Options for inclusion in the study
The retrieval strategy is shown in Fig. 1. A total of 234 studies were identified; since 87 articles were repeated, they were deleted. The titles and abstracts of the remaining 147 articles were screened. Of these, 98 articles were considered insignificant and were therefore excluded. The remaining 49 articles were screened, and 18 articles were excluded for various reasons, such as lack of a control group, lack of data, studies published in languages other than English or Chinese, unclear measurement units, case reports, and reviews. Finally, 25 papers [11, 12, 22–44] comprising 31 studies were included in the current meta-analysis, as showed in Fig. 1.
General characteristics of the included studies
Tables 1 and 2 listed the general characteristics of the included studies. A total of 25 articles involving 31studies were included in this meta-analysis. These studies were published from 2002 to 2022. Among these studies, 11 studies [11, 23–25, 28, 29, 31, 35, 40–44] reported the relationship between plasma IL-10 levels and OSAHS, A study reported differences in IL-10 levels in tonsil tissue between OSAHS patients and non-OSHAS patient9[22], and 19 studies [12, 22, 26, 27, 30, 32–34, 36–39] reported the relationship between the serum IL-10 levels and OSA. The determination methods of IL-10 varied across the studies; however, most of them use enzyme-linked immunosorbent assay (ELISA).
Table 1 Characteristics of included studies
Author
|
Year
|
Country
|
Ethnicity
|
Case/Control
(n)
|
Sample Source
|
Assay approach
|
NOS
|
Study design
|
PSG degree
|
Alberti A et al.[22]
|
2003
|
Italy
|
Caucasian
|
18/20
|
plasma
|
ELISA
|
8
|
Case-control
|
I
|
Bhatt SP et al.[23]
|
2021
|
India
|
Asian
|
190/57
|
Serum
|
ELISA
|
9
|
Cross-sectional
|
I
|
Leon-Cabrera S et al.[12]
|
2015
|
Mexico
|
Latino
|
10/29
|
Serum
|
ELISA
|
8
|
Case-control
|
I
|
Chen VG et al. [24]
|
2018
|
Brazil
|
Latino
|
17/17
|
Tonsil tissue
|
ELISA
|
6
|
Cross-sectional
|
I
|
Dalesio NM et al. (Obese and OSAS) [25]
|
2020
|
America
|
Caucasian
|
8/18
|
Plasma
|
Mass spectrometry
|
8
|
Case-control
|
I
|
Dalesio NM et al. (OSAS Only) [25]
|
2020
|
America
|
Caucasian
|
7/18
|
Plasma
|
Mass spectrometry
|
8
|
Case-control
|
I
|
Dyugovskaya L et al. [26]
|
2002
|
Israel
|
Caucasian
|
8/6
|
Serum
|
Flow
|
8
|
Case-control
|
I
|
cytometer
|
|
|
Galati D et al. [27]
|
2020
|
Italy
|
Caucasian
|
45/30
|
Serum
|
Flow
|
6
|
Cohort
|
I
|
cytometer
|
|
|
Gozal D et al. [11]
|
2008
|
America
|
Caucasian
|
20/20
|
Plasma
|
ELISA
|
8
|
Case-control
|
I
|
Hirsch D et al. [28]
|
2018
|
Australia
|
Caucasian
|
23/16
|
Serum
|
A multiplex bead-based
|
7
|
Cross-sectional
|
I
|
assay
|
|
|
Huang YS et al. [29]
|
2020
|
China
|
Asian
|
55/32
|
Serum
|
ELISA
|
8
|
Case-control
|
I
|
Jiang H et al.(Mild) [30]
|
2015
|
China
|
Asian
|
45/94
|
Serum
|
ELISA
|
7
|
Case-control
|
I
|
Jiang H et al.(Moderate) [30]
|
2015
|
China
|
Asian
|
44/94
|
Serum
|
ELISA
|
7
|
Case-control
|
I
|
Jiang H et al.(Severe) [30]
|
2015
|
China
|
Asian
|
46/94
|
Serum
|
ELISA
|
7
|
Case-control
|
I
|
Ke D et al. [31]
|
2019
|
Canada
|
Caucasian
|
10/5
|
Plasma
|
ELISA
|
8
|
Case-control
|
I
|
Alonso-Fernandez et al. [44]
|
2021
|
Spain
|
Caucasian
|
22/11
|
Plasma
|
ELISA
|
7
|
Case-control
|
I
|
Serednytskyy O et al. [43]
|
2022
|
Spain
|
Caucasian
|
17/34
|
Plasma
|
ELISA
|
7
|
Case-control
|
I
|
Li Y et al.(Serum Mild) [32]
|
2009
|
China
|
Asian
|
22/22
|
Serum
|
ELISA
|
8
|
Cross-sectional
|
I
|
Li Y et al.(Serum Moderate) [32]
|
2009
|
China
|
Asian
|
22/22
|
Serum
|
ELISA
|
8
|
Cross-sectional
|
I
|
Li Y et al.(Serum Severe) [32]
|
2009
|
China
|
Asian
|
24/22
|
Serum
|
ELISA
|
8
|
Cross-sectional
|
I
|
Niżankowska-Jędrzejczyk A et al. [33]
|
2014
|
Poland
|
Caucasian
|
22/16
|
Serum
|
ELISA
|
6
|
Case-control
|
I
|
Qian X et al. [34]
|
2012
|
China
|
Asian
|
30/40
|
Serum
|
ELISA
|
8
|
Case-control
|
I
|
Rogers VE et al. [35]
|
2018
|
America
|
Caucasian
|
20/7
|
Serum
|
ELISA
|
7
|
Cross-sectional
|
I
|
Ryan S et al.(Mild to Moderate) [36]
|
2006
|
Poland
|
Caucasian
|
35/30
|
Serum
|
ELISA
|
9
|
Cohort
|
I
|
Ryan S et al.(Severe) [36]
|
2006
|
Poland
|
Caucasian
|
31/30
|
Serum
|
ELISA
|
9
|
Cohort
|
I
|
Sahlman J et al. [37]
|
2010
|
Finland
|
Caucasian
|
84/40
|
Plasma
|
ELISA
|
7
|
Cross-sectional
|
I
|
Silva WA et al. [39]
|
2021
|
Brazil
|
Latino
|
188/520
|
Serum
|
ELISA
|
8
|
Cross-sectional
|
I
|
Su MS et al. [40]
|
2017
|
China
|
Asian
|
42/48
|
Plasma
|
ELISA
|
7
|
Case-control
|
I
|
Sarinc Ulasli S et al. [38]
|
2015
|
Turkey
|
Caucasian
|
28/20
|
Serum
|
ELISA
|
8
|
Cross-sectional
|
I
|
Wang W et al. [41]
|
2021
|
China
|
Asian
|
22/20
|
Plasma
|
ELISA
|
7
|
Case-control
|
I
|
Zhang Z et al. [42]
|
2017
|
China
|
Asian
|
50/52
|
Serum
|
Flow cytometry
|
6
|
Case-control
|
I
|
|
|
|
|
|
|
|
|
|
|
NOS: Newcastle-Ottawa scale; ELISA: Enzyme linked immunosorbent assay; NA: not available
Table 2
Participants’ characteristics of included studies
Author
|
IL-10 (Mean ± SD)
|
|
BMI (Mean ± SD)
|
|
Age (Mean ± SD)
|
|
AHI(Mean ± SD)
|
Case
|
Control
|
|
Case
|
Control
|
|
Case
|
Control
|
|
Case
|
Control
|
Alberti A et al.[22]
|
3.5 ± 2.7
|
5.4 ± 1.8
|
|
26.5 ± 2.2
|
22.1 ± 3.4
|
|
52.7 ± 12.0
|
51.3 ± 13.2
|
|
18.2 ± 15.1
|
< 5
|
Bhatt SP et al.[23]
|
2.62 ± 0.39
|
2.1 ± 0.28
|
|
27.1 ± 6.5
|
27.4 ± 2.8
|
|
10.7 ± 3.1
|
11.8 ± 2.6
|
|
13.3 ± 6.5
|
1.88 ± 1.11
|
Leon-Cabrera S et al.[12]
|
74.4 ± 17
|
113.2 ± 12
|
|
45.2 ± 8.4
|
23.6 ± 2.1
|
|
37.2 ± 11.4
|
43.4 ± 11.5
|
|
NA
|
NA
|
Chen VG et al.[24]
|
13 ± 11.77
|
10.69 ± 12.95
|
|
NA
|
NA
|
|
NA
|
NA
|
|
NA
|
NA
|
Dalesio NM et al.(Obese and OSAS) [25]
|
1 ± 0.7
|
0.46 ± 0.16
|
|
NA
|
NA
|
|
7 ± 2.5
|
8.3 ± 2.12
|
|
NA
|
NA
|
Dalesio NM et al.(OSAS Only)[25]
|
0.59 ± 0.17
|
0.46 ± 0.16
|
|
NA
|
NA
|
|
6.0 ± 1.8
|
8.3 ± 2.12
|
|
NA
|
NA
|
Dyugovskaya L et al.[26]
|
6.1 ± 4.8
|
14.7 ± 4.9
|
|
29.36 ± 5.5
|
27.3 ± 3.4
|
|
53.5 ± 12.9
|
53.2 ± 10.4
|
|
NA
|
NA
|
Galati D et al.[27]
|
2.44 ± 0.91
|
1.1 ± 0.68
|
|
28 ± 2.2
|
26.3 ± 1.8
|
|
53.9 ± 11.6
|
55 ± 5.8
|
|
43.5 ± 38.5
|
< 5
|
Gozal D et al.[11]
|
195.2 ± 33.6
|
458.5 ± 102.2
|
|
17.3 ± 0.6
|
17.1 ± 0.5
|
|
6.5 ± 0.6
|
6.4 ± 0.7
|
|
13.3 ± 2.8
|
0.0 ± 0.0
|
Hirsch D et al.[28]
|
2.08 ± 1.3
|
1.83 ± 0.77
|
|
NA
|
NA
|
|
10.0 ± 1.7
|
10.7 ± 1.2
|
|
NA
|
NA
|
Huang YS et al.[29]
|
2.74 ± 2.94
|
2.1 ± 1.41
|
|
16.8 ± 4.0
|
17.44 ± 30
|
|
7.6 ± 2.6
|
7.0 ± 0.6
|
|
15.7 ± 22.6
|
0.46 ± 0.28
|
Jiang H et al.(Mild)[30]
|
6.68 ± 4.74
|
8.76 ± 5.25
|
|
NA
|
27.5 ± 2.58
|
|
NA
|
47.2 ± 13.5
|
|
11.52 ± 3.9
|
1.6 ± 1.6
|
Jiang H et al.(Moderate)[30]
|
5.77 ± 4.06
|
8.76 ± 5.25
|
|
NA
|
27.5 ± 2.58
|
|
NA
|
47.2 ± 13.5
|
|
21.06 ± 8.64
|
1.6 ± 1.6
|
Jiang H et al.(Severe)[30]
|
4.87 ± 3.84
|
8.76 ± 5.25
|
|
NA
|
27.5 ± 2.58
|
|
NA
|
47.2 ± 13.5
|
|
35.1 ± 24. 7
|
1.6 ± 1.6
|
Ke D et al.[31]
|
102.28 ± 115.08
|
257.15 ± 100.02
|
|
15.4 ± 0.6
|
16.5 ± 1.0
|
|
2.7 ± 0.4
|
4.0 ± 0.6
|
|
NA
|
NA
|
Li Y et al.(Serum Mild) [32]
|
46.7 ± 4.6
|
50.1 ± 6.9
|
|
25.7 ± 4.2
|
23.3 ± 2.0
|
|
48 ± 12
|
43.0 ± 93.0
|
|
14.1 ± 3.5
|
2.9 ± 1.3
|
Li Y et al.(Serum Moderate) [32]
|
35.7 ± 5.3
|
50.1 ± 6.9
|
|
28.8 ± 5.3
|
23.3 ± 2.0
|
|
44 ± 13
|
43.0 ± 93.0
|
|
29.7 ± 5.5
|
2.9 ± 1.3
|
Li Y et al.(Serum Severe) [32]
|
24.6 ± 5.1
|
50.1 ± 6.9
|
|
28.67 ± 4.2
|
23.3 ± 2.0
|
|
44 ± 8
|
43.0 ± 93.0
|
|
70.1 ± 18.1
|
2.9 ± 1.3
|
Niżankowska-Jędrzejczyk A et al.[33]
|
5.45 ± 4.02
|
3.97 ± 1.37
|
|
30.1 ± 2.7
|
28.0 ± 3.3
|
|
52.5 ± 8.3
|
54.0 ± 2.0
|
|
23.6 ± 12.3
|
2.26 ± 1.97
|
Qian X et al.[34]
|
26.1 ± 20.98
|
38.6 ± 10.6
|
|
29.4 ± 2.1
|
29.4 ± 2.1
|
|
45.0 ± 9.0
|
46.3 ± 8.1
|
|
NA
|
NA
|
Rogers VE et al.[35]
|
1.7 ± 1.06
|
2.03 ± 1.76
|
|
NA
|
NA
|
|
NA
|
NA
|
|
13.1 ± 9.8
|
0.8 ± 0.3
|
Ryan S et al.(Mild to Moderate) [36]
|
1.19 ± 0.42
|
1.19 ± 0.41
|
|
32.9 ± 6.03
|
30.7 ± 3.1
|
|
42.0 ± 8.0
|
41.0 ± 8.0
|
|
15.9 ± 7.7
|
1.2 ± 1.0
|
Ryan S et al.(Severe)[36]
|
1.14 ± 0.42
|
1.19 ± 0.41
|
|
32.1 ± 3.5
|
30.7 ± 3.1
|
|
43.0 ± 9.0
|
41.0 ± 8.0
|
|
56.6 ± 20.9
|
1.2 ± 1.0
|
Sahlman J et al.[37]
|
1.28 ± 2.34
|
0.7 ± 1.51
|
|
32.5 ± 3.3
|
31.5 ± 3.5
|
|
50.4 ± 9.3
|
45.6 ± 11.5
|
|
9.6 ± 2.9
|
1.9 ± 1.4
|
Silva WA et al.[39]
|
2.59 ± 0.37
|
2.51 ± 0.52
|
|
28.7 ± 4.0
|
25.1 ± 3.7
|
|
47 ± 6
|
45 ± 5
|
|
27.0 ± 13.1
|
5.4 ± 5.1
|
Su MS et al.[40]
|
261.7 ± 18.5
|
338.2 ± 25.2
|
|
NA
|
NA
|
|
NA
|
NA
|
|
8.68 ± 4.56
|
2.40 ± 2.75
|
Sarinc Ulasli S et al.[38]
|
34.7 ± 12.4
|
32.5 ± 14
|
|
32.6 ± 4.4
|
30.4 ± 8
|
|
53.7 ± 12.7
|
45.3 ± 14
|
|
34.08 ± 52.7
|
2.3 ± 2.9
|
Wang W et al.[41]
|
5.51 ± 1.53
|
5.7 ± 1.2
|
|
21.6 ± 2.2
|
18.9 ± 1.7
|
|
6.58 ± 0.61
|
5.69 ± 0.78
|
|
33.4 ± 5.6
|
1.1 ± 0.2
|
Zhang Z et al.[42]
|
3.45 ± 0.19
|
3.01 ± 0.17
|
|
NA
|
NA
|
|
NA
|
NA
|
|
NA
|
NA
|
Alonso-Fernandez et al.[44]
|
1.08 ± 0.19
|
0.95 ± 0.13
|
|
27.2 ± 2.4
|
26.6 ± 3.2
|
|
33.9 ± 5
|
35.6 ± 3.1
|
|
7.79 ± 3.48
|
0.37 ± 0.49
|
Serednytskyy O et al.[43]
|
0.64 ± 0.59
|
0.64་0.49
|
|
28.8 ± 6.06
|
26.04 ± 4.57
|
|
37.64 ± 4.04
|
35.35 ± 5.42
|
|
8.94 ± 5.90
|
0.5 ± 0.54
|
Quality assessment
The quality assessment results of the included studies are shown in Table 1. Among the included studies, 27 were of high quality and the other four were of medium quality.
Meta-analysis
Comparison of plasma IL-10 concentrations between OSAHS patients and control group
The outcomes of a pooled analysis of plasma IL-10 concentrations in patients are illustrated in Fig. 2A. The results indicated that no difference in the plasma levels of IL-10 was detected between OSAHS patients and controls (SMD=-0.68, 95%CI= -1.58-0.21, P = 0.136). The random-effects model was selected for investigation as a result of the high heterogeneity (I2 = 94.3%), as showed in Fig. 2A.
Comparison of serum IL-10 concentration among OSAHS patients and control group
The correlation of serum IL-10 concentrations in patients with OSAHS with those in the control group was investigated in our meta-analysis, which included 19 eligible observational studies. Serum levels of IL-10 were no significant differences between the two groups (SMD=-0.12, 95%CI= -0.55-0.32, P = 0.591). For further investigation, the random-effects model was selected as a result of the increased heterogeneity (I2 = 94.4%, Fig. 2B).
Sensitivity analysis
According to the literature on IL-10 expression levels in plasma and serum of patients with OSAHS, no studies were found showing an impact on the results after excluding each study in turn. Sensitivity analysis showed that the meta-analysis was stable (Fig. 3A, B).
Comparison of palatine tonsils of IL-10 concentration among OSAHS patients and control group
Chen VG et al. collected tonsillar tissue samples from 34 children who underwent tonsillectomy (including 17 children with OSAHS and 17 children without OSAHS) and detected the expression of IL-10 and other inflammatory factors in the tissues using enzyme linked immunosorbent assay (ELISA). Chen VG et al. found that the expression of IL-10 cytokines in tonsils of children with OSAHS was significantly increased (OSAHS group vs. control group: 13.08 ± 11.77 pg/ml vs. 10.69 ± 12.95 pg/ml, p = 0.04).
Subgroup analysis of plasma IL-10 concentration
Age
Plasma IL-10 levels in children with and without OSAHS were studied in four investigations. According to the findings, there was no major variation in plasma IL-10 levels between children with OSAHS and the control group (SMD=-1.05, 95%CI= -2.74-0.65, P = 0.226). Four studies compared plasma IL-10 concentrations in adults with and without OSAHS, indicating plasma IL-10 levels did not differ between the two groups (SMD=-0.17, 95%CI= -0.84-0.49, P = 0.613, Table 3).
Table 3
Subgroup analyses of the association between the interleukin-10 (IL-10) levels and OSAHS in the meta-analysis
Subgroup analysis of plasma levels(n)
|
SMD(95% CI), P-value, I2(%), Ph
|
|
Subgroup analysis of serum levels(n)
|
SMD(95% CI), P-value, I2(%), Ph
|
Overall (11)
|
|
|
Overall (19)
|
|
Ethnicity
|
|
|
Ethnicity
|
|
Caucasian (9)
|
-0.41(-1.21,0.40), 0.320, 90.4%, ,<0.001
|
|
Caucasian (7)
|
0.31(-0.18,0.81), 0.214, 96.5%, < 0.001
|
Asian (2)
|
-1.78(-5.00,1.44), 0.279, 98.1%, ,<0.0001
|
|
Asian (10)
|
-0.28(-1.05,0.48), 0.471, 96.8%, < 0.001
|
Latino
|
|
|
Latino (2)
|
-1.09(-3.65,1.48), 0.407, 79.8%, <0.001
|
BMI
|
|
|
BMI
|
|
Mean BMI ≥ 30 (3)
|
-0.61(-3.24,2.02), 0.650, 98.2%, < 0.001
|
|
Mean BMI ≥ 30 (5)
|
-0.12(-0.55,0.32), 0.389, 95.5%, < 0.001
|
Mean BMI༜30 (8)
|
-0.69(-1.56,0.18), 0.121, 89.3%, < 0.0001
|
|
Mean BMI༜30 (14)
|
-0.05(-0.57,0.48), 0.854, 86.3%, < 0.001
|
Degree of severity
|
|
|
Degree of severity
|
|
Mean AHI ≥ 30 (2)
|
-0.68(-1.58,0.21), 0.290, 95.3%, < 0.001
|
|
Mean AHI ≥ 30 (7)
|
-0.71(-1.72,0.30), 0.171, 95.6%, ,<0.001
|
Mean AHI༜30 (9)
|
-0.67(-1.75,0.41), 0.221, 70.4%, < 0.001
|
|
Mean AHI༜30 (12)
|
0.18(-0.29,0.66), 0.447, 93.6%, ,<0.001
|
Age
|
|
|
Age
|
|
Adult (5)
|
-0.17(-0.84,0.49), 0.613, 80.7%, < 0.001
|
|
Adult (14)
|
-0.44(-0.89,0.001), 0.050, 92.9%, < 0.001
|
Nonage (6)
|
-0.68(-1.58,0.21), 0.226, 96.0%, < 0.001
|
|
Nonage (5)
|
0.85(-0.03,1.73, ,0.05, ༌93.4, ༌<0.001
|
Design
|
|
|
Design
|
|
Cross-sectional study (1)
|
NA
|
|
Cross-sectional study (8)
|
-0.60(-1.42,0.23), 0.155, 95.9%, < 0.001
|
Case-control study (10)
|
-0.79(-1.82,0.25), 0.138, 94.2%, < 0.001
|
|
Case-control study (8)
|
0.07(-0.61,0.75), 0.838, 94.3%, < 0.001
|
Cohort study
|
NA
|
|
Cohort study (3)
|
0.49(-0.58,1.56), 0.365, 94.4%, <0.001
|
BMI
We performed subgroup analysis depending on whether the mean BMI was greater than 30, which was reported in all the included studies. Three studies reported plasma IL-10 concentrations in patients whose mean BMI was over 30, suggesting that plasma IL-10 concentrations in the OSAHS group were not lower or greater in comparison with those in the healthy control group (SMD=-0.61, 95%CI= -3.24-2.02, P = 0.650). According to eight studies, the plasma IL-10concentrations in patients with an average BMI lower than 30 had no major variation between the OSAHS and the subjects of the control groups (SMD=-0.69, 95%CI= -1.56-0.18, P = 0.121, Table 3).
Severity of OSAHS
A subgroup meta-analysis of OSAHS severity was also undertaken. Plasma IL-10 levels were measured in two studies with an average AHI ≥ 30 and nine studies reported plasma IL-10 levels in patients with an average AHI < 30. The outcomes revealed that regardless of whether the mean AHI of patients was greater than 30, there was no major variation in plasma IL-10levels between OSAHS patients and the control group (SMD=-0.66, 95%CI= -1.88-0.56, P = 0.290; SMD=-0.67, 95%CI=-1.75 ~ 0.41, P = 0.221, Table 3).
Ethnicity
The subgroup study of plasma IL-10 concentrations in OSAHS patients belonging to different ethnicities is compiled in Table 3. Caucasian and Asian populations are shown as the major subgroups. In the Caucasian population, there was no major variation in plasma IL-10 concentration between the OSAHS and control groups (SMD=-0.41, 95%CI=-0.121 ~ 0.40, P = 0.320). In the Asian population group, at the plasma IL-10 levels, similar result was observed (SMD=-1.79, 95%CI= -5.00-1.44, P = 0.279, Table 3).
Research design type
We conducted a matching subgroup analysis because the differences in design types included in the studies might alter the heterogeneity of the results. No major variation was seen in plasma IL-10 concentration between OSAHS patients and the control group (SMD=-0.44, 95 percent CI= -0.89-0.00, P = 0.05) according to a subgroup analysis of case-control studies. One study was a cross-sectional study and the majority were case-control studies. (Table 3)
Subgroup analysis of serum IL-10 concentrations
Age
The serum IL-10 levels of OSAHS children and healthy children were evaluated in five investigations. The combined results revealed that there was no major variation in serum IL-10 levels between children with OSAHS and the control group (SMD = 0.85, 95% CI= -0.03-1.73, P = 0.059). Fourteen studies compared serum IL-10 concentrations in adults with and without OSAHS, indicating serum IL-10 levels did not differ between the two groups (SMD=-0.44, 95 percent CI= -0.89-0.00, P = 0.05, Table 3).
BMI
Based on BMI, we conducted a subgroup meta-analysis of 19 studies. Five studies were based on serum IL-10 concentrations of patients with an average BMI ≥ 30, and the outcomes of the analysis suggested that serum IL-10 levels did not differ from those of healthy controls in patients with OSAHS (SMD=-0.32, 95%CI= -1.04-0.41, P = 0.389). Fourteen publications gave us data on serum IL-10 concentration in patients with an average BMI < 30, and the outcomes of the analysis suggested that serum IL-10 levels did not differ from those of healthy controls in patients with OSAHS (SMD=-0.32, 95%CI= -1.04-0.41, P = 0.389, Table 3) nether.
OSAHS of severity
Many studies have found a close link between serum IL-10 concentration to the AHI value in patients with OSAHS. In the current report, a comparison of serum IL-10 concentrations between OSAHS patients with an average AHI ≥ 30 and the control group were carried out in 7 studies, and the findings suggested that serum IL-10 levels in these patients were not significant different as compared to those in the control group (SMD=-0.71, 95%CI= -1.72-0.30, P = 0.171). Twelve studies were based on a comparison of the serum IL-10 concentrations between OSAHS patients with an average AHI < 30 and the control group, and the outcomes indicated that serum IL-10 concentrations in these patients were not significant different when compared with subjects of the control group (SMD = 0.18, 95%CI= -0.29-0.66, P = 0.447, Table 3).
Ethnicity
There were 19 reports from Caucasians, Asians, and Latin Americans included in the study. The serum IL-10 concentrations of OSAHS patients in these three groups exhibited no significant differences between OSAHS group and the control group, according to subgroup analysis based on different demographics and ethnicities. (See Table 3)
Research design type
In cross-sectional studies, blood IL-10 concentrations in patients with OSAHS presented with comparable levels as those in control group (SMD=-0.60, 95% CI=-1.42-0.22, P = 0.155). Similarly, serum IL-10 concentrations in patients with OSAHS were not statistically different between OSAHS patients and controls in case-control and cohort studies (SMD = 0.07, 95 percent CI= -0.61-0.75, P = 0.838; SMD = 0.49, 95 percent CI=-0.57-1.56, P = 0.591, Table 3).
Meta-regression
All the included studies had an I2 value of 94.3% in plasma IL-10 and an I2 value of 94.4% in serum IL-10, indicating a high level of heterogeneity. As a result, we explored the possible sources of this heterogeneity using meta-regression. The Table 4 demonstrates the meta-regression of serum/plasma IL-10 concentration. Unfortunately, the potential sources of heterogeneity were explored through meta-regression analysis; however, they were not determined.
Table 4
Meta-regression analysis of variables predicting serum and plasma levels of IL-10
Variables
|
R
|
Adjusted R2
|
P
|
Age
|
|
|
|
Serum
|
− 1.339
|
0.139
|
0.079
|
Plasma
|
0.786
|
-0.040
|
0.461
|
BMI
|
|
|
|
Serum
|
-0.274
|
-0 057
|
0 733
|
Plasma
|
0.096
|
-0.121
|
0.936
|
Severity
|
|
|
|
Serum
|
− 0.873
|
0. 029
|
0.223
|
Plasma
|
− 0.066
|
-0.118
|
0.963
|
Ethnicity
|
|
|
|
Serum
|
-0.281
|
-0.032
|
0.454
|
Plasma
|
-0.672
|
0.019
|
0.317
|
Design
|
|
|
|
Serum
|
0.031
|
-0.067
|
0.950
|
Plasma
|
1.062
|
-0.077
|
0.560
|
Publication bias
Funnel plots were used to investigate the probability of publication bias in research investigating the link between IL-10 concentration and OSAHS, and our funnel plots appear to be symmetric. Begg's and Egger's tests (plasma: Begg's (P = 0.276), Egger's (t=-0.93, P = 0.379); serum: Begg's (P = 0.162), Egger's (t=-0.77, P = 0.453)) could not report any publication bias in the studies of patients with OSAHS (Fig. 4A and Fig. 4B).
Bioinformatic results
Expression of IL-10 mRNA in subcutaneous fat tissue (GSE135917) and visceral adipose tissue (GSE38792) of OSAHS patients was not significantly different from that of the control group, as shown in Fig. 5A, B.