INTIMP-03663; No of Pages 7 International Immunopharmacology xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

International Immunopharmacology journal homepage: www.elsevier.com/locate/intimp

F

a

7

a r t i c l e

8 9 10 11 12

Article history: Received 17 December 2014 Received in revised form 1 April 2015 Accepted 15 April 2015 Available online xxxx

13 14 15 16 17 18 19 20 21

Keywords: S100B Carbon monoxide poisoning Correlation Biomarker Diagnosis Treatment Case–control study Meta-analysis

i n f o

R O

Department of Emergency, The First Affiliated Hospital of Dalian Medical University, Dalian 116000, PR China Department of Abdominal Emergency, The First Affiliated Hospital of Dalian Medical University, Dalian 116000, PR China Department of Emergency, The First Affiliated Hospital of China Medical University, Shenyang 110001, PR China

a b s t r a c t

Objective: To investigate the correlation between serum S100B level and carbon monoxide (CO) poisoning by meta-analysis. Methods: By searching both English and Chinese language-based electronic databases (PubMed, EBSCO, Ovid, Springerlink, Wiley, Web of Science, Wanfang databases, China national knowledge infrastructure (CNKI), VIP database, etc.) thoroughly, we tabulated and analyzed the collected data with the use of Comprehensive Meta-analysis 2.0 (CMA 2.0). Results: Totally 108 studies have been searched initially (92 studies in Chinese, 16 studies in English). Nine case– control studies (4 studies in English, 5 in Chinese) were chosen for an updated meta-analysis including 542 patients with CO poising and 236 healthy controls. Results identified that the serum S100B level were obviously higher than that in healthy controls (SMD = 1.600, 95% CI = 1.055–2.145, P b 0.001). A subgroup based on the ethnicities revealed that the serum S100B level in Caucasian and Asian subgroups was clearly higher than serum S100B level in healthy controls (Asians: SMD = 2.0624, 95% CI = 1.736–3.511, P b 0.001; Caucasians: SMD = 0.447, 95% CI = 0.197–0.697, P b 0.001). Conclusion: Serum S100B level may be correlated with the CO poisoning and could be effective biomarker for early diagnosis and treatment monitoring in CO poisoning. © 2015 Published by Elsevier B.V.

P

c

D

b

41 39 38

R

40

1. Introduction

43

Carbon monoxide (CO) is formed during the incomplete combustion of any carbon-containing compound and it has been estimated that CO poisoning might be responsible for more than half of the fatal poisonings worldwide [1]. Despite the aggressive treatment approaches taken in many industrialized countries, the morbidity and mortality rates from CO poisoning have remained high as half of them may have cognitive sequelae between 3 days and 4 weeks after the event [2]. CO poisoning results in generalized hypoxia in the body, and the central nervous system (CNS) is the most sensitive area to this poisoning, especially for severe neurologic and psychiatric sequelae [3]. Approximately 67% of the patients with CO poisoning has been found with cardiovascular and CNS prominent, and brain injury also might occur depending on the severity of the clinical picture in patients with CO poisoning which the diagnosis and prognosis was still controversial [3]. Some patients with acute CO poisoning will experience delayed neurological sequelae

52 53 54 55 56 57

N C O

50 51

U

48 49

R

42

46 47

22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

E

37

44 45

O

4 5 6

E

3

Hui Liu a, Yu Zhang a, Yan-Bo Ren a, Jian Kang a, Jing Xing a, Qing-Hui Qi b, Dong-Na Gao a,⁎, Tao Ma c, Xiao-Wei Liu c, Zhi Liu c

T

2Q2

Serum S100B level may be correlated with carbon monoxide poisoning

C

1Q1

⁎ Corresponding author at: Department of Emergency, The First Affiliated Hospital of Dalian Medical University, Zhongshan Road No. 222, Xigang District, Dalian 116000, PR China. Tel.: +86 411 83635963 3186. E-mail addresses: [email protected], [email protected] (D.-N. Gao).

and several factors associated with it have been reported, but these factors are unsatisfactory for the assessment of unconscious patients [4]. In the literature, the use of brain injury markers in poisoning has been investigated in several studies such as S100B, neuron specific enolase, and glial fibrillary acidic protein [5,6]. Early elevation of S100B in cerebrospinal fluid after CO poisoning could be a suitable predictor of subsequent development of persistent vegetative state [7]. S100B, a member of the S100 family, is an acidic protein with a molecular weight of 21 kDa and play an important role in neurodevelopment, differentiation, and brain construction, and extracellular concentration of S100B increases following brain damage, which easily penetrates into cerebrospinal fluid in brain damage and then into the blood [8,9]. S100B as one of the brain specific biomarkers found in the past decades, is expressed and produced by astrocytes in vertebrate brains and in the CNS, and the astrocytes are the major cells producing S100B protein in gray matter, as well as oligodendrocytes are the predominant S100B in protein producing cells in white matter [10,11]. A normal S100B level has constantly been applied to predict the absence of major CNS damage and increased serum S100B level are not essentially related to neuroglia damage but could also reveal the current failure of the blood brain barrier [5]. A number of reports examined serum S100B level after CO poisoning, as it has been reported that serum S100B level was considerably high during initial examination of patients with CO poisoning, suggesting

http://dx.doi.org/10.1016/j.intimp.2015.04.027 1567-5769/© 2015 Published by Elsevier B.V.

Please cite this article as: H. Liu, et al., Serum S100B level may be correlated with carbon monoxide poisoning, Int Immunopharmacol (2015), http://dx.doi.org/10.1016/j.intimp.2015.04.027

58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80

H. Liu et al. / International Immunopharmacology xxx (2015) xxx–xxx

R

E

C

T

Included

E

D

P

Eligibility

R O

O

Screening

F

n

2

O

R

Fig. 1. Flow chart shows study selection procedure. Nine studies were included in this meta-analysis.

100

2.2. Study selection

106

90

2. Materials and methods

107

91

2.1. Literature search

92

The following computerized bibliographic databases were applied to identify relevant articles related to the association of Serum S100B level and CO poisoning without restrictions with respect to data collection: PubMed, EBSCO, Ovid, Springerlink, Wiley, Web of Science, Wanfang databases, China national knowledge infrastructure (CNKI), VIP databases (last updated search in September, 2014) utilizing selected common keywords regarding S100B protein and carbon monoxide poisoning. (“S100B protein” or “S100B protein” or “S100 calcium-

After reading the abstract, full papers were retrieved and assessed for their suitability with the following inclusion criteria: (1) only those case–control studies clarifying the correlation between serum level of S100B and CO poisoning were enrolled; (2) subjects investigated in case and control groups should be CO poisoning patients and healthy controls, respectively; (3) eligible articles must provide sufficient complete data; (4) language was restricted to Chinese and English for eligible articles. The studies were excluded based on the following exclusion criteria: (1) supply incomplete data; (2) significant differences existed in baseline characteristics between cases and controls; (3) duplicate publications or studies with overlapping data; (4) no clear diagnostic criteria applied in CO poisoning patients.

86 87

93 94 95 96 97 98 99

N

84 85

U

82 83

C

binding protein β subunit” or “S100 Calcium Binding Protein beta Subunit”) and (“Carbon Monoxide Poisoning” or “CO poisoning” or “CMP” or “Gas Poisoning”) were selected as exposure factors and outcome factors in databases for thoroughly searches with no restriction set to the language of the article. In addition, manual searches were applied to scan other relevant studies.

88 89

that serum S100B level might be correlated with loss of consciousness in patients with CO poisoning [6,7,12]. However, previews report has also found that, S100B protein level had not increased in response to hypoxic brain damage in children [3]. Moreover, it has been suggested that there was no consistent evidence for a correlation between disease severity and concentrations of S100B in cerebrospinal fluid or serum and thus S100B has limited usefulness for monitoring disease progression [13]. Accordingly, this study aimed to investigate the correlations with serum S100B level and carbon monoxide poisoning by metal analysis.

81

Please cite this article as: H. Liu, et al., Serum S100B level may be correlated with carbon monoxide poisoning, Int Immunopharmacol (2015), http://dx.doi.org/10.1016/j.intimp.2015.04.027

101 102 103 104 105

108 109 110 111 112 113 114 115 116 117 118

H. Liu et al. / International Immunopharmacology xxx (2015) xxx–xxx

Year

Ethnicity

Country

Detection methods

Degree of poisoning

Caucasians

Denmark

NR

2005

Asians

China

ELISA

2006

Asians

China

ELISA

2007

Asians

China

ELISA

2009

Caucasians

Turkey

NR

2009

Asians

China

ELISA

2011

Asians

China

ELISA

2013

Caucasians

Turkey

NR

2014

Caucasians

Turkey

NR

Mild Moderate Mild Moderate Severe Mild Moderate Severe Mild Moderate Severe Mild Moderate Mild Moderate Mild Moderate Mild Moderate Severe Mild Moderate

M: male; F: female; NR: not reported; ELISA: enzyme-linked immunosorbent assay.

119

2.3. Data extraction and quality assessment

120

Two investigators were arranged for extracting data from each included study using a standard reporting form, and the following descriptive information were collected: surname and initials of the first author, year of publication, country, ethnicity, language, disease, gender, study design, the degree of CO poisoning detection methods, numbers of cases and controls, etc. The discrimination basis of the degree of CO poisoning is as follows: short Mild: poisoning time with

16/4

NR

26/72

20

10/14

8/12

26.3 ± 7.8

25.7 ± 8.0

38

37/28

NR

42.5 ± 3.8

NR

20

16/12

12/8

8.4 (2.3–14.0)

7.9 (2.1–13.8)

20

25/45

NR

36.6 ± 16.3

NR

30

97/88

13/17

42.5 ± 3.8

41.6 ± 2.5

33

35/30

17/16

58.18 ± 8.95

57.56 ± 9.01

Control

30

17/13

7/23

7.88 ± 3.75

8.16 ± 3.05

25

22/33

9/16

45.6 ± 20.3

NR

10–20% concentrations of carboxyhemoglobin, and the clinical presentations are headache and vertigo; Moderate: slightly long poisoning time with 30–40% concentrations of carboxyhemoglobin, and the clinical presentations are collapse or coma state; Severe: long poisoning time with 50% concentrations of carboxyhemoglobin, and the clinical presentations are deep coma, areflexia, fecal and urinary incontinence, cold limbs, blood pressure decrease, tachypnea, limb paralysis or even death. Two reviewers assessed the methodological quality of the included trials using the Newcastle–Ottawa Scale (NOS) criteria [14],

R

E

C

20

R

125 126

9 11 9 9 6 45 14 6 5 17 6 57 13 120 65 32 33 21 7 2 24 31

N C O

123 124

Case

U

121 122

T

t1:27

Control

P

2004

Age (years)

Case

D

Rasmussen LS-a [23] Rasmussen LS-b [23] Li FX-a [24] Li FX-b [24] Li FX-c [24] Du XD-a [25] Du XD-b [25] Du XD-c [25] Cao YH-a [26] Cao YH-b [26] Cao YH-c [26] Yardan T-a [28] Yardan T-b [28] Wang JG-a [27] Wang JG-b [27] Han GH-a [29] Han GH-b [29] Akelma AZ-a [35] Akelma AZ-b [35] Akelma AZ-c [35] Akdemir H-a [35] Akdemir H-b [35]

Gender (M/F)

Control

E

t1:5 t1:6 t1:7 t1:8 t1:9 t1:10 t1:11 t1:12 t1:13 t1:14 t1:15 t1:16 t1:17 t1:18 t1:19 t1:20 t1:21 t1:22 t1:23 t1:24 t1:25 t1:26

Number Case

F

First author

t1:4

O

t1:3

Table 1 Baseline characteristics of the individual studies in this meta-analysis.

R O

t1:1 t1:2

3

Fig. 2. Newcastle–Ottawa Scale scores for 9 eligible studies in present meta-analysis investigating the correlation between serum S100B level and carbon monoxide poisoning.

Please cite this article as: H. Liu, et al., Serum S100B level may be correlated with carbon monoxide poisoning, Int Immunopharmacol (2015), http://dx.doi.org/10.1016/j.intimp.2015.04.027

127 128 129 130 131 132 133 134 135

4

H. Liu et al. / International Immunopharmacology xxx (2015) xxx–xxx

used to evaluate the funnel plot asymmetry whose asymmetric plot 155 revealed possible publication bias [21,22]. 156

3. Results

157

3.1. Included studies

158

Sums of 108 studies were identified from our electronic database searching and manual searching. We excluded 90 according to the title and abstracts, and 5 studies were canceled by further reading, as well as 2 passages without completely statistics, remaining 9 studies ultimately [3,23–30] including 542 patients with CO poisoning and 236 normal people qualified for our meta-analysis between 2004 and 2014 (Fig. 1). Five studies in all were performed in populations of Asians while others were Caucasians. The characteristics of these 9 included studies were presented in Table 1 and NOS for 9 included studies was shown in Fig. 2.

159

O

F

p

2.4. Statistical analysis

142

Present study was performed with the use of Comprehensive Metaanalysis 2.0 (Biostat Inc., Englewood, New Jersey, USA). The difference of serum level of S100B between case and control groups was estimated by the standardized mean difference (SMD) and 95% confidence interval (95% CI). Z test were conducted to evaluate the significance of the overall effect size. We used Cochran's Q-statistic (P b 0.05 was considered significant) and I2 tests to quantify heterogeneity among studies [16]. In order to calculate the pool SMDs, fixed/random effects model were used; random effects model was applied for the evidence of significant heterogeneity (P b 0.05 or I2 test exhibited N 50%), whereas SMDs were pooled based on the fixed-effects model [17,18]. Further, the effect of publication bias was detected by classic fail-safe N [19,20] and Egger's linear regression test (P b 0.05 was considered significant) which can be

154

T

C

E

R

152 153

R

151

O

149 150

C

147 148

168

170

N

145 146

166 167

The total 9 studies reported the differences of serum S100B level between patients with CO poisoning and normal people. Considering existing heterogeneity (I2 = 92.320%, P b 0.001), random effects model was applied and most of the studies appeared in the middle domain of the Bivariate Boxplot while two of them out of the region which suggested the heterogeneity (Fig. 3). The major findings of the present meta-analysis revealed that the serum S100B level in patients with CO poisoning were evidently higher than that in healthy controls (SMD = 1.600, 95% CI = 1.055–2.045, P b 0.001) (Fig. 4). The subgroup analysis based on ethnicity implied that serum S100B level was associated with CO poisoning in ethnicity subgroup (Asians: SMD = 2.624, 95% CI = 1.736–3.511, P b 0.001; Caucasians: SMD = 0.447, 95% CI = 0.197–0.697, P b 0.001) (Fig. 5A). According to the degree of CO poisoning, a subgroup analysis was conducted. The results demonstrated that the serum level of S100B in patients with CO poisoning of mild, moderate and severe was significantly higher than that of the healthy controls, and the differences existed statistical significance (Mild: SMD = 0.692, 95% CI = 0.317– 1.066, P b 0.001; Moderate: SMD = 1.620, 95% CI = 0.692–2.548, P = 0.001; Severe: SMD = 6.078, 95% CI = 1.964–10.192, P = 0.004) (Fig. 5B).

U

143 144

164 165

169

P

141

138 139

162 163

3.2. Pooled outcome of meta-analysis

D

140

independently. Three aspects were considered in the NOS criteria: (1) subject selection: 0–4; (2) comparability of subject: 0–2; (3) clinical outcome: 0–3. The range of NOS scores is from 0 to 9; and a score ≥7 means a good quality [15]. Disagreement on the inclusion of one single literature was determined by consultation with another investigator.

E

136 137

R O

Fig. 3. Bivariate boxplot applied for heterogeneity testing in present meta-analysis investigating the correlation between serum S100B level and carbon monoxide poisoning.

160 161

Fig. 4. Forest plots for pooled SMDs in present meta-analysis investigating the correlation between the serum level of S100B and carbon monoxide poisoning.

Please cite this article as: H. Liu, et al., Serum S100B level may be correlated with carbon monoxide poisoning, Int Immunopharmacol (2015), http://dx.doi.org/10.1016/j.intimp.2015.04.027

171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190

H. Liu et al. / International Immunopharmacology xxx (2015) xxx–xxx

5

R O

O

F

A

R

E

C

T

E

D

P

B

N C O

R

Fig. 5. Subgroup analyses based on ethnicity in present meta-analysis investigating the correlation between the serum level of S100B and carbon monoxide poisoning. A, ethnicity; B, degree of CO poisoning.

3.3. Sensitivity analysis and publication bias

192 193

198

Sensitivity analysis was performed, and the results in our study demonstrated that no single study had the weight to impact on the overall estimate of the association between serum S100B level and CO poisoning (Fig. 6A). Publication bias was assessed with funnel plots demonstrating the connection between study sample sizes and estimating precision. Classic fail-safe N and Egger regression asymmetry further indicated minimal evidence of publication bias (P b 0.001) (Fig. 6B).

199

4. Discussion

200

The main findings of our study suggested that serum S100B level may be correlated with the CO poisoning and could be effective biomarker for early diagnosis and treatment monitoring in CO poisoning. Present study showed that tissue hypoxia after CO exposure is one of the most common causes of toxic brain injury, and Hippocampus, ganglia, and the white substance are frequently affected in the central nervous system [31]. As carboxyhemoglobin (COHb) level in patients with CO poisoning were not correlated with clinical findings or symptoms, new markers were in need to identify the central nervous system damage [32]. S100B was a calcium binding protein produced by astroglia in

194 195 196 197

201 202 203 204 205 206 207 208 209

U

191

the brain and has been used as a marker of neuronal damage after brain injury [7]. It was released into the peripheral blood as a result of an impaired blood–brain barrier and hypoxic damage caused by ischemia and edema with high intracranial pressure [28]. It has been shown to be a useful neurobiochemical marker of brain damage in cardiac arrest, stroke, subarachnoid hemorrhage, and traumatic head injury [33]. This may be explained by the fact that the half-life of S100B was around 60 min on average and S100B can immediately leak to the serum shortly after brain injury or anoxia [6]. With respect to that, the raising serum S100B level may implied brain edema with a disturbance of astroglial cell membrane integrity and blood–brain barrier function, thus the increasing serum S100B level may indicate the ongoing destruction of neurons and the release of this intracellular protein into the serum [28,34]. Increased S100B protein levels mediated by traumatic brain damage have been reported in children, and additionally, several studies performed on adults reported that S100B protein might act as an indicator of CO-induced hypoxic brain damage [35]. Serum S100B level seem to increase in correlation with the clinical findings associated with CO poisoning induced hypoxic brain damage and decrease after oxygen therapy [3]. Agreed with our results, several studies reported that the increased serum S100B level was observed in patients with CO poisoning [4,6,7]. Cakir et al. suggested that the level of S-100β

Please cite this article as: H. Liu, et al., Serum S100B level may be correlated with carbon monoxide poisoning, Int Immunopharmacol (2015), http://dx.doi.org/10.1016/j.intimp.2015.04.027

210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 Q3

6

H. Liu et al. / International Immunopharmacology xxx (2015) xxx–xxx

C

T

E

D

P

R O

O

F

A

U

N

C

O

R

R

E

B

Fig. 6. Sensitivity analysis and funnel plot of publication biases in present meta-analysis investigating the correlation between serum S100B level and carbon monoxide poisoning. A, sensitivity analysis; B, funnel plot of publication biases.

232 233 234 235 236 237 238 239

seems to be quite useful in determining early-period neuronal injury in CO intoxications [31]. Eunjung Park et al. speculated that the blood– brain barrier may have been disrupted by acute CO injury, and they suggested that the degree of brain injury might be reflected in the serum S100B protein level, and thus acute CO poisoning an increased serum S100B protein level might be useful as a predictor of the development of neurological sequelae [36]. The evidence from this study suggested that serum S100B level could be employed as useful markers for

predicting clinical severity and hypoxic brain damage in patients with CO poisoning. In consideration of other related factors may influence the correlation between serum S100B level and CO poisoning, subgroups analysis on the basis of ethnicity and the degree of CO poisoning were performed. The subgroup analyzed the influence of genetic differences of different race on correlation between serum S100B level and CO poisoning in Caucasian and Asian patients with CO poisoning compared with

Please cite this article as: H. Liu, et al., Serum S100B level may be correlated with carbon monoxide poisoning, Int Immunopharmacol (2015), http://dx.doi.org/10.1016/j.intimp.2015.04.027

240 241 242 243 244 245 246 247

H. Liu et al. / International Immunopharmacology xxx (2015) xxx–xxx

265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284

287

The authors have declared that no competing interests exist. Acknowledgments

289 290 Q5 291

This study was funded by the National Natural Science Foundation of China (No. 81273920). We would like to acknowledge the reviewers for their helpful comments on this paper.

292

References

N C O

288

[1] A.K. Erenler, T. Yardan, A. Baydin, M. Gunay, R. Amanvermez, Heart-type fatty acidbinding protein as a potential biomarker of acute carbon monoxide poisoning, Am. J. Emerg. Med. 31 (2013) 1165–1169. [2] X. Sun, H. Xu, X. Meng, J. Qi, Y. Cui, Y. Li, et al., Potential use of hyperoxygenated solution as a treatment strategy for carbon monoxide poisoning, PLoS ONE 8 (2013) e81779. [3] A.Z. Akelma, A. Celik, O. Ozdemir, F. Kavak Akelma, A. Abaci, C.H. Razi, et al., Neuronspecific enolase and S100B protein in children with carbon monoxide poisoning: children are not just small adults, Am. J. Emerg. Med. 31 (2013) 524–528. [4] E. Park, J. Ahn, Y.G. Min, Y.S. Jung, K. Kim, J. Lee, et al., The usefulness of the serum s100b protein for predicting delayed neurological sequelae in acute carbon monoxide poisoning, Clin. Toxicol. (Phila.) 50 (2012) 183–188. [5] S. Shahsavand, A.H. Mohammadpour, R. Rezaee, E. Behravan, R. Sakhtianchi, S.A. Moallem, Effect of erythropoietin on serum brain-derived biomarkers after carbon monoxide poisoning in rats, Iran J. Basic Med. Sci. 15 (2012) 752–758. [6] H. Akdemir, T. Yardan, C. Kati, L. Duran, H. Alacam, Y. Yavuz, et al., The role of S100B protein, neuron-specific enolase, and glial fibrillary acidic protein in the evaluation of hypoxic brain injury in acute carbon monoxide poisoning, Hum. Exp. Toxicol. 33 (2014) 1113–1120.

U

293 294 295 296 297 298 299 Q6 300 301 302 Q7 303 304 305 306 307 308 309 310 311 397

F

O

263 264

R O

261 262

C

259 260

E

257 Q4 258

R

255 256

R

254

P

Competing interests

252 253

[7] T. Ide, Y. Kamijo, A. Ide, K. Yoshimura, T. Nishikawa, K. Soma, et al., Elevated S100B level in cerebrospinal fluid could predict poor outcome of carbon monoxide poisoning, Am. J. Emerg. Med. 30 (2012) 222–225. [8] A.G. Kartal, S. Yilmaz, E. Yaka, M. Pekdemir, H.T. Sarisoy, M.B. Cekmen, et al., Diagnostic value of S100B protein in the differential diagnosis of acute vertigo in the emergency department, Acad. Emerg. Med. 21 (2014) 736–741. [9] A. Rodriguez-Rodriguez, J.J. Egea-Guerrero, A. Leon-Justel, E. Gordillo-Escobar, J. Revuelto-Rey, A. Vilches-Arenas, et al., Role of S100B protein in urine and serum as an early predictor of mortality after severe traumatic brain injury in adults, Clin. Chim. Acta 414 (2012) 228–233. [10] M. Shakeri, A. Mahdkhah, F. Panahi, S100B protein as a post-traumatic biomarker for prediction of brain death in association with patient outcomes, Arch. Trauma Res. 2 (2013) 76–80. [11] T. Gos, M.L. Schroeter, W. Lessel, H.G. Bernstein, H. Dobrowolny, K. Schiltz, et al., S100B-immunopositive astrocytes and oligodendrocytes in the hippocampus are differentially afflicted in unipolar and bipolar depression: a postmortem study, J. Psychiatr. Res. 47 (2013) 1694–1699. [12] Z. Cakir, S. Aslan, Z. Umudum, H. Acemoglu, A. Akoz, S. Turkyilmaz, et al., S-100beta and neuron-specific enolase levels in carbon monoxide-related brain injury, Am. J. Emerg. Med. 28 (2010) 61–67. [13] J. Steiner, B. Bogerts, M.L. Schroeter, H.G. Bernstein, S100B protein in neurodegenerative disorders, Clin. Chem. Lab. Med. 49 (2011) 409–424. [14] A. Stang, Critical evaluation of the Newcastle–Ottawa scale for the assessment of the quality of nonrandomized studies in meta-analyses, Eur. J. Epidemiol. 25 (2010) 603–605. [15] C.K. Lo, D. Mertz, M. Loeb, Newcastle–Ottawa scale: comparing reviewers' to authors' assessments, BMC Med. Res. Methodol. 14 (2014) 45. [16] E. Zintzaras, J.P. Ioannidis, HEGESMA: genome search meta-analysis and heterogeneity testing, Bioinformatics 21 (2005) 3672–3673. [17] E. Zintzaras, J.P. Ioannidis, Heterogeneity testing in meta-analysis of genome searches, Genet. Epidemiol. 28 (2005) 123–137. [18] J.P. Higgins, S.G. Thompson, Quantifying heterogeneity in a meta-analysis, Stat. Med. 21 (2002) 1539–1558. [19] J.A. Sterne, M. Egger, Funnel plots for detecting bias in meta-analysis: guidelines on choice of axis, J. Clin. Epidemiol. 54 (2001) 1046–1055. [20] E.A. Wikstrom, S. Naik, N. Lodha, J.H. Cauraugh, Balance capabilities after lateral ankle trauma and intervention: a meta-analysis, Med. Sci. Sports Exerc. 41 (2009) 1287–1295. [21] F. Song, S. Gilbody, Bias in meta-analysis detected by a simple, graphical test. Increase in studies of publication bias coincided with increasing use of metaanalysis, BMJ 316 (1998) 471. [22] J.L. Peters, A.J. Sutton, D.R. Jones, K.R. Abrams, L. Rushton, Comparison of two methods to detect publication bias in meta-analysis, JAMA 295 (2006) 676–680. [23] L.S. Rasmussen, M.G. Poulsen, M. Christiansen, E.C. Jansen, Biochemical markers for brain damage after carbon monoxide poisoning, Acta Anaesthesiol. Scand. 48 (2004) 469–473. [24] F.X. Li, W.J. Li, C.Z. Liao, X.C. Wang, Plasma s-100 protein levels and its clinical significance in patients with acute carbon monoxide poisoning, J. Med. Theory Pract. (2005) 900–901. [25] X.D. Du, Prognostic value of serum S100B protein in patients with acute carbon monoxide poisoning, SiChuan Med. J. (Si Chuan Yi Xue) 27 (2006) 40–41. [26] Y.H. Cao, G.Y. Zhang, G.C. zHANG, C.L. Ding, R.Y. Li, Significance of serum s-100β protein detection in patients with acute carbon monoxide poisoning, J. Appl. Clin. Pediatr. (Shi Yong Er Ke Lin Chuang Za Zhi) (2007) 1398–1399. [27] J.G. Wang, W.J. Fu, L.F. Peng, Z.X. Chen, L. Liu, S. He, Dynamic changes and significance of serum S100b, NSE and MBP in patients with acute CO poisoning, Int. Med. Health Guid. News 9–11 (2009). [28] T. Yardan, Y. Cevik, O. Donderici, C. Kavalci, F.M. Yilmaz, G. Yilmaz, et al., Elevated serum S100B protein and neuron-specific enolase levels in carbon monoxide poisoning, Am. J. Emerg. Med. 27 (2009) 838–842. [29] G.H. Han, J.K. Wang, R.J. GU, W.Q. Li, P. Zhang, F. Zhang, et al., Serum S100B protein and GFAP levels and their clinical significance in patients with delayed encephalopathy after acute carbon monoxide poisoning, Chinese J. Behav. Med. Brain Sci. 20 (2011) 1107–1110. [30] H. Akdemir, T. Yardan, C. Kati, L. Duran, H. Alacam, Y. Yavuz, et al., The role of S100B protein, neuron-specific enolase, and glial fibrillary acidic protein in the evaluation of hypoxic brain injury in acute carbon monoxide poisoning, Hum. Exp. Toxicol. (2014). [31] Z. Cakir, S. Aslan, Z. Umudum, H. Acemoglu, A. Akoz, S. Turkyilmaz, et al., S-100beta and neuron-specific enolase levels in carbon monoxide-related brain injury, Am. J. Emerg. Med. 28 (2010) 61–67. [32] N.B. Hampson, S.L. Dunn, U.C.C.P.S. Group, Symptoms of carbon monoxide poisoning do not correlate with the initial carboxyhemoglobin level, Undersea Hyperb. Med. 39 (2012) 657–665. [33] T. Ide, Y. Kamijo, A. Ide, K. Yoshimura, T. Nishikawa, K. Soma, et al., Elevated S100B level in cerebrospinal fluid could predict poor outcome of carbon monoxide poisoning, Am. J. Emerg. Med. 30 (2012) 222–225. [34] H. Wolf, S. Frantal, G. Pajenda, J. Leitgeb, K. Sarahrudi, S. Hajdu, Analysis of S100 calcium binding protein B serum levels in different types of traumatic intracranial lesions, J. Neurotrauma (2014). [35] A.Z. Akelma, A. Celik, O. Ozdemir, F. Kavak Akelma, A. Abaci, C.H. Razi, et al., Neuronspecific enolase and S100B protein in children with carbon monoxide poisoning: children are not just small adults, Am. J. Emerg. Med. 31 (2013) 524–528. [36] E. Park, J. Ahn, Y.G. Min, Y.S. Jung, K. Kim, J. Lee, et al., The usefulness of the serum s100b protein for predicting delayed neurological sequelae in acute carbon monoxide poisoning, Clin. Toxicol. (Phila.) 50 (2012) 183–188.

D

286

250 251

T

285

healthy controls. Our result revealed that a higher serum S100B level was detected in Caucasian and Asian patients with CO poisoning than that in normal controls, which suggests that genetic differences had a minimum influence on the correlation between serum S100B level and CO poisoning and serum S100B could become a new effective marker assessing the clinical status of patients with CO poisoning regardless of ethnicity differences. On the other hand, according to the degree of CO poisoning, our results demonstrated that the serum level of S100B in patients with CO poisoning of mild, moderate and severe was significantly higher than that of the healthy controls. Several limitations should be noted in our meta-analysis. First, some of the included literatures may have a relatively small sample size, which may limit the power to identify the results of this metaanalysis. Second, this analysis may have publication and reporting bias. While we attempted to use several Chinese and English databases as complete as possible to minify the publication bias, we were not able to take unpublished papers and abstracts into consideration entirely. Third, the selective bias and language bias may exist in this metaanalysis for the included studies were only involved papers published with languages in English or Chinese. Fourth, the crude classification criteria of ethnicity into “Caucasian” or “Asian”, may lead the study prone to bias. The studies with ethnicity of Asian were all conducted in China and almost the studies grouped into Caucasian were performed in Turkey, as well as the Chinese occupied the big proportion, to achieve the possible ethnic differences in correlation between serum S100B level and CO poisoning, more investigations from various populations are needed to further identify the present results. Fifth, the original data in the included studies of timing and age were not sufficient, and thus the effect of timing on correlation and difference in correlation in adults or children were not presented clearly in the meta-analysis, which may decrease the value of the study. Finally, although we attentively evaluated the included studies, there are no sufficient data showed about delayed complications which decrease the value of the study. To conclude, the increased serum S100B level was associated with CO poisoning, and therefore Serum S100B could be a useful marker in the assessment of clinical status in CO poisoning. Well-designed further researches with larger sample sizes will contribute to the reliability of the results.

E

248 249

7

Please cite this article as: H. Liu, et al., Serum S100B level may be correlated with carbon monoxide poisoning, Int Immunopharmacol (2015), http://dx.doi.org/10.1016/j.intimp.2015.04.027

312 Q8 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 Q9 379 380 381 382 383 384 385 386 387 388 389 390 Q10 391 392 393 394 395 396

Serum S100B level may be correlated with carbon monoxide poisoning.

To investigate the correlation between serum S100B level and carbon monoxide (CO) poisoning by meta-analysis...
2MB Sizes 0 Downloads 8 Views