1. General information of the pediatric patients:
Among the 60 cases, there were 29 boys and 31 girls, with an average age of 3.74 ± 3.51 years. Pathogen analysis: viral infections were present in 44 cases (42 cases of EBV, 1 case of cytomegalovirus, and 1 case of measles), bacterial infections in 12 cases, Leishmania infection in 2 cases, and Mycoplasma pneumoniae infection in 2 cases (see Figure 1).
Our follow-up time was 8 weeks, observing survival rates at 8 weeks. Of the 60 cases included, 53 survived at 8 weeks, and 7 died, with a survival rate of 88.3%.
2. Comparison of two groups:
2.1 General clinical data comparison:
Among the included cases, there were 19 critically ill patients (organ dysfunction details see Figure 2) and 41 non-critically ill patients. In the critically ill group, there were 7 males and 12 females, with a median age of 1.00 (4.00) years; in the non-critically ill group, there were 22 males and 19 females, with a median age of 3.00 (5.00) years. There was no statistically significant difference in gender between the two groups (χ2=1.470, P=0.225); the age of critically ill patients was younger than that of non-critically ill patients, with a statistically significant difference (Z=-2.635, P=0.008).
There were no significant differences in total fever duration, hospital stay, and time of diagnosis between the two groups; there was no significant difference in the rate of hepatosplenomegaly between the two groups (P>0.05). See Table 2.
Table 2. Comparison of general clinical data
General Clinical Data
|
Critical Group (n=19)
N(%)
|
Non-Critical Group (n=41)
N(%)
|
Z/χ2/t/
|
P -value
|
Gender
Male
Female
|
7(36.8)
12(50)
|
22(53.7)
19(52.3)
|
1.470
|
0.225
|
Age (year)
|
1.00(4.00)
|
3.00(5.00)
|
-2.635
|
0.008
|
Duration of fever (d)
|
18(12)
|
11(10)
|
-1.290
|
0.197
|
Time to Diagnosis (d)
|
2(5)
|
3(4)
|
-1.611
|
0.107
|
Time to Diagnosis (d)
Hepatomegaly
Splenomegaly
|
20(15)
15(78.9)
13(68.4)
|
18(9)
30(73.2)
34(82.9)
|
-0.769
0.026
0.868
|
0.445
0.873
0.351
|
2.2 Comparison of laboratory data between two groups:
The critically ill group showed significantly higher levels of ALT, AST, LDH, GGT, DD, and TG, as well as a more significant decrease in ALB compared to the non-critically ill group (P<0.05). The differences between the critically ill and non-critically ill groups were mainly manifested in liver function, DD, and TG. There were no significant differences between the two groups in blood routine, creatinine, ferritin, CRP, PCT, total lymphocyte count, cytokines, and sCD25. See Table 3.
Table 3. Comparison of functional indexes of various organs
Laboratory Data
|
Critical Group (n=19)
M(Q)
|
Non-Critical Group (n=41)
M(Q)
|
Z/χ2/t
|
P-value
|
WBC(×109/L)
NEU(×109/L)
PLT(×109/L)
HGB(×1012/L) ★
|
2.80(3.30)
1.02(1.57)
74(133)
94.63±13.463
|
2.70(2.80)
1.06(1.16)
64(55)
94.49±19.293
|
-0.135
-0.127
-0.533
0.950
|
0.893
0.899
0.594
0.346
|
CRP(g/L)
|
15.69(26.01)
|
23.22(28.94)
|
-0.493
|
0.622
|
PCT(ng/mL)
|
0.576(2.221)
|
0.84(3.180)
|
-0.251
|
0.802
|
FERR(μg/L)
|
6986(29515)
|
3057(8683)
|
-1.390
|
0.164
|
ALT(U/L)
AST(U/L)
LDH(U/L)
TB(μ mol/L)
GGT(U/L)
ALB (g/L) ★
APTT(Sec) ★
PT(Sec) ★
DD(mg/L)
FIB(g/L)
HDL(mmol/L)
TG(mmol/L)
Cr(μ mol/L)
CD4+(%)★
CD8+(%)★
CD56+(%)
IL-10(p g /ml)
IL-6(p g /ml)
TNF-α(p g/ml)
IFN-γ(p g/ml)
sCD25(p g /ml)
|
365(500)
471(903)
1263(2316)
12.70(42.50)
315(372)
29.93±3.99
36.04±9.70
13.05±2.62
9.96(14.41)
1.41(0.81)
0.52(0.56)
2.77(1.58)
32(18)
30.94±16.27
38.90±28.39
3.1(6.89)
57.83(276.97)
12.66(33.68)
5.23(4.52)
39.98(750.10)
20871(26431)
|
90(229)
112(297)
764(698)
10.50(11.80)
35(140)
35.02±6.10
34.72±9.70
12.50±1.92
3.85(8.57)
1.56(1.38)
0.36(0.45)
2.28(1.80)
27(17)
34.66±16.28
39.49±21.46
4.99(4.22)
39.53(182)
21.05(29.63)
4.18(6.18)
28.78(7103.38)
16294(26913)
|
-3.004
-3.750
-2.439
-1.478
-3.934
-3.282
-0.534
-0.924
-2.106
-1.542
-0139
-2.106
-1.177
0.966
0.106
-0.168
-0.153
-0.714
-1.376
-0.866
-1.644
|
0.003
0.000
0.015
0.139
0.000
0.001
0.602
0.359
0.035
0.123
0.890
0.035
1.239
0.339
0.916
0.867
0.878
0.475
0.169
0.386
0.100
|
★Mean (Standard Deviation)
2.3. Multifactor logistic regression analysis:
In this part of the study, multifactor logistic regression analysis was conducted on age, TG, and ALB, which showed significant differences between the critically ill group and the non-critically ill group. It was found that the younger the age and the lower the ALB, the higher the risk of MODS occurrence in children with infection-related HLH. Age and ALB were identified as independent risk factors for critically ill children. See Table 4.
Table 4. Multivariate Logistic regression analysis
Variable
|
b Value
|
Standard Error of B
|
Wald Chi-Square Value
|
P-value
|
OR Value
|
95%CI of OR Value
|
ALB
|
-0.216
|
0.084
|
6.714
|
0.010
|
0.805
|
0.684-0.949
|
TG
|
0.321
|
0.197
|
2.644
|
0.104
|
1.379
|
0.936-2.030
|
Age
|
-0.289
|
0.141
|
4.174
|
0.041
|
0.749
|
0.568-0.988
|
2.4. ROC Curves for Age and ALB Predicting Children with Infection-Related HLH Developing Critical HLH:
As shown in Table 5 and Figure 3, the area under the ROC curve for ALB predicting critically ill children with infection-related HLH is 0.765 (95% CI: 0.643-0.888, P=0.011), with an optimal cutoff value of 32.50 g/L (sensitivity=68.3%, specificity=84.2%); the area under the ROC curve for age predicting critically ill children with infection-related HLH is 0.711 (95% CI: 0.570-0.851, P=0.009), with an optimal cutoff value of 1.50 years (sensitivity=70.7%, specificity=68.4%).
Table 5. Optimal cut-off of ROC curve
Parameters
|
AUC
|
P-value
|
95%CI
|
Youden index
|
Sensitivity
|
Sensitivity
|
Optimal cut-off value
|
ALB
Age
|
0.765
0.711
|
0.001
0.009
|
0.643-0.888
0.570-0.851
|
0.525
0.392
|
0.683
0.707
|
0.842
0.684
|
32.50
1.50
|