Patient Population
A total of 85 patients were analyzed. 44 (52%) patients received preoperative evaluation with echocardiogram and CTA, while 41 (48%) of patients received an echocardiogram only. Demographic data are presented in Table 1. For all patients who underwent CTA, the median effective radiation dose was 0.23 (IQR 0.20, 0.31) Sv. The median time from surgery to record review was 8.8 (5.9, 11.3) years.
Table 1: Demographic comparison of CTA + echo patients versus echo only patients.
|
CTA + Echo
(N = 44)
|
Echo Only
(N = 41)
|
P-value
|
Birth weight (kg)
|
3.1 (2.8, 3.6)
|
3.0 (2.8, 3.8)
|
0.81
|
Age at repair (days)
|
11 (7.3, 42)
|
14 (9.5, 26.5)
|
0.73
|
Weight at repair (kg)
|
3.4 (3.1, 3.9)
|
3.1 (2.8, 3.9)
|
0.32
|
Height at repair (cm)
|
50 (47.3, 54)
|
50 (48, 52)
|
0.45
|
Male, n (%)
|
27 (61%)
|
31 (76%)
|
0.17
|
Term, n (%)
|
40 (91%)
|
37 (90%)
|
1.00
|
Genetic syndrome, n (%)
|
7 (15%)
|
5 (12%)
|
076
|
Prenatal diagnosis, n (%)
|
17 (39%)
|
8 (20%)
|
0.05
|
Data reported as median (interquartile range) or count (percentage). CTA = computed tomographic angiography.
Comparison of Echocardiogram Measurements Between CTA + Echo and Echo Only Groups
Echocardiographic measurements between the two study groups were compared (Table 2). Measurements of the transverse arch and isthmus were similar between groups. However, the CTA + echo group had a smaller mitral valve diameter and Z-score (p = 0.01) and smaller median ascending aorta measurements and Z-score (p = 0.01) than the echo only group. All echocardiographic and CTA measurements showed an intra-observer variability intraclass correlation coefficient > 0.85 and an inter-observer variability intraclass correlation coefficient > 0.80, showing strong reproducibility of the measurements.
Table 2: Echocardiographic measurement comparison of CTA + echo patients versus echo only patients.
Echo Measurements
|
CTA + Echo
|
Echo Only
|
P-value
|
Aortic Isthmus (mm)
|
2.6 (2.3, 2.9)
|
2.7 (2.5, 3.1)
|
0.17
|
Aortic Isthmus Z-score
|
-2.9 (-3.1, -2.4)
|
-2.7 (-3.0, -2.1)
|
0.32
|
Isthmus: PDA ratio
|
0.7 (0.5, 1.0)
|
1.0 (0.8, 1.4)
|
<0.01
|
Distal Transverse Arch (mm)
|
3.7 (3.3, 4.6)
|
4.0 (3.6, 4.7)
|
0.34
|
Distal Transverse Z-score
|
-2.6 (-3.2, -2.1)
|
-2.5 (-3.0, -1.9)
|
0.68
|
Proximal Transverse Arch (mm)
|
4.5 (3.9, 5.2)
|
4.4 (3.5, 5.4)
|
0.97
|
Ascending Aorta (mm)
|
7.5 (6.6, 8.4)
|
8.2 (7.4, 8.9)
|
0.01
|
Ascending Aorta Z-score
|
-0.6 (-1.5, -0.2)
|
-0.1 (-0.9, 0.4)
|
<0.01
|
Aortic Valve Diameter (mm)
|
5.5 (5.0, 6.7)
|
6.2 (5.3, 6.9)
|
0.21
|
Aortic Valve Diameter Z-score
|
-1.7 (-2.7, -0.6)
|
-1.4 (-2.2, -0.8)
|
0.37
|
Isthmus Flow Velocity (m/s)
|
1.9 (1.5, 2.6)
|
2.3 (1.4, 2.9)
|
0.52
|
Mitral Valve Diameter (mm)
|
8.7 (7.5, 10.0)
|
9.7 (8.7, 10.0)
|
0.01
|
Mitral Valve Z-score
|
-1.5 (-2.1, -0.4)
|
-0.7 (-1.4, -0.1)
|
<0.01
|
LVEDV (mL)
|
7.4 (5.8, 12.7)
|
8.7 (6.8, 12.4)
|
0.24
|
LVEDV Z-score
|
-0.5 (-2.0, 1.5)
|
0.6 (-0.7, 1.8)
|
0.12
|
Data reported as median (interquartile range). CTA = computed tomographic angiography. LVEDV = left ventricular end-diastolic volume.
PDA = patent ductus arteriosus.
Agreement of CTA and Echocardiographic Measurements in the CTA + Echo Group
Echocardiographic measurements were compared to CTA measurements within the CTA + echo group. Results are presented utilizing the Bland-Altman plots in Figure 2. Measurements were on average larger on CTA when compared to echocardiogram measurements in the same patient, including aortic valve annulus (mean difference = -1.3 mm, p < 0.01), ascending aorta (mean difference = -0.6 mm, p < 0.01), proximal transverse aortic arch (mean difference = -1.0 mm, p < 0.01), distal transverse aortic arch (mean difference = -0.6mm, p < 0.01), and aortic isthmus (mean difference = -1.0 mm, p < 0.01).
Predictors of Surgical Approach
In the CTA + echo group, 10 (23%) patients underwent sternotomy versus 6 (15%) in the echo only group, p = 0.41. When comparing the thoracotomy (n = 69) versus sternotomy (n = 16) groups, there was no statistically significant differences in sex, birth weight, age at repair, weight at repair, height at repair, genetic disorder, or presence of prenatal diagnosis. Echo measurements showed no statistically significant differences between the thoracotomy and sternotomy groups, though, patients in the sternotomy group trended toward having a smaller distal transverse arch Z-score [-3.1 (-3.9, -2.1) versus -2.4 (-2.9, -1.8), p = 0.10] and aortic isthmus Z-score [-3.0 (-3.3, -2.6) versus -2.6 (-3.0, -2.6), p = 0.14] compared to the thoracotomy group. In patients with CTA measurements (n = 44), patients in the sternotomy group trended toward having a smaller distal transverse arch Z-score compared to the thoracotomy group [-2.3 (-2.9, -2.1) versus -2.1 (-2.5, -2.1), p = 0.12]. Upon multivariable logistic regression analysis, no statistically significant predictors of surgical approach were identified, including presence of pre-operative CTA.
Predictors of Composite Outcome – Presence of Residual Defect and/or Need for Re-intervention
In the CTA + echo group, 13/44 (30%) patient reached the composite outcome of need for re-intervention or residual arch/isthmus gradient > 16 mmHg on echocardiogram versus 17/41 (42%) of patients in the echo only group, p = 0.25. In the whole group, demographic and echocardiographic differences in patients who reached the composite outcome (n = 30) versus those who did not (n = 55) are reported in Table 3. In patients who had a pre-operative CTA (n = 44), patients in the composite outcome group had smaller CTA-based aortic valve annulus Z-score [-1.0 (-1.7, -0.5) versus 0.2 (-0.7, 1.4)] and aortic root Z-score
[-1.6 (-2.0, -1.2) versus -0.5 (-1.4, 0.6)] compared to those patients who did not reach the composite outcome, p < 0.01 for both. There were no differences in CTA measurements of the ascending aorta (p = 0.25), distal transverse arch (p = 0.95), or aortic isthmus (p = 0.72) Z-scores between groups.
In the whole cohort that included demographic, operative, and echo measurements only, multivariable logistic regression (Nagelkerke R2 = 0.08) showed the only independent predictor of reaching the composite outcome was LV EDV Z-score (OR 0.75, p = 0.02). Surgical approach trended toward significance (p = 0.07) and remained in the final model. No echo measures of aortic size were predictive of reaching the composite outcome.
In patients who had a pre-operative CTA, multivariable logistic regression (Nagelkerke R2 = 0.52) showed smaller CTA based aortic annulus Z-score (OR 0.34, p < 0.01) and thoracotomy (OR 30.6, p = 0.02) were independently associated with reaching the composite outcome. Upon ROC analysis, CTA based aortic annulus Z-score displayed an area under the curve of 0.80, p < 0.01 in predicting the composite outcome. A pre-operative aortic valve Z-score < -1.0 had 87% sensitivity and 62% specificity in identifying patients who reached the composite outcome.
Table 3: Demographic and echocardiographic measurement comparison of patients who reached the composite outcome versus those who did not.
|
No Gradient/
Re-intervention
(n=55)
|
Gradient of >16 mmHg
Or Re-intervention
(n=30)
|
P-value
|
Birth Weight (kg)
|
3.1 (2.9, 3.6)
|
3.0 (2.8, 3.6)
|
0.41
|
Age at Repair (days)
|
14 (9, 42)
|
11 (7, 21)
|
0.16
|
Weight at Repair (kg)
|
3.4 (3.0, 4.1)
|
3.2 (3.0, 3.6)
|
0.14
|
Height at repair (cm)
|
50 (48, 54)
|
49 (48, 51)
|
0.18
|
Male
|
38 (66%)
|
17 (63%)
|
0.82
|
Term
|
51 (93%)
|
26 (87%)
|
0.36
|
Genetic Disorder
|
9 (16%)
|
3 (10%)
|
0.53
|
Fetal Diagnosis
|
38 (69%)
|
22 (73%)
|
0.68
|
Abnormal Aortic Valve
|
23 (42%)
|
15 (50%)
|
0.47
|
Abnormal Arch Branching
|
37 (67%)
|
21 (70%)
|
0.80
|
Echo Data
|
|
|
|
Isthmus Z-score
|
-2.7 (-3.1, -2.3)
|
-2.7 (-3.0, -2.4)
|
0.82
|
Isthmus (mm)
|
2.7 (2.4, 3.1)
|
2.7 (2.3, 2.9)
|
0.64
|
Isthmus: Duct
|
0.9 (0.6, 1.1)
|
0.9 (0.8, 1.4)
|
0.48
|
Distal Transverse Arch Z-score
|
-2.4 (-3.0, -2.0)
|
-2.8 (-3.2, -2.2)
|
0.08
|
Distal Transverse Arch (mm)
|
4.0 (3.6, 4.8)
|
3.7 (3.3, 4.2)
|
0.08
|
Proximal Transverse Arch (mm)
|
4.7 (3.9, 5.7)
|
4.1 (3.4, 4.7)
|
0.07
|
Ascending Aorta Z-score
|
-0.4 (-1.2, 0.1)
|
-0.6 (-1.3, 0.0)
|
0.54
|
Ascending Aorta (mm)
|
8.2 (7.0, 9.0)
|
7.6 (6.9, 8.2)
|
0.04
|
Aortic Valve Diameter Z-score
|
-1.5 (-2.2, -0.3)
|
-2.9 (-3.1, -1.1)
|
0.35
|
Aortic Valve Diameter (mm)
|
6.2 (5.4, 7.0)
|
5.8 (4.9, 6.2)
|
0.04
|
Isthmus Flow Velocity (m/s)
|
2.1 (1.4, 2.9)
|
1.8 (1.4, 2.7)
|
0.65
|
Mitral Valve Z-score
|
-1.1 (-1.9, -0.2)
|
-1.2 (-2.0, -0.5)
|
0.41
|
LVEDV Z-score
|
0.6 (-0.8, 1.8)
|
-0.6 (-2.6, 1.5)
|
0.06
|
Data reported as median (interquartile range) or count (percentage). LVEDV = left ventricular end-diastolic volume.