2.1 Comparison of general data between children with DMD and normal control group
A total of 287 children entered this study, except that 80 cases had not completed blood routine and muscle enzyme tests as required, 40 cases lost the follow up because they did not follow-up on time or change their contact information, 22 cases withdrew because of pulmonary infection and diarrhea during treatment or follow up, and the remaining 145 children completed the study.150 children in the control group were healthy children in the same period. The level of eosinophils was 0 at 0-0.1x109/l, and greater than or equal to 0.1x109/l was 1. The results of gene testing showed that 145 cases completed gene testing, including 92 cases of gene deletion (80 cases of large fragments and 12 cases of small fragments), 22 cases of gene duplication and variation (4 cases of single exon duplication, 16 cases of multi exon duplication) and 31 cases of point mutation (4 missense variants, 16 nonsense variants, 6 frameshift variants, 5 splice variants).The mean follow up time of DMD patients was 1.67(0.50,3.25)years.The comparison of age, blood routine and biochemical results between the two groups is shown in Table 1.
Table 1 Comparison of baseline data between patients with DMD and normal control group
|
Case group
N=145
|
Control group
N=150
|
Value
|
P
|
Age
|
3.83(1.42,6.08)
|
2.92(1.17,5.92)
|
-0.795
|
0.427
|
Total protein
|
66.18±4.13
|
68.66±4.41
|
4.820
|
0.000
|
Albumin
|
42.52±3.71
|
46.74±2.39
|
11.042
|
0.000
|
Globulin
|
23.67±3.87
|
21.93±3.02
|
4.193
|
0.000
|
Urea nitrogen
|
3.82±1.31
|
4.58±1.07
|
4.463
|
0.000
|
White blood cell
|
8.00(6.39,9.82)
|
7.55(6.40,9.41)
|
0.880
|
0.379
|
Neutrophils
|
3.43(2.52,4.36)
|
2.87(2.30,3.86)
|
2.723
|
0.006
|
Lymphocyte
|
3.34(2.69,4.03)
|
3.47(2.87,4.73)
|
-1.407
|
0.159
|
Monocyte
|
0.35(0.27,0.49)
|
0.41(0.32,0.59)
|
-2.896
|
0.004
|
Red blood cell
|
4.65±0.37
|
4.79±0.39
|
3.060
|
0.002
|
Hemoglobin
|
122.83±10.29
|
128.51±9.89
|
4.651
|
0.000
|
Eosinophils
|
0.16(0.09,0.29)
|
0.21(0.12,0.32)
|
2.163
|
0.031
|
Basophil
|
0.01(0.003,0.02)
|
0.01(0.002,0.03)
|
0.940
|
0.347
|
Platelet
|
311.24±92.23
|
308.32±65.48
|
0.297
|
0.767
|
Vitamin D3
|
47.91±11.48
|
82.60±23.12
|
6.439
|
0.000
|
Creatinine
Proportion of eosinophils
|
18.41±7.79
0.02(0.01,0.03)
|
35.25±10.18
0.03(0.02,0.05)
|
13.375
3.454
|
0.000
0.001
|
A total of 145 children with DMD completed this follow up study. Compared with the first visit, the eosinophil count, erythrocyte, hemoglobin and uric acid,etc, at the follow-up visit were statistically significant (P < 0.05 or 0.01).The comparison of baseline data between the initial visit and the follow-up visit is shown in Table 2.
Table 2 Comparison of data between the first visit and follow-up visit of children with DMD
|
At visit
N=145
|
Follow
Up
N=145
|
Value
|
P
|
Age
|
3.83(1.42,6.08)
|
5.25(3.16,7.16)
|
10.449
|
0.000
|
Total bilirubin
|
7.89(6.00,10.90)
|
7.92(6.30,10.49)
|
-0.484
|
0.629
|
Direct bilirubin
|
2.70(1.90,3.66)
|
2.33(1.60,3.20)
|
1.733
|
0.083
|
Indirect bilirubin
|
5.30(3.80,7.58)
|
5.70(4.20,7.30)
|
0.566
|
0.577
|
Total protein
|
66.18±4.13
|
66.24±4.29
|
0.425
|
0.672
|
Albumin
|
42.52±3.71
|
42.62±3.35
|
0.049
|
0.961
|
Globulin
|
23.67±3.87
|
23.62±2.97
|
0.398
|
0.692
|
Creatinine
|
18.41±7.79
|
23.41±9.17
|
2.933
|
0.006
|
Urea nitrogen
|
3.82±1.31
|
3.68±0.98
|
0.233
|
0.817
|
Uric acid
|
216.59±71.59
|
249.46±58.45
|
2.245
|
0.035
|
Total bile acid
|
4.40(3.00,8.07)
|
3.40(2.20,4.87)
|
-2.867
|
0.004
|
CK
|
4177.00(1698.70,9831.30)
|
7108.30(1993.2,11755.30)
|
2.074
|
0.038
|
CK-MB
|
188.8(109.76,298.00)
|
184.09(119.35,284.58)
|
1.441
|
0.149
|
LDH
|
1203.6(790.00,1322.00)
|
994.00(760.00,1262.60)
|
-1.039
|
0.229
|
Alanine aminotransferase
|
322.70±181.07
|
303.75±140.55
|
1.240
|
0.218
|
Aspartate aminotransferase
|
244.2(139.60,315.00)
|
195.90(146.90,292.70)
|
2.001
|
0.045
|
Eosinophils
|
0.16(0.09,0.29)
|
0.11(0.07,0.19)
|
2.251
|
0.032
|
Basophil
|
0.01(0.003,0.02)
|
0.01(0.01,0.02)
|
0.766
|
0.444
|
Red blood cell
|
4.65±0.37
|
4.85±0.42
|
4.577
|
0.000
|
Hemoglobin
|
122.83±10.29
|
129.99±12.50
|
5.521
|
0.000
|
CRP
|
0.75(0.18,1.21)
|
0.50(0.30,0.84)
|
1.521
|
0.128
|
Platelet
|
311.24±92.23
|
299.76±66.14
|
1.079
|
0.284
|
Vitamin D3
|
47.91±11.48
|
45.22±9.23
|
1.155
|
0.307
|
Vignos
scale
|
2(1,3)
|
2(1,3)
|
0.780
|
0.435
|
Neutrophils
|
3.43(2.52,4.36)
|
3.19(2.71,4.41)
|
0.895
|
0.371
|
Lymphocyte
|
3.34(2.69,4.03)
|
3.51(2.51,4.27)
|
0.330
|
0.741
|
Monocyte
|
0.35(0.27,0.49)
|
0.40(0.33,0.52)
|
-1.108
|
0.268
|
Proportion of eosinophils
|
0.02(0.01,0.03)
|
0.02(0.01,0.03)
|
1.105
|
0.269
|
2.3 The effect of eosinophils on muscle strength score
The correlation coefficient between eosinophils and vitamin D3 in the normal control group was ρ= 0.299,p=0.012. The correlation coefficient between eosinophils and vitamin D3 in DMD group was ρ= 0.563,p=0.001. The correlation coefficient between eosinophils and vitamin D3 was ρ= 0.267,p=0.207.
Correlation coefficient between eosinophils and CRP at visit ρ=-0.271,p=0.046,eosinophils and CRP at follow-up visit ρ=- 0.189,p=0.035.
Correlation coefficient between age and visit score was ρ=0.528, p =0.000, correlation coefficient between follow-up age and follow-up score was ρ=0.512, p=0.000,that was Statistically significant, The partial correlation analysis of controlling age found that the partial correlation coefficient between eosinophil count and visit score at visit was -0.155, p= 0.047, and the partial correlation analysis at follow-up visit was 0.082, p = 0.489.
Correlation coefficient between eosinophils to visit and score of visit was ρ=-0.245, p=0.040. Correlation coefficient with lymphocytes was ρ= 0.257, p=0.004, correlation coefficient with monocytes was ρ=-0.195, p =0.032. These were statistically significant.
Correlation coefficient between follow up visit eosinophils and fo-llow-up scores was ρ=-0.137, p=0.032. There was statistically signif-icant.Correlationcoefficient between eosinophils with lymphocytes at follow-up visit was ρ=0.038,p=0.716. Correlation coefficient betweeneosinophils and monocytes at follow-up visit was ρ=0.46,p=0.656.T-here was no statistical significance. Correlation coefficient between eosinophils and lymphocytes to the first visit was ρ=0.244, p=0.007,monocyte correlation coefficient was ρ=0.182, p=0.044.
2.4 Effects of different treatment methods on eosinophil count
There were 95 children in the prednisone treatment group. The number of eosinophils was 0.15 (0.08,0.25) at the time of treatment and 0.08 (0.05,0.17) at the time of follow-up. The comparison before and after treatment was Z=3.157,p=0.002. The correlation coefficient between the follow-up score and follow-up eosinophils was ρ=-0.092,p=0.573, and the correlation coefficient between visit score and eosinophils was ρ=-0.259,p=0.041.
There were 50 children in the conservative treatment group.The number of eosinophils in the conservative treatment group was 0.19 (0.10,0.31) at the time of visit and 0.20 (0.11,0.27) at the time of follow-up. The comparison before and after treatment was Z=0.270, p= 0.787. The correlation coefficient between the follow-up score and eosinophils was ρ=-0.233,p=0.004, and the correlation coefficient between visit score and eosinophils was ρ=-0.212,p=0.043. Comparison of eosinophils in the conservative treatment group and prednisone treatment group before treatment, Z=1.464, p=0.143, There was no statistical difference. There was statistical difference after treatment, Z=4.559, p=0.000.
2.5 Eosinophils affected muscle strength score of DMD
Single factor regression analysis of DMD muscle strength score revealed that the age of visit (OR=1.453,95% CI 1.180-1.789. p=0.000),creatinine (OR=0.865,95% CI 0.754-0.953. p=0.039),lymphocyte (OR =0.646,95%CI0.434-0.960 p=0.031),mononuclear(OR=0.046,95%CI0.003-0.849. p=0.038),eosinophils (OR=0.007,95% CI 0.001-0.276. p=0.008) were statistically significant. Age was a risk factor and others were protective factors for muscle strength score.
In the multivariate regression controlling for age and other factors, it was found that eosinophils (OR=0.038,95% CI 0.002-0.752. p=0.032) were protective factors of muscle strength score. The higher the eosinophils, the lower the score, the higher the muscle strength. The results of multivariate regression of influencing factors of muscle strength at the first visit are shown in Table 3.
Table 3 Multivariate regression analysis of muscle strength scores in children with DMD
|
B
|
SE
|
Wald
|
Exp B
|
95%CI
|
P
|
Age
|
0.962
|
0.331
|
8.474
|
2.619
|
1.369-5.003
|
0.004
|
Lymphocyte
|
-0.633
|
0.412
|
2.362
|
0.531
|
0.237 -1.190
|
0.124
|
Eosinophils
|
-3.324
|
1.551
|
4.590
|
0.038
|
0.002 -0.752
|
0.032
|
Constant
|
1.094
|
0.893
|
5.991
|
2.983
|
|
0.672
|
Using univariate regression of follow-up score we found the age of follow-up(OR=1.334,95%CI1.134-1.570,p=0.000),treatment(OR=0.404,95%CI0.178-0.918.p=0.030),lymphocytes(OR=0.630,95%CI0.439-0.904. p=0.012),monocytes (OR=0.015,95%CI 0.001-0.405.p=0.013),Visit score (OR=11.543,95% CI5.607-26.297.p=0.000).Following up eosin-ophils (OR=0.012,95% CI0.001-0.645.p=0.029),Eosinophils at visit (OR=0.016,95%CI0.002-0.586,p=0.024) were statistically significant.
Further control of age and other factors was discovered by multivariate regression,treatment(OR=0.167,95%CI0.030-0.931,p=0.041),Visit score (OR=13.582,95% CI 1.332-138.492. p=0.028),The effect of follow-up eosinophils (OR=0.033,95%CI 0.001-0.8 21,p=0.039) on muscle strength score was statistically significant.The results of multivari-ate regression of muscle strength influencingfactors during follow-up are shown in Table 4.
Table 4 Multivariate regression table of DMD follow-up muscle strength score
|
B
|
SE
|
Wald
|
Exp B
|
95%CI
|
P
|
Follow up age
|
1.042
|
0.626
|
2.768
|
2.834
|
0.831-9.668
|
0.096
|
Treatment
|
-1.788
|
0.876
|
4.167
|
0.167
|
0.030-0.931
|
0.041
|
Score at visit
|
2.609
|
1.185
|
4.849
|
13.582
|
1.332-138.492
|
0.028
|
Follow up eosinophils
|
-5.721
|
2.764
|
4.283
|
0.033
|
0.001-0.821
|
0.039
|
Constant
|
-3.738
|
1.583
|
5.577
|
0.024
|
|
0.018
|
2.6 Analysis of the effect of eosinophils on the efficacy of DMD
Using univariate Cox regression analysis we found that the age of visit (HR=0.851,95% CI 0.771-0.939,p=0.001), treatment (HR=3.362,95% CI1.222-5.607,p=0.000) and visit score (HR=0.637,95% CI 0.454-0.863. p=0.002),Lymphocytes(HR=1.056,95%CI1.003-1.111,p=0.037),Granulocytes(HR=0.785,95%CI0.457-0.954,p=0.048) follow up gra-nulocytes (HR=0.481,95% CI0.254-0.912. p=0.025), Albumin (HR=1.131,95% CI 1.060-1.206.p=0.002),Follow up eosinophils(HR=2.739,95%CI1.294-5.979. p=0.008) there were statistically significant.
After further analysis of other factors, multivariate regression we found age at visit (HR=0.827,95% CI0.701-0.974.p=0.023),visit sco-re (HR=0.354,95% CI0.465-0.942.p=0.008),treatment(HR=7.596,95% CI2.138-26.990,p=0.002),eosinophils (HR=1.127,95% CI 1.109-1.246,p=0.020), these were statistically significant. It suggests that the higherthe eosinophils, the better the curative effect. Multivariate Cox regression results affecting the efficacy of DMD in children are shown in Table 5.
Table 5 Cox multivariate regression table of effective treatment in children with DMD
|
B
|
SE
|
Wald
|
HR
|
95%CI
|
P
|
Follow up eosinophils
|
0.119
|
0.051
|
5.399
|
1.127
|
1.109-1.246
|
0.020
|
Visit score
|
-0.471
|
0.156
|
7.006
|
0.354
|
0.465-0.942
|
0.008
|
Treatment
|
2.208
|
0.647
|
9.825
|
7.596
|
2.138-26.990
|
0.002
|
Age of visit
|
-0.191
|
0.084
|
5.176
|
0.827
|
0.701-0.974
|
0.023
|
Kaplan-Meier analysis of different levels of eosinophils at follow-up showed that there were significant differences in the effective maintenance time of different levels of eosinophils for the treatment of DMD, see Fig. 1. The higher the level of eosinophils, the longer the effective maintenance time of treatment. There are also significant differences in the effective maintenance time of DMD treatment by different treatment methods during follow-up visit, see Fig. 2 The curative effect of prednisone treatment is significantly better than that of conservative treatment.
In Cox regression, it was found that there were also significant differences in the efficacy of different levels of eosinophils on DMD, see Fig. 3 The higher the level of eosinophils, the better the curative effect.