Neonatal coarctation of the aorta (CoA) is often associated with hypoplastic left-sided cardiac structures. Limited data are available regarding the long-term clinical outcomes. Our purposes were to assess the following in neonates with CoA and left-sided cardiac hypoplasia: (1) left-sided cardiac structure growth over time, (2) echocardiographic parameters predicting reintervention, and (3) long-term outcomes and reintervention rates. Data were collected for all patients born with isolated CoA, along with a mitral or aortic Z-score of <−2, who underwent repair before 2 months of age from January 1993 to April 1997. Recent follow-up data were available for 51 of 63 patients (81%) aged 15.4 ± 1.5 years (range 11 to 18). Of the 51 patients, 46 (90%) had a recent echocardiogram, all with normal left ventricular systolic function. The mitral and aortic valve annulus Z-scores increased significantly from the initial measurements: −3.29 ± 1.54 to −0.94 ± 0.58 (p <0.0001) and −3.25 ± 1.98 to 0.08 ± 1.81 (p <0.0001). No significant change was seen from intermediate (6.1 ± 1.6 years) to the latest follow-up examination. Of the 51 patients, 12 (24%) required reintervention at 3 ± 4.5 years; 9 for repeat CoA, 4 for subaortic stenosis, and 2 for aortic stenosis (some in combination). The initial echocardiographic parameters were not associated with reintervention. The overall freedom from reintervention was 76% at 15 years of follow-up. In conclusion, the long-term outcomes after isolated CoA repair with associated left-sided cardiac hypoplasia were excellent. Both aortic and mitral valve sizes had increased substantially by intermediate follow-up but tended to normalize afterward. Although 24% of the patients required reintervention, significant left ventricular inflow or outflow tract obstruction was uncommon.
Hypoplasia of the structures on the left side of the heart is common in the setting of coarctation of the aorta (CoA). Determining the adequacy of the left side of the heart in this setting is critical if a 2-ventricular repair (preferred option) is to be performed. A staged single-ventricle repair using an initial Norwood procedure or cardiac transplantation might be recommended as alternatives to a 2-ventricle repair when the structures on the left side of the heart are deemed too small. Several investigators, including our group, have shown that one cannot extrapolate the scoring system for critical aortic stenosis to CoA. We have shown that hypoplastic structures on the left side of the heart increase in size after CoA repair in infancy, with a low incidence of reintervention at mid-term follow-up. Concerns remain, however, regarding the long-term outcomes of these patients. These concerns include the development of left ventricular (LV) inflow or outflow obstruction, recurrent CoA, and/or pulmonary hypertension related to chronic LV diastolic dysfunction and pulmonary venous hypertension. The purposes of the present study were to assess the growth of hypoplastic left-sided cardiac structures over time, to identify the echocardiographic parameters at initial presentation that are predictive of the future need for reintervention, and to evaluate the long-term clinical outcomes and reintervention rates after isolated, neonatal CoA repair.
Methods
The patients were recruited from the original cohort of the study by Tani et al. The following were the criteria for inclusion at presentation: (1) an echocardiographic diagnosis of CoA from January 1, 1993 to April 30, 1997, with surgical repair at the Primary Children’s Medical Center at <2 months of age, (2) patency of all 4 cardiac valves and a mitral or aortic valve Z-score of <−2, (3) normally related and concordantly connected arterial trunks, (4) absence of a true atrial septal defect or anomalous pulmonary venous connection, and (5) absence of significant valvar or subvalvar aortic stenosis (peak Doppler gradient <20 mm Hg). The CoA was repaired from a left thoracotomy with resection and extended end-to-end anastomosis in all patients.
The medical records were reviewed for current patient age, body surface area, date of initial surgery, follow-up duration, outcomes, need for any cardiac reintervention, and most recent echocardiogram. The institutional review board at the University of Utah and Primary Children’s Medical Center approved the present study.
The initial and mid-term echocardiographic data were taken from our earlier studies and have been previously reported. For the present study, the most recent echocardiogram was reviewed, and optimal images were selected for offline measurements using previously described techniques. The dimensions of the aortic valve annulus, mitral valve annulus (anteroposterior diameter) were measured from the parasternal long-axis view. M-mode measurements of the LV end-diastolic dimension and LV shortening fraction were obtained from the parasternal short-axis view. The tricuspid regurgitation gradient, LV ejection fraction (Simpson’s method), mitral valve inflow echo pulsed-wave Doppler and E/e′ ratio, and LV outflow tract echo pulsed-wave Doppler were measured from appropriate imaging windows. Calculation of the Z-scores was performed using the Boston Z-score data set.
Mitral and aortic valve growth analysis was performed by comparing the Z-scores obtained at long-term follow-up to their initial preoperative and intermediate follow-up values. Student t tests were used to evaluate the average values of the difference under validated normality assumptions; otherwise, signed rank tests were performed. Survival analysis was performed to study the timing of freedom from any catheter-based or surgical reintervention. The Kaplan-Meier product limit method was used to estimate the overall probability of freedom from the event. Cox proportional hazard regression analysis was used to determine whether the initial echocardiographic parameters were associated with the need for reintervention. The descriptive statistics of outcomes are presented as mean ± SD. Statistical significance was inferred when p <0.05. All analyses were conducted using SAS, version 9.2 (SAS, Cary, North Carolina).
Results
A total of 63 patients met the original inclusion criteria and underwent surgical repair of CoA at our institution. Of these 63 patients, 51 (81%) had recent clinical follow-up data (15.4 ± 1.5 years) available, and all were alive and well. Echocardiograms were available for review for 46 of the 51 patients (90%). The remaining 17 patients were lost to follow-up or did not have a recent follow-up echocardiogram.
The 46 patients with follow-up echocardiographic data all had a normal LV shortening fraction (39 ± 5%) and ejection fraction (67 ± 5%), and all but 1 patient had an LV end-diastolic dimension Z-score >−2 (−0.24 ± 1.10). The mitral and aortic valve Z-scores ( Figures 1 and 2 ) increased significantly from preoperatively to long-term follow-up: −3.3 ± 1.5 to −0.9 ± 0.6 (p <0.0001) and −3.3 ± 2.0 to 0.1 ± 1.8 (p <0.0001), respectively. They did not, however, change significantly from intermediate (6.1 ± 1.6 years) to long-term follow-up for either the mitral valve (−0.6 ± 1.6 to −0.9 ± 0.6, p = 0.23) or the aortic valve (0.7 ± 1.6 to 0.1 ± 1.8; p = 0.28). At most, mild tricuspid regurgitation was present in 5 of 46 patients (11%), with none to trivial in the remainder. A regurgitant jet by Doppler was available in 11 patients (24%) and was 22 ± 6.9 mm Hg. No patient had indirect (ventricular septal flattening) evidence of pulmonary hypertension. The mitral valve E/e′ was available for 30 of 46 patients (65%; 8.0 ± 2.9, range 1.7 to 15.5), with only 1 patient having a value >12.
The initial echocardiographically derived mitral valve, aortic valve, and transverse arch diameter Z-scores were analyzed by regression analysis and were not found to correlate with the need for reintervention ( Table 1 ).
| Initial Mitral Valve Z-Score | Initial Aortic Valve Z-Score | Initial Transverse Arch Diameter Z-Score | |
|---|---|---|---|
| Reintervention (n = 12) | −2.6 ± 1.7 | −3.3 ± 2.7 | −2.4 ± 0.7 |
| No reintervention (n = 39) | −3.5 ± 1.4 | −3.2 ± 1.8 | −3.0 ± 1.2 |
| Hazard ratio | 1.56 | 0.80 | 2.16 |
| p Value | 0.07 | 0.23 | 0.11 |
Of the 51 patients with available follow-up data, 50 were asymptomatic (New York Heart Association functional class I). Class II symptoms of exercise intolerance were present in 1 patient, likely secondary to morbid obesity (body mass index 38.5 kg/m 2 ), given the lack of any significant echocardiographic abnormalities. All patients survived their initial CoA repair, and none required conversion to single ventricle physiology. Freedom from reintervention was 84% at 5 years, 82% at 10 years, and 78% at 15 years of follow-up ( Figure 3 ). A total of 12 patients (24%) required reintervention at 3 ± 4.5 years (range 0.3 to 14) after surgery. Of these 12 patients, 8 (16%) had recurrent CoA alone and 4 required treatment of multiple defects ( Table 2 ). Only 1 patient required reintervention between mid- and long-term follow-up: stent treatment of recurrent CoA >14 years postoperatively.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
