Author + information
- Received February 14, 2013
- Revision received May 20, 2013
- Accepted June 11, 2013
- Published online September 10, 2013.
- Morgan L. Brown, MD, PhD∗,
- Harold M. Burkhart, MD∗∗ (, )
- Heidi M. Connolly, MD†,
- Joseph A. Dearani, MD∗,
- Frank Cetta, MD†,‡,
- Zhuo Li, MS§,
- William C. Oliver‖,
- Carole A. Warnes, MD† and
- Hartzell V. Schaff, MD∗
- ∗Division of Cardiovascular Surgery, Mayo Clinic, Rochester, Minnesota
- †Division of Cardiovascular Diseases, Mayo Clinic, Rochester, Minnesota
- ‡Division of Pediatric Cardiology, Mayo Clinic, Rochester, Minnesota
- §Division of Biostatistics, Mayo Clinic, Rochester, Minnesota
- ‖Division of Anesthesiology, Mayo Clinic, Rochester, Minnesota
- ↵∗Reprint requests and correspondence:
Dr. Harold M. Burkhart, Mayo Clinic, 200 1st Street SW, Rochester, Minnesota 55905.
Objectives The objective of our study was to review the long-term outcomes of patients undergoing surgical repair of aortic coarctation.
Background Surgical repair of aortic coarctation has been performed at the Mayo Clinic, Rochester, Minnesota, for over 60 years.
Methods Between 1946 and 2005, 819 patients with isolated coarctation of the aorta underwent primary operative repair. Medical records were reviewed and questionnaires mailed to the patients.
Results Mean age at repair was 17.2 ± 13.6 years. The majority (83%) had pre-operative hypertension. Operations included simple and extended end-to-end anastomosis (n = 632), patch angioplasty (n = 72), interposition grafting (n = 49), bypass grafting (n = 30), and subclavian flap or “other” (n = 35). Overall early mortality (<30 days) was 2.4%. In the previous 30 years (n = 225), there were no operative deaths. Mean follow-up was 17.4 ± 13.9 years, with a maximum of 59.3 years. Actuarial survival rates were 93.3%, 86.4%, and 73.5% at 10, 20, and 30 years, respectively. When compared to an age- and sex-matched population, long-term survival was decreased (p < 0.001). Older age at repair (>20 yrs) and pre-operative hypertension were associated with decreased survival (p < 0.001). Patients age <9 years age at repair had significantly less hypertension at 5 to 15 years of follow-up (p < 0.001). Rates of freedom from re-intervention on the descending aorta were 96.7%, 92.2%, and 89.4% at 10, 20, and 30 years, respectively. Younger age at time of repair (p < 0.001) and an end-to-end anastomosis technique (p < 0.001) were independently associated with lower rates of re-intervention on the descending aorta.
Conclusions Primary repair of isolated coarctation of the aorta was performed with a low rate of mortality. However, long-term survival was reduced compared with that in an age- and sex-matched population, and many patients required further reoperation. These findings emphasize that patients with aortic coarctation need early recognition and intervention, as well as lifelong informed follow-up.
Aortic coarctation is defined as a narrowed aortic segment, most commonly located near the ligamentum arteriosum adjacent to the left subclavian artery (1). It can be discrete, long, or complex, involving the aortic arch or isthmus, and may have collateral vessels (1). Coarctation of the aorta has an estimated incidence of 1 in 2,500 live births, with a 2:1 predominance in males (2,3). In historical series, untreated aortic coarctation led to a mortality of more than 80% by age 50 years due to complications including aortic rupture, heart failure, and intracranial hemorrhage (4).
The first successful repair of aortic coarctation was described by Dr. Crafoord in 1945 (5), and the first repair at the Mayo Clinic was in 1946. Following surgical repair of aortic coarctation, patients experience improved survival, with age at time of repair being the most important predictor of survival (6). In general, significant coarctation repair should occur as early as possible. However, the optimal timing of this intervention is still not clear, as it is a balance between the risks for early and late mortality, reoperation, and the long-term effects of hypertension.
Our objective was to review the experience of repair of aortic coarctation over 60 years at the Mayo Clinic to identify predictors of long-term survival, reoperation, and late hypertension.
Institutional review board approval was obtained, which obviated individual patient consent. From August 1946 to July 2005, 819 patients had primary operative repair of isolated aortic coarctation. We excluded patients who had complex left-sided cardiac lesions, interrupted aortic arch, or Shone’s complex.
Perioperative data were collected using all available medical records. Follow-up data were collected using all medical records, and surveys were mailed to all patients not known to be deceased. Survival was ascertained through Accurint.com.
As a part of quality assurance at the Mayo Clinic, all patients who have undergone cardiac surgery are sent a survey on a scheduled basis. This survey includes a question that asks whether a diagnosis of high blood pressure was present, as well as inquiring regarding current medications and any cardiac catheterization procedures or surgical procedures performed. In patients in whom an intervention is reported, medical records are requested from outside institutions in an attempt to confirm the procedure indication and type. The presence of hypertension was determined through repeated elevated blood pressure measurements using standard definitions of hypertension (adults: systolic blood pressure ≥140 mm Hg and diastolic blood pressure ≥140 mm Hg) (7) and for children as systolic and/or diastolic blood pressure in the ≥95th percentile (8), the prescription or report of antihypertensive medications, or the self-report of a diagnosis of hypertension.
Statistical analyses were performed using SAS version 9.1 (SAS Institute Inc, Cary, North Carolina). Data are presented as mean ± SD, median (range), or number (%), as appropriate. Univariate and stepwise multivariate models were created using Cox proportional hazards. Kaplan-Meier curves were compared using a log-rank test. Cutoff values were identified using an SAS macro designed to find the best cutoff point of a continuous variable based on a chi-square statistic for a binomial outcome.
Pre-operative characteristics are listed in Table 1. Operations included end-to-end anastomosis in 631 patients, patch angioplasty in 73, interposition graft in 49, bypass graft in 30, and subclavian flap or other in 36. These operations were predominantly performed prior to the 1980s (n = 612; 74.7%). The type of repair has changed over the decades, with older patients and more complex repairs predominating in later decades (Fig. 1). Overall early mortality (<30 days) was 2.4%. In the previous 30 years (n = 225), there were no operative deaths.
Mean follow-up was 17.4 ± 13.9 years, with a maximum of 59.3 years. Four hundred fifty-nine patients had follow-up >30 days, and an additional 122 surveys were returned. Thus, data on mid- or late-term follow-up was obtained from 70.8% of all patients.
Actuarial survival rates were 93.3%, 86.4%, and 73.5% at 10, 20, and 30 years, respectively. Mean age of death was 34.2 ± 20.1 years. When compared to an age- and sex-matched population, long-term survival was decreased (p < 0.001) (Fig. 2). Older age at repair (>20 years) and pre-operative hypertension were associated with decreased survival (p < 0.001). On multivariate analysis, only increasing age was associated with mortality (hazard ratio [HR]: 1.06; 95% CI: 1.04 to 1.07; p < 0.001) (Fig. 3).
Rates of freedom from re-intervention on the descending aorta were 96.7%, 92.2%, and 89.4% at 10, 20, and 30 years, respectively (Fig. 4). Older age (HR: 0.93; 95% CI: 0.91 to 0.96; p < 0.001) and an end-to-end anastomosis technique (HR: 0.11; 95% CI: 0.07 to 20; p < 0.001) were independently associated with lower re-intervention rates on the descending aorta. Patients who had initial repair before age 1 year (n = 116), had a 31.1% re-intervention rate on the descending aorta rate at 30 years' follow-up. The highest rate of re-intervention on the descending aorta was in patients who were <5 years of age at the time of operative repair (73.3%; 95% CI: 64.1% to 83.9% at 30 years; p < 0.001; n = 192). In patients <5 years of age, simple and extended end-to-end anastomosis technique was associated with a decreased re-intervention rate, p < 0.013.
Rates of freedom from any cardiac reoperation were 92.8%, 85.7%, and 76.6% at 10, 20, and 30 years, respectively. The majority of reoperations were for repair of recoarctation or aortic valve disease (Table 2).
Data on follow-up regarding blood pressure status were available from 551 patients (67.3%). Mean follow-up was 17.4 ± 13.8 years, with a maximum of 59.3 years. The percentages of patients with hypertension at various follow-up intervals are presented in Fig. 5. The only predictor of hypertension up to 5 years postoperatively was older age at the time of operation (p < 0.001). A cutoff value of 9 years was identified; patients 9 years of age or older at the age of repair had significantly more hypertension at 5 to 15 years of follow-up (relative risk: 4.1; 95% CI: 2.1 to 8.4; p < 0.001). Pre-operative hypertension was not associated with late hypertension (p = 0.06).
Our study reports late outcomes in one of the largest published cohorts with isolated coarctation of the aorta. A previous report on this cohort from the Mayo Clinic found late survival of 91% at 10 years, 84% at 20 years, and 72% at 30 years (6). This was similar to our current report of survival (93% at 10 years, 86% at 20 years, and 74% at 30 years postoperatively). Surprisingly, improvements in operative technique, improved medical management of hypertension, surveillance for comorbid cardiovascular disease, and monitoring for complications related to the aorta have not made as much of an impact on late survival as expected.
The cause of death was not known in the majority of our patients due to the lack of autopsies performed. One might expect that these premature deaths are likely related to concomitant cardiovascular disease (6). A recent study from Quebec, Canada, questioned the prevalence of coronary artery disease in these patients later in life (9). However, others have found that patients with successfully repaired coarctation of the aorta have both endothelial dysfunction (10) and impaired arterial dilation in the pre-coarctation vascular bed (11), suggesting a possible vascular cause of late death. Normotensive coarctation patients have also been found to have increased left ventricular mass despite successful aortic coarctation repair (12). This increase in muscle mass may lead to subendocardial ischemia, late arrhythmias, and diastolic heart failure.
There have been changes in operative strategy for aortic coarctation over this 60-year period. Extended end to end is the current preferred technique for repair of aortic coarctation in neonates and children, while more complex techniques are reserved for complicated coarctations predominantly in older patients. The Mayo Clinic is a referral center for adult aortic coarctations, and as such, many repair techniques were utilized including ascending-to-descending bypass and aortic interposition grafting.
Reoperations or catheter-based re-interventions on the descending aorta were frequent. Age <1 year at the time of initial repair, and a repair technique other than simple or extended end-to-end anastomosis, were independently associated with the need for reoperation or re-intervention on the aorta. Younger age at operation has been reported as a risk factor for recoarctation and the need for late re-intervention in several studies (13,14). It is likely that in small children, the technical challenge of resection of all periductal tissue, as well as the small anastomosis size, may lead to an increased risk for late recoarctation. Patients who are diagnosed and present at a young age may also have more severe coarctation and therefore are also predisposed to late vascular complications. In general, an extended end-to-end anastomosis is the preferred type of repair. Patients who received other types of repairs such as subclavian flap or ascending-to-descending aortic bypass are more likely to have longer segments of stenosis or arch involvement. Unfortunately, not all forms of coarctation of the aorta are amenable to an end-to-end anastomosis.
In this cohort, a large number of other cardiovascular procedures were performed later in life, including aortic valve procedures, ascending aortic procedures, coronary artery bypass grafting, and mitral valve procedures. These data further underline the importance of ongoing follow-up in patients with aortic coarctation, and that repair of the coarctation does not mean a “cure.” We recommend annual clinical evaluation to assess blood pressure. Regular cardiac imaging is also recommended to assess the aorta (ascending aorta, arch, and descending aorta), valves, and ventricles. These patients should be followed up by physicians who have expertise in congenital heart disease.
In the most recent 225 patients, there were no operative deaths in our series. By comparison, Birmingham Children’s Hospital, Birmingham, Alabama, reported a 1% mortality (15), and Children’s Memorial Hospital, Chicago, Illinois, reported a 2% mortality (16) in surgical repair of aortic coarctation. Our results confirm that surgical correction of isolated aortic coarctation can be safely performed and any newer technologies, such as catheter-based intervention, should be held to these high standards.
Blood pressure data
Data on blood pressure readings or a self-report of hypertension was available in over 70% of patients. Older age at the time of repair was independently associated with late hypertension. Patients with repair of aortic coarctation before age 9 years had less hypertension at 5 to 15 years of follow-up. However, pre-operative hypertension was not associated with late hypertension. The presence and degree of hypertension in native aortic coarctation did not impact prognosis following repair, suggesting that the vascular system is adversely affected in patients with unrepaired coarctation and that intervention is advisable for important coarctation at the time of diagnosis.
Timing of intervention
Identifying the ideal time for repair of aortic coarctation remains challenging, and indications for intervention were refined during the study period. Our patient population represented referral bias, including more adults than in most other series. We refer patients with severe coarctation for intervention when the diagnosis is made, regardless of age. Survival was not reduced until patients had reached adulthood (>20 years). Patients who had a repair of aortic coarctation at younger than 5 years had the greatest risk for reoperation. Patients who were older than 9 years of age at the time of repair had a greater prevalence of residual hypertension, suggesting that repair may be best performed between the ages of 5 and 9 years, unless a patient presents with severe symptoms. This is similar to a report from the Netherlands wherein children who were <10 years of age at the time of repair had the best survival (97% at 30 years post surgical repair) (14).
Our study spans over 60 years of operative history at the Mayo Clinic, and there have been many developments in diagnostic techniques, medical options, and interventional procedures. These factors must be considered when interpreting our results. In addition, we included only patients who underwent surgical coarctation repair. Recent American Heart Association guidelines on percutaneous management of congenital cardiac disorders included aortic coarctation or recoarctation, irrespective of patient age, as a Class I indication for catheter-based interventions (17). While we also perform catheter-based interventions in patients with aortic coarctation and recoarctation, these patients were not included in this series.
Due to the observational nature of this study, follow-up interval and availability of blood pressure data varied. We were able to identify the presence or absence of late hypertension in two-thirds of our patients, but we were unable to determine other important factors, including the type and number of medical therapies that these patients with late hypertension required. We utilized direct blood pressure measurements when available, although we also used a self-report of hypertension. Previous work from the Mayo Clinic (18) reported in 25,415 patients that the sensitivity of a patient-reported diagnosis of high blood pressure was 73.2% (95% CI: 72.1% to 74.2%) and that the specificity was 94.6% (95% CI: 94.3% to 94.9%).
As well, due to the retrospective nature of this study, we may have not identified all re-intervention events, and as such, our estimates, while consistent with previously published rates, including those from our institution (6,13), may be low. We are also unable to comment on the prevalence of coronary artery disease, as we considered only those patients who required re-intervention such as angioplasty or coronary artery bypass grafting. This likely resulted in an underestimation of coronary artery disease in this population.
Surgical repair of aortic coarctation can be performed with minimal mortality. Long-term outcomes demonstrate that surgery is not a cure; patients have high rates of hypertension, need for re-intervention, and decreased survival. These results stress the importance of lifelong informed follow-up and aggressive identification and treatment of sequelae in patients with aortic coarctation.
Dr. Oliver has received a scientific grant from GTC Biotherapeutics. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- atrial septal defect
- New York Heart Association
- patent foramen ovale
- Received February 14, 2013.
- Revision received May 20, 2013.
- Accepted June 11, 2013.
- American College of Cardiology Foundation
- Warnes C.A.,
- Williams R.G.,
- Bashore T.M.,
- et al.
- Hiratzka L.F.,
- Bakris G.L.,
- Beckman J.A.,
- et al.
- Abbott M.E.
- Crafoord C.,
- Nylin G.
- Cohen M.,
- Fuster V.,
- Steele P.M.,
- Driscoll D.,
- McGoon D.C.
- Roifman I.,
- Therrien J.,
- Ionescu-Ittu R.,
- et al.
- Gardiner H.M.,
- Celermajer D.S.,
- Sorensen K.E.,
- et al.
- Ong C.M.,
- Canter C.E.,
- Gutierrez F.R.,
- Sekarski D.R.,
- Goldring D.R.
- Koller M.,
- Rothlin M.,
- Senning A.
- Brouwer R.M.,
- Erasmus M.E.,
- Ebels T.,
- Eijelaar A.
- Sakurai T.,
- Stickley J.,
- Stümper O.,
- et al.
- Feltes T.F.,
- Bacha E.,
- Beekman R.H. 3rd.,
- et al.