We identified a cohort of 18 patients with COA requiring surgical intervention as neonates and matched them with 18 control patients who met inclusion and exclusion criteria listed above. See Table 1 for patient group characteristics. There were no significant differences in gender or ethnicity. Of the control group, 9 patients (50%) had some degree of anomalous pulmonary venous return, 4 (22%) had neonatal arrhythmias, 2 (11%) had valvar abnormalities and 3 others had cardiomyopathy, pulmonary hypertension and a pericardial tumor (representing 6% each).
Patient Cohort Characteristics.
|
Control
|
Coarctation
|
Sex
|
Male
|
11
|
61%
|
9
|
50%
|
Female
|
7
|
39%
|
9
|
50%
|
Race/Ethnicity
|
African American
|
8
|
44%
|
5
|
28%
|
White
|
8
|
44%
|
13
|
72%
|
Hispanic
|
2
|
11%
|
3
|
17%
|
Other/Unanswered
|
2
|
11%
|
0
|
0%
|
Diagnoses of Control Group
|
Pulmonary Vein
|
9
|
50%
|
|
|
Arrhythmia
|
4
|
22%
|
|
|
Valvar Abnormality
|
2
|
11%
|
|
|
Cardiomyopathy
|
1
|
6%
|
|
|
Pulmonary Hypertension
|
1
|
6%
|
|
|
Tumor
|
1
|
6%
|
|
|
Table 1. Here we report the demographic information of each cohort and report the primary diagnosis resulting neonatal admission of the control group to the cardiac intensive care unit.
Preoperative COA vs unaffected controls:
There was a near linear association between the average rate of rise and fall compared to heart rate (Supplemental Figure). Presurgical waveform characteristics between COA and control patients were calculated and shown in Table 2. The two characteristics that were significant (p-value < 0.01) were the MRF in the lower extremity and the difference between the MRF of the upper extremity minus the lower extremity (∆MRF on Table 2).
Preoperative Waveform Analysis for Upper and Lower Extremities and Their Differences.
Preop
Feature
|
Control
|
Coarctation
|
|
Median [Q1, Q3]
|
p value
|
ARR_u
|
4.66 [4.38, 5.07]
|
4.52 [4.41, 4.88]
|
0.16
|
ARF_u
|
-1.72 [-1.87, -1.53]
|
-1.67 [-1.75, -1.53]
|
0.96
|
MRR_u
|
5.05 [4.71, 5.46]
|
4.98 [4.76, 5.26]
|
0.24
|
MRF_u
|
-2.63 [-2.81, -2.47]
|
-2.64 [-2.73, -2.51]
|
0.6
|
ARR_l
|
4.9 [4.42, 5.32]
|
4.9 [4.47, 5.31]
|
0.91
|
ARF_l
|
-1.53 [-1.65, -1.41]
|
-1.49 [-1.59, -1.35]
|
0.38
|
MRR_l
|
5.27 [4.77, 5.68]
|
5.32 [4.87, 5.7]
|
0.84
|
MRF_l**
|
-2.6 [-2.67, -2.52]
|
-2.34 [-2.5, -2.3]
|
0.009
|
∆ARR
|
-0.22 [-0.52, -0.13]
|
-0.12 [-0.45, 0.02]
|
0.5
|
∆ARF
|
-0.12 [-0.31, -0.02]
|
-0.2 [-0.26, -0.01]
|
0.22
|
∆MRR
|
-0.22 [-0.44, -0.11]
|
-0.12 [-0.46, 0.02]
|
0.54
|
∆MRF**
|
-0.07 [-0.2, 0.08]
|
-0.24 [-0.38, -0.13]
|
0.001
|
Table 2. Units for slope are normalized amplitude / time and values are reported as the median, 1st and 3rd quartiles of PPG features. Subscripts indicate upper (u) vs lower (l) extremity acquisition site. The ∆ indicates the difference between the upper and lower extremity for that feature. Features with statistical significance between the 2 groups are marked as ** - p value <0.01. Abbreviations: ARR average rate of rise, ARF average rate of fall, MRR maximum rate of rise, MRF maximum rate of fall, COA coarctation of the aorta, Q1 1st quartile, Q3 3rd quartile.
Postoperative COA vs unaffected controls:
Of the 18 patients with COA, one patient did not have upper extremity PPG data recorded post operatively, therefore they were withheld from the post op statistical analysis. Table 3 shows the same waveform characteristics evaluated postoperatively. Here we note statistical significance in the MRF in both the upper and lower extremities but not in the ∆MRF indicating that the difference normalized after surgical correction of the COA (see Figure 4).
Postoperative Waveform Analysis for Upper and Lower Extremities and Their Differences.
Postop
Feature
|
Control
|
Coarctation
|
|
Median [Q1, Q3]
|
p value
|
ARR_u
|
4.66 [4.38, 5.07]
|
4.68 [4.44, 4.86]
|
0.46
|
ARF_u
|
-1.72 [-1.87, -1.53]
|
-1.65 [-1.79, -1.54]
|
0.59
|
MRR_u
|
5.05 [4.71, 5.46]
|
5.04 [4.84, 5.36]
|
0.44
|
MRF_u**
|
-2.63 [-2.81, -2.47]
|
-2.47 [-2.56, -2.26]
|
0.046
|
ARR_l
|
4.9 [4.42, 5.32]
|
4.75 [4.56, 5.21]
|
0.79
|
ARF_l
|
-1.53 [-1.65, -1.41]
|
-1.51 [-1.62, -1.37]
|
0.4
|
MRR_l
|
5.27 [4.77, 5.68]
|
5.12 [4.92, 5.58]
|
0.62
|
MRF_l**
|
-2.6 [-2.67, -2.52]
|
-2.27 [-2.46, -2.23]
|
0.008
|
∆ARR
|
-0.22 [-0.52, -0.13]
|
-0.04 [-0.32, 0.13]
|
0.66
|
∆ARF
|
-0.12 [-0.31, -0.02]
|
-0.07 [-0.16, 0.02]
|
0.99
|
∆MRR
|
-0.22 [-0.44, -0.11]
|
-0.05 [-0.32, 0.17]
|
0.61
|
∆MRF
|
-0.07 [-0.2, 0.08]
|
-0.04 [-0.29, -0.02]
|
0.93
|
Table 3. This table compares values for COA s/p surgical repair and control subjects. Units for slope are normalized amplitude / time and values are reported as the median, 1st and 3rd quartiles of PPG features. Subscripts indicate upper (u) vs lower (l) extremity acquisition site. The ∆ indicates the difference between the upper and lower extremity for that feature. The two-sided Wilcoxon rank sum test was used to test for significance. Features with statistical significance between the 2 groups are marked as ** - p value <0.05. Abbreviations: ARR average rate of rise, ARF average rate of fall, MRR maximum rate of rise, MRF maximum rate of fall, COA coarctation of the aorta, Q1 1st quartile, Q3 3rd quartile.
Preoperative COA vs Postoperative COA:
Preoperative vs Postoperative Waveform Characteristics for Differences Between Upper and Lower Extremities.
Waveform Features
|
Preoperative
|
Postoperative
|
|
Median [Q1, Q3]
|
p value
|
∆ARR
|
-1.2 [-4.5, 0.19]
|
-0.16 [-3.3, 2.6]
|
0.13
|
∆ARF
|
-2 [-2.6, -0.085]
|
-0.63 [-1.7, 0.34]
|
0.2
|
∆MRR
|
-1.2 [-4.6, 0.24]
|
-0.14 [-3.3, 3]
|
0.16
|
∆MRF**
|
-2.4 [-3.8, 1.3]
|
-0.3 [-2.9, 0.71]
|
0.028
|
Table 4. Reported as the median, 1st and 3rd quartiles of PPG features evaluated for COA before and after surgical repair. Subscripts indicate upper (u) vs lower (l) extremity acquisition site. The ∆ indicates the difference between the upper and lower extremity for that feature. The two-sided Wilcoxon rank sum test was used to test for significance. Features with statistical significance between the 2 groups are marked as ** - p value <0.05. Abbreviations: ARR average rate of rise, ARF average rate of fall, MRR maximum rate of rise, MRF maximum rate of fall, COA coarctation of the aorta, Q1 1st quartile, Q3 3rd quartile.
Leave One Out Cross Validation:
In order to test the validity of our algorithm to identify and appropriately characterize patients with COA we implemented LOOCV stratified by patient. The MRF in the upper and lower extremity are used to classify subjects in the COA or control groups. An LDA classifier achieved an average accuracy of 72%. The receiver operating characteristic curve (ROC) is given in Figure 6. The area under the curve (AUROC) is 0.78. The suggested operating point is indicated by a red circle with a sensitivity of 0.61 and a specificity of 0.94.