Abstract
Background
Stenting of coarctation of the aorta (CoA) generally results in good angiographic results and a decrease in transcoarctation pressure gradient. However, effect on blood pressure control is less clear. The goal of the current retrospective analysis was to investigate the effects of CoA stenting on blood pressure control.
Methods
A retrospective analysis was conducted in consecutive adult patients with a CoA who underwent a percutaneous intervention at one of the three participating hospitals. Measurements included office blood pressure, invasive peak-to-peak systolic pressure over the CoA, diameter of the intima lumen at the narrowest part of the CoA and use of medication. The follow-up data were obtained, based on the most recent examination date.
Results
There were 26 native CoA and 17 recurrent CoAs (total n = 43). Seven of them underwent two procedures. Mean peak-to-peak gradient decreased from 27 mmHg to 3 mmHg ( p < 0.001), and minimal diameter increased from a mean of 11 mm to 18 mm ( p < 0.001). Mean systolic blood pressure decreased from 151 ± 18 mmHg to 135 ± 19 mmHg at first follow-up of 3.8 ± 1.9 months and 137 ± 22 mmHg at latest follow-up of 19.5 ± 10.9 months ( p = 0.001 and p = 0.009, compared to baseline, respectively). The total number of hypertensive patients decreased from 74% to 27% at latest follow-up. No significant change in antihypertensive medication was observed.
Conclusion
A clinically significant decrease in systolic blood pressure of approximately 16 mmHg was shown after (re)intervention in CoA patients, which sustained at follow-up. This sustained decrease of blood pressure can be expected to lead to less future adverse cardiovascular events.
1
Introduction
Aortic coarctation (CoA) is a congenital restriction of the lumen of the aorta, commonly occurring as stenosis in the juxtaductal position [ ]. It accounts for 5–8% of all congenital heart disorders, but can also have an acquired cause (i.e. Takayasu) [ ]. CoA may be seen in isolation but frequently occurs in combination with other cardiac lesions like bicuspid aortic valve or ventricular septal defect [ ]. Its natural history carries a mean life expectancy of approximately 35 years, mainly resulting from hypertension related complications such as left ventricular failure (28%), intracranial haemorrhage (12%), aortic rupture/dissection (21%) and premature coronary artery disease, but also from infective endocarditis (18%) or associated heart defects [ ]. Unrepaired CoA frequently causes problems in early infancy, with 60% of the untreated patients with symptomatic high-grade CoA and 90% of the patients with complicated CoA dying during the first year of life. CoA may be corrected surgically with a high degree of anatomical success, but patients typically show high rates of hypertension, may suffer from restenosis, and in up to 10% recurrent CoA occurs with the need of a re-intervention [ ]. Up to one third of the patients develop hypertension during follow-up [ ].
As of the 1990s, invasive percutaneous treatment became widely accepted as treatment option for CoA. Initially, percutaneous treatment only consisted of expandable balloon dilatation, and relatively high rates of complications such as restenosis and balloon rupture were observed [ ]. With the introduction of the Cheatham Platinum stent (NuMED, Hopkinton, NY, USA) a way to overcome these drawbacks was introduced. Placement of a balloon-expandable stent soon became standard of care. Percutaneous stent placement as treatment for CoA leads to fewer acute complications and acute angiographic and long-term results comparable with surgical repair [ ]. Although the therapy virtually abolishes the risk of early left ventricular failure, even after relief of obstruction to a residual transaortic pressure gradient of <10 mmHg, long-term blood pressure control proves to be suboptimal in many patients [ ]. It is hypothesized that an underlying generalized arteriopathy is accountable for the persisting systemic hypertension even after treatment of the focal stenosis. Although not undebated, aortic arch hypoplasia and/or gothic arch have been described to be a substrate for systemic arterial hypertension [ , ]. Furthermore, reduced baroreceptor sensitivity and residual arch gradients may also be contributing factors [ ]. Alternatively, a sub-optimal result of the (surgical or catheter) intervention may account for residual aortic obstruction resulting in reactive hypertension. Age of initial CoA repair is also known to be of great influence on the risk of long term hypertension [ ].
Regardless of its aetiology, hypertension should be abolished, as it is associated with structural changes in the heart and blood vessels, leading to significant cardiovascular mortality and morbidity (i.e. stroke, acute myocardial infarction, aortic dilatation and dissection) [ ].
Recently, we observed a rather striking improvement in systemic blood pressure regulation in several CoA cases that underwent (re-)intervention with stent placement. This observation supports the notion that blood pressure control may be more directly related to residual CoA relief than generally held. The aim of this study was to investigate the effects of stenting on blood pressure control in adult CoA patients.
2
Methods
2.1
Patient population
We performed a retrospective observational chart review of all patients from the University Medical Center Utrecht, Radboud University Medical Center Nijmegen and St. Antonius Hospital Nieuwegein who underwent stent placement for CoA from 2009 to 2015. Data on demography, biometry, medication use, blood pressure, cardiac imaging parameters, and stent procedures were collected from the electronic patient file. Patients who underwent surgical repair or balloon dilatation without stent placement were excluded. Patients were divided in native CoA (no previous intervention), recurrent CoA, and patients with multiple interventions (interventions >1 year apart).
2.2
Measurements
In accordance to the European Society of Cardiology (ESC) protocol [ ], blood pressure was measured by cuff sphygmomanometer on the right arm and leg. In case of arteria lusoria, the left arm was used. Hypertension was defined as systolic blood pressure > 140 mmHg on the upper limb, regardless of medication [ ]; when reliable office measurements were not available, ambulatory blood pressure monitoring was analysed [ ]. A non-invasive arm-leg gradient of >20 mmHg was considered suggestive of CoA [ ]. The echocardiographic parameters used to specifically assess presence of CoA were spectral and/or colour Doppler patterns in descending or abdominal aorta. Furthermore, left ventricle (LV) dimensions, LV systolic function and presence of LV hypertrophy were evaluated from 2D short and/or long-axis views. Non-invasive aortograms were obtained through Computed Tomography or Magnetic Resonance Imaging of the chest. The aortic lumen at the narrowest part of the CoA was measured and compared to the diameter of aorta at the diaphragm level. The presence and amount of collaterals was also assessed. To quantify the burden of the anti-hypertensive regime, Defined Daily Dose (DDD) index of the WHO [ ] was calculated, both prior to intervention and at follow-up.
2.3
Indication for catheterization/stenting
As determined by the multidisciplinary Grown Up Congenital Heart disease team, CoA patients were considered candidate for intervention if either of the following was present: (a) upper-limb hypertension or significant LV hypertrophy and a significant transcoarctation gradient (20 mmHg non-invasive, 15 mmHg invasive), (b) upper-limb hypertension or significant LV hypertrophy with an invasive gradient <15 mmHg in the presence of extensive collateral formation and clear substrate, or (c) ≥ 50% luminal narrowing at the CoA site as compared to the diaphragm level [ ].
2.4
Catheterization procedure
All patients underwent diagnostic catheterization for invasive pressure gradient measurements, (three-dimensional) angiography and, if indicated and deemed technically feasible, subsequent stent implantation. The procedures were performed under conscious sedation or general anaesthesia. Vascular access was achieved through the femoral artery using the standard Seldinger technique. Patients were heparinized (1.000 IE/10 kg). Using an open-end catheter, the gradient over the CoA was assessed. Stent placement was performed if the gradient was considered significant (>15 mmHg) and/or in case of the presence of hypertension with a clear substrate. The required size of the stent was determined by measuring the angiographic diameter of the transvers aortic arch. In case of aortic arch hypoplasia stent target diameter was chosen to fit the descending aorta at level of diaphragm. In recent cases, three-dimensional rotational angiography was used. In general, covered stents were preferred if estimated safe (i.e. side branch access). The stents that were used were the Cheatham Platinum (CP) covered and non-covered stent (NuMED, Hopkinton, NY, USA), Adventa V12 stent (Atrium, Mijdrecht, Netherlands), ev3 Max LD IntraStent (Covidien, Irvine, California, USA) and Andra Stent XXL (Andramed, GmbH, Reutlingen Germany). The stent was expanded on a Balloon-in-Balloon (BiB; NuMED, Hopkinton, NY, USA) and if deemed necessary post-dilated with a ultra-high-pressure balloon (Atlas; Bard Medical, Covington, GA). We prefer to flare the stent exit with the use of a Cristal Balloon (Balt, Montmorency, France). In case of severe narrowing (so called subatretic CoA), a relative under expansion was accepted, to lower the risk of dissection and dilate to final diameter during second procedure [ , ]. In case of intentional under expansion, a staged procedure for optimal post-dilatation was performed with a second procedure approximately 4–6 months after the index procedure. Femoral arterial access was closed with the use of two Perclose ProGlide suture-mediated closure systems (Abbott Vascular, Clonmel, Tipperary, Ireland). One day after the intervention, a Computed Tomography scan aortography was made to rule out dissection before discharge.
2.5
Multiple interventions
For this study, the catheterization procedures specifically planned for post-dilatation were considered an integral part of the index procedure; only after the post-dilatation procedure was performed, the index procedure was regarded as complete and follow-up began. Only patients who underwent a catheterization for suspicion of re-CoA at least one year after the index procedure were considered to have had multiple interventions. For patients who underwent multiple interventions, data on PG, diameter and blood pressure were collected after the last procedure.
2.6
Follow-up
Patients received follow-up care based on advice from the Grown-Up Congenital Heart disease team and clinical protocol [ ]. Data were collected at two time points after stent implantation. The first was scheduled after approximately three months. For the second and latest follow-up, data on the most recent examination or hospital visit were used. At both time points, upper limb blood pressure was determined.
2.7
Statistical analysis
All data were entered in IBM® SPSS® Statistics 21 software. Descriptive statistics were used for demographic and other descriptive data. Quantitative data were expressed as mean (SD) or frequency. Paired/unpaired student’s t -test was used for comparing group means. Wilcoxon signed rank test was used if indicated. Regression and correlation tests were done by calculating Pearson or Spearman’s coefficients. A p -value <0.05 was considered to represent statistical significance.
2
Methods
2.1
Patient population
We performed a retrospective observational chart review of all patients from the University Medical Center Utrecht, Radboud University Medical Center Nijmegen and St. Antonius Hospital Nieuwegein who underwent stent placement for CoA from 2009 to 2015. Data on demography, biometry, medication use, blood pressure, cardiac imaging parameters, and stent procedures were collected from the electronic patient file. Patients who underwent surgical repair or balloon dilatation without stent placement were excluded. Patients were divided in native CoA (no previous intervention), recurrent CoA, and patients with multiple interventions (interventions >1 year apart).
2.2
Measurements
In accordance to the European Society of Cardiology (ESC) protocol [ ], blood pressure was measured by cuff sphygmomanometer on the right arm and leg. In case of arteria lusoria, the left arm was used. Hypertension was defined as systolic blood pressure > 140 mmHg on the upper limb, regardless of medication [ ]; when reliable office measurements were not available, ambulatory blood pressure monitoring was analysed [ ]. A non-invasive arm-leg gradient of >20 mmHg was considered suggestive of CoA [ ]. The echocardiographic parameters used to specifically assess presence of CoA were spectral and/or colour Doppler patterns in descending or abdominal aorta. Furthermore, left ventricle (LV) dimensions, LV systolic function and presence of LV hypertrophy were evaluated from 2D short and/or long-axis views. Non-invasive aortograms were obtained through Computed Tomography or Magnetic Resonance Imaging of the chest. The aortic lumen at the narrowest part of the CoA was measured and compared to the diameter of aorta at the diaphragm level. The presence and amount of collaterals was also assessed. To quantify the burden of the anti-hypertensive regime, Defined Daily Dose (DDD) index of the WHO [ ] was calculated, both prior to intervention and at follow-up.
2.3
Indication for catheterization/stenting
As determined by the multidisciplinary Grown Up Congenital Heart disease team, CoA patients were considered candidate for intervention if either of the following was present: (a) upper-limb hypertension or significant LV hypertrophy and a significant transcoarctation gradient (20 mmHg non-invasive, 15 mmHg invasive), (b) upper-limb hypertension or significant LV hypertrophy with an invasive gradient <15 mmHg in the presence of extensive collateral formation and clear substrate, or (c) ≥ 50% luminal narrowing at the CoA site as compared to the diaphragm level [ ].
2.4
Catheterization procedure
All patients underwent diagnostic catheterization for invasive pressure gradient measurements, (three-dimensional) angiography and, if indicated and deemed technically feasible, subsequent stent implantation. The procedures were performed under conscious sedation or general anaesthesia. Vascular access was achieved through the femoral artery using the standard Seldinger technique. Patients were heparinized (1.000 IE/10 kg). Using an open-end catheter, the gradient over the CoA was assessed. Stent placement was performed if the gradient was considered significant (>15 mmHg) and/or in case of the presence of hypertension with a clear substrate. The required size of the stent was determined by measuring the angiographic diameter of the transvers aortic arch. In case of aortic arch hypoplasia stent target diameter was chosen to fit the descending aorta at level of diaphragm. In recent cases, three-dimensional rotational angiography was used. In general, covered stents were preferred if estimated safe (i.e. side branch access). The stents that were used were the Cheatham Platinum (CP) covered and non-covered stent (NuMED, Hopkinton, NY, USA), Adventa V12 stent (Atrium, Mijdrecht, Netherlands), ev3 Max LD IntraStent (Covidien, Irvine, California, USA) and Andra Stent XXL (Andramed, GmbH, Reutlingen Germany). The stent was expanded on a Balloon-in-Balloon (BiB; NuMED, Hopkinton, NY, USA) and if deemed necessary post-dilated with a ultra-high-pressure balloon (Atlas; Bard Medical, Covington, GA). We prefer to flare the stent exit with the use of a Cristal Balloon (Balt, Montmorency, France). In case of severe narrowing (so called subatretic CoA), a relative under expansion was accepted, to lower the risk of dissection and dilate to final diameter during second procedure [ , ]. In case of intentional under expansion, a staged procedure for optimal post-dilatation was performed with a second procedure approximately 4–6 months after the index procedure. Femoral arterial access was closed with the use of two Perclose ProGlide suture-mediated closure systems (Abbott Vascular, Clonmel, Tipperary, Ireland). One day after the intervention, a Computed Tomography scan aortography was made to rule out dissection before discharge.
2.5
Multiple interventions
For this study, the catheterization procedures specifically planned for post-dilatation were considered an integral part of the index procedure; only after the post-dilatation procedure was performed, the index procedure was regarded as complete and follow-up began. Only patients who underwent a catheterization for suspicion of re-CoA at least one year after the index procedure were considered to have had multiple interventions. For patients who underwent multiple interventions, data on PG, diameter and blood pressure were collected after the last procedure.
2.6
Follow-up
Patients received follow-up care based on advice from the Grown-Up Congenital Heart disease team and clinical protocol [ ]. Data were collected at two time points after stent implantation. The first was scheduled after approximately three months. For the second and latest follow-up, data on the most recent examination or hospital visit were used. At both time points, upper limb blood pressure was determined.
2.7
Statistical analysis
All data were entered in IBM® SPSS® Statistics 21 software. Descriptive statistics were used for demographic and other descriptive data. Quantitative data were expressed as mean (SD) or frequency. Paired/unpaired student’s t -test was used for comparing group means. Wilcoxon signed rank test was used if indicated. Regression and correlation tests were done by calculating Pearson or Spearman’s coefficients. A p -value <0.05 was considered to represent statistical significance.
3
Results
3.1
Demographic data
We identified a total of 57 CoA patients who underwent an intervention between 2009 and 2015, of which 43 underwent balloon dilatation with stent placement. Data on first follow up (mean 3.8 ± 1.9 months) were available for 35 patients. Data on latest follow up (mean 19.5 ± 10.9 months) were available for 26 patients. Population characteristics are shown in Table 1 .
| Variable | Patients (n = 43) |
|---|---|
| Age, ( years) | 41 ± 15 |
| Male | 26 (60.5%) |
| Body Mass Index (kg/m 2 ) | 25 ± 4.6 |
| Classification | |
| Native CoA | 26 (60.5%) |
| Recurrent CoA | 17 (39.5%) |
| Early surgical repair | |
| Patch angioplasty | 5 (11.6%) |
| End-to-end anastomosis | 1 (2.3%) |
| Aorta graft prosthesis | 1 (2.3%) |
| Complex procedure | 6 (14%) |
| Unknown | 8 (18.6%) |
| Associated cardiac anomalies | |
| Bicuspid aortic valve | 22 (51.2%) |
| Patent ductus arteriosus | 5 (11.6%) |
| Ventricular septal defect | 3 (7%) |
| Comorbidities | |
| Turner syndrome | 1 (2.7%) |
| Noonan syndrome | 1 (2.7%) |
| Alagille syndrome | 1 (2.7%) |
3.2
Procedural data
A total of 50 interventions were performed; 43 were index procedures, 7 patients underwent two interventions. In the current study, procedural success, defined as a PG ≤ 10 mmHg, was achieved in 88% of cases. Stent implantation was performed in 3 out of the 7 of the unplanned procedures during follow-up. Eight patients had two stents implanted during the re-intervention.
There were no deaths related to the procedures. During 2 procedures (4%), tearing of the intimal lining of the aorta occurred. The ensuing local dissections were treated successfully with covered stent implantation without sequelae on subsequent imaging. One patient suffered an intra-cerebral haemorrhage several hours after the procedure. CT showed no circle of Willis anomalies, all haemostasis parameters were found to be normal. The haemorrhage was most likely due to the combination of heparinization and hypertensive episodes during and after the procedure. Mild neurological deficits remained during follow up.
3.3
Peak-to-peak gradient
At baseline, the mean peak-to-peak pressure gradient (PG) was 27.5 ± 19.2 mmHg and decreased to a mean PG post intervention of 3.5 ± 5.0 mmHg ( p < 0.001). Data are shown in Table 2 . No significant relation was found between difference in systolic blood pressure and initial PG ( p = 0.07; Fig. 1 ).
