Patient characteristics
In total, 380 patients were diagnosed with KD upon admission to participating institutions during the study period. We first excluded the patients for whom there was no available TN-C data from either before or after their initial treatment (n = 103), patients who had a recurrent case (n = 12), those with a concurrent infection (n = 3), those with underlying congenital heart disease (n = 2), those who were not finally diagnosed with KD (n = 2), and those who did not receive IVIG treatment (n = 36) (Fig. 1). The remaining 222 patients were categorized into two groups on the basis of whether their Kobayashi score was ≤4 points (low-risk patients, n = 162) or ≥5 points (high-risk patients, n = 60) at the time of diagnosis (Table 1).
Table 1. Scoring systems that predict initially IVIG-resistant patients
Kobayashi Score (Cut off: ≥5 points;Sensitivity 76%、Specificity 80%)
|
Risk Factor
|
Points
|
Illness days at diagnosis
|
< 4 days
|
2
|
Serum sodium level
|
< 133 mmol/l
|
2
|
AST
|
≥ 100 IU/l
|
2
|
Neutrophil rate
|
≥ 80 %
|
2
|
CRP
|
≥ 10mg/dl
|
1
|
Platelet count
|
≤30.0×104/mm3
|
1
|
Age at diagnosis
|
≤12 month
|
1
|
Regarding the low-risk patients, we further excluded those whose treatment included added ulinastatin (n = 11) or steroids (n = 7). This left 144 enrolled patients who were administered IVIG as a first-line therapy. Among them, 116 (80.6%) patients responded to the IVIG-treatment and did not require a second-line therapy (low-risk initial treatment-responsive group), while 28 patients were resistant to the IVIG and did require a second-line therapy (low-risk initial treatment-resistant group). Regarding the high-risk patients, we excluded those who were not administered steroids (n = 23). This left 37 patients who were administered IVIG+steroid as a first-line therapy. Among them, 27 (73.0%) patients responded to the IVIG+steroid and did not require a second-line therapy (high-risk initial treatment-responsive group), while 10 patients did not respond to the IVIG+steroid and did require a second-line therapy (high-risk initial treatment-resistant group).
The baseline characteristics of the high-risk patients and low-risk patients are shown in Tables 2 and 3, respectively. In both the low- and high-risk groups, there were no significant differences in terms of age, sex, and laboratory data between the initial treatment-responsive group and the initial treatment-resistant group.
Table 2. Characteristics and data of high-risk patients in the initial treatment (IVIG+steroid)-responsive and initial treatment-resistant groups
|
IVIG +steroid responder group
|
IVIG +steroid resistant group
|
p-value
|
Number
|
27
|
10
|
|
Age (months)
|
42 [11 - 80]
|
44 [9 - 86]
|
0.880
|
Male gender, n (%)
|
17 (63.0)
|
8 (80.0)
|
0.285
|
Illness day at diagnosis
|
4 [2 - 8]
|
3 [2 - 7]
|
0.220
|
Kobayashi score
|
6 [5 - 10]
|
6 [5 - 10]
|
0.216
|
< Laboratory data before 1st line therapy >
|
TN-C, ng/mL
|
117.6 [35.0 – 324.8]
|
175.8 [80.4 – 380.9]
|
0.037
|
WBC, ×103/µL
|
14.8 [6.6 – 33.2]
|
18.6 [6.9 – 36.8]
|
0.242
|
Neutrophils, %
|
83 [60 - 95]
|
88 [68 - 94]
|
0.191
|
Platelets, ×104/mL
|
26.2 [13.1 – 59.4]
|
23.9 [13.5 – 36.6]
|
0.555
|
CRP, mg/dL
|
10.0 [2.5 – 24.0]
|
10.1 [5.2 – 21.7]
|
0.853
|
Albumin, g/dL
|
3.6 [2.7 – 4.1]
|
3.6 [2.8 – 4.4]
|
0.801
|
T-bilirubin, mg/dL
|
0.7 [0.3 – 5.5]
|
1.4 [0.5 – 4.6]
|
0.391
|
AST, IU/L
|
57 [20 – 787]
|
551 [25 – 2725]
|
0.013
|
ALT, IU/L
|
83 [8 – 937]
|
518 [9 – 1435]
|
0.067
|
Sodium, mEq/L
|
133 [127 – 137]
|
132 [128 – 135]
|
0.578
|
TN-C: tenascin-C, WBC: white blood cell, CRP: C-reactive protein, AST: aspartate aminotransferase, ALT: alanine aminotransferase.
* The Kobayashi score was ≥5 points in all cases.
* In all cases, the first-line therapy was IVIG, prednisolone, and aspirin.
Table 3. Characteristics and data of low-risk patients in the initial treatment (IVIG)-responsive and initial treatment-resistant groups
|
IVIG-responder group
|
IVIG-resistant group
|
p value
|
number
|
116
|
28
|
|
Age in month
|
29.5 [4 - 130]
|
22 [4 - 107]
|
0.352
|
Male gender, n (%)
|
56 (48.3)
|
13 (46.4)
|
0.861
|
Illness day at diagnosis
|
5 [2 - 10]
|
5 [3 - 8]
|
0.436
|
Kobayashi score
|
2 [0 - 4]
|
3 [0 - 4]
|
0.030
|
< Laboratory data before IVIG >
|
TN-C, ng/mL
|
106.6 [29.1 – 449.6]
|
113.5 [46.6 – 483.4]
|
0.432
|
WBC, ×103/µL
|
13.0 [6.1 – 32.3]
|
12.7 [6.3 – 22.0]
|
0.435
|
Neutrophil, %
|
66 [24 - 91]
|
67 [26 - 88]
|
0.418
|
Platelet, ×104/mL
|
33.3 [16.6 – 53.3]
|
32.9 [19.4 – 62.5]
|
0.612
|
CRP, mg/dL
|
6.7 [1.3 – 31.4]
|
8.8 [1.5 – 20.0]
|
0.190
|
Albumin, g/dL
|
3.6 [2.6 – 4.8]
n = 115
|
3.6 [2.6 – 4.3]
|
0.998
|
T-bilirubin, mg/dL
|
0.5 [0.1 – 3.1]
n = 114
|
0.6 [0.2 – 3.4]
|
0.064
|
AST, IU/L
|
34 [15 – 298]
|
33 [18 – 236]
|
0.677
|
ALT, IU/L
|
19 [5 – 442]
|
24 [8 – 237]
|
0.608
|
Sodium, mEq/L
|
136 [127 – 143]
|
135 [127 – 138]
|
0.093
|
TN-C: tenascin-C, WBC: white blood cell, CRP: C-reactive protein, AST: aspartate aminotransferase, ALT: alanine aminotransferase.
* The Kobayashi score was <5 points in all cases.
* In all cases, the first-line therapy was IVIG and aspirin.
Serum TN-C levels
First, we compared the serum TN-C levels on admission between the high-risk patients and the low-risk patients. The median level of TN-C for the high-risk patients was significantly higher than that of the low-risk patients (median: 121.6 [35.0–380.9] ng/ml vs. 110.2 [29.1–293.6] ng/ml, p = 0.028) (Fig. 2).
Among the high-risk patients, the median TN-C level on admission for the first-line treatment-resistant group (IVIG+ASA+steroid) was significantly higher than that of the first-line treatment-responsive group (median: 175.8 [80.4–380.9] ng/ml vs. 117.6 [35.0–324.8] ng/ml, p = 0.037) (Fig. 2a). After the first line-treatment was initiated, the level of TN-C was significantly reduced in the initial treatment-responsive group (median: 117.6 [35.0–324.8] ng/ml to 88.7 [23.8–263.3] ng/ml, p = 0.011), whereas no significant change in TN-C level was found for the initial treatment-resistant group (median: 175.8 [80.4–380.9] ng/ml to 166.1 [86.2–696.2] ng/ml, p = 0.878). Hence, the post-first treatment median TN-C level of the patients who required a second-line treatment was significantly higher than that of the patients who did not need additional treatment (median: 166.1 [86.2–696.2] ng/ml vs. 88.7 [23.8–263.3] ng/ml, p = 0.004).
Among the low-risk patients, no significant difference in the level of TN-C upon admission was found between the initial treatment-responsive group and the initial treatment (IVIG+ASA)-resistant group (median: 106.6 ng/ml [29.1–293.6] vs. 113.5 [46.6–277.4] ng/ml, p = 0.432) (Fig. 2b). As in the high-risk patients, the first-line treatment significantly reduced the TN-C level in the initial treatment-responsive group of low-risk patients (median: 106.6 [29.1–293.6] ng/ml to 81.1 [22.4–181.4] ng/ml, p < 0.001), whereas no significant change was found in the initial treatment-resistant group of low-risk patients (median: 113.5 [46.6–277.4] ng/ml to 107.3 [35.1–218.5] ng/ml, p = 0.212). Again, the post-first-line treatment TN-C levels of the initial treatment-resistant patients were significantly higher than those of the group who did not require additional treatment (median: 107.3 [35.1–218.5] ng/ml vs. 81.1 [22.4–181.4] ng/ml, p = 0.016).
Coronary artery lesions
Among the 37 high-risk patients, three (8.1%) had coronary aneurysms (z-score: ≥2.5). The serum TN-C level upon admission was not significantly different between the CAL-positive and CAL-negative groups of high-risk patients (119.5±33.0 ng/ml vs. 120.4±75.3 ng/ml, p = 0.291). Among the 144 low-risk patients, there were seven (4.9%) who had coronary aneurysms (z-score: ≥2.5). Similarly, the serum TN-C level upon admission in the low-risk patients was not significantly different between the CAL-positive and CAL-negative groups (150.4±67.5 ng/ml vs. 110.2±43.6 ng/ml, p = 0.835).