVIDd: end diastolic left ventricular internal diameter. LVPWd: end diastolic left ventricular posterior wall. RWT: relative wall thickness. LVMi: left ventricular mass index (normalized for body surface area). LVM OBG: left ventricular mass normalized for an allometric power of height using OBG-Ospedale Bambino Gesù formula (Chinali et al. [
At Kaplan Meier survival analysis, CoA-PS group significantly showed a higher re-coartaction rate (log rank
Coarctation of the aorta is a common congenital cardiac malformation mainly diagnosed in infancy and early childhood and surgical repair has been for many years the treatment of choice in neonates, infants and young children [
In this heated debate about which technique could be best for isolated CoA treatment, literature lacks of long-term results after different repair techniques, particularly for percutaneous stent positioning in the pediatric age. Previous studies showed that surgical repair is associated to lower incidence of reCoA, lower need of re-intervention and lower residual aortic gradient compared to balloon angioplasty [
As far as we know, this retrospective study for the first time directly compares long-term effects of aortic surgery versus aortic stenting in a large cohort of children and adults, focusing on blood pressure control and freedom from re-coarctation. We only included isolated forms of CoA as we recently demonstrated that the presence of associated cardiac anomalies results in a significantly lower incidence of late-onset arterial hypertension [
As mentioned above, chronic arterial hypertension is known to be the major long-term morbidity even after successful aortic repair and in the absence of aortic re-obstruction [
Consistent with this, our data showed that patients with aortic stenting have a higher need for HTN therapy compared to those with previous surgical or hybrid CoA repair. Moreover, we observed at ABPM significantly higher PP values in CoA-PS and CoA-H groups compared to CoA-S group. PP is a simple surrogate of arterial stiffness measure and it has been reported to be abnormally elevated in patients with repaired CoA [
Finally, in order to avoid bias between surgery and percutaneous stenting (mostly due to significant difference in age at first repair), we demonstrated at multivariate Cox regression analysis that stenting was the strongest predictor of reCoA in the long-term follow up, irrespective from age at first aortic repair, gender, HTN medications and elevated BMI.
In conclusion, this study evaluated the long-term ABPM and TTE outcomes in a large cohort of children and adults with previous isolated CoA treated with surgery, stent implantation or both. We demonstrated that patients with aortic stenting had higher prevalence of late arterial hypertension with higher need for HTN therapy and higher rate of significant aortic gradient. Much more, aortic stenting emerged as the best independent predictor of re-coartaction, irrespective from age at repair, need and type of antihypertensive therapy or other CoA repair techniques.
Although being yet far from the solution of the problem (i.e., which technique is best for isolated CoA repair), our data suggest that surgery may probably be still the best option for native CoA in the pediatric age. On the other hand, looking at recoarctation scenarios, treatment should be probably tailored to each patient. Children and adolescents before puberty may experience the cons of stent positioning for reCoa treatment (need of multiple dilation procedures with multiple radiation exposures, stent’s failure to adapt to a growing aorta, hypertensive burden as suggested by our data). Instead, patients after pubertal development may be akin to adults and stenting procedure may be the best choice for reCoa correction in this context.
The main limitation of our study is the heterogeneity of groups, mostly regarding age at first repair, which is higher in patients with aortic stenting. This is not perfectible because percutaneous treatment needs by definition higher age and weight of patients, compared to surgery. Thus, age-matched groups are not feasible. As much as possible, we reduced this discrepancy by applying multivariate Cox regression analysis, and primary end-point proved to be irrespective of age at repair.