Neurology

S100B serum level: A relevant biomarker for the management of non-traumatic headaches in emergency care?

a b s t r a c t

Background: The diagnostic of primary or secondary headaches in emergency units is mostly based on brain im- aging, which is expensive and sometimes hardly accessible. An increase in serum S100B protein has already been found in several neurological conditions inducing brain damage. The objective of this study was to assess the diagnostic performance of S100B serum assay to distinguish primary and secondary headaches among patients with non-traumatic headaches in the emergency department.

Methods: This was a Phase 2, prospective, monocentric diagnostic study. Eighty-one adult patients with non- traumatic headaches in the emergency department were included. In addition to the usual management, a blood assay of the S100B protein was performed in the emergency department, as well as a brain MRI between 48 and 96 h if not performed during the initial management. The primary or secondary headache diagnosis was made at one month by an expert committee, blindly of the results of the S100B assay. The primary outcome was the blood assay of the S100B protein.

Results: There was 63 patients for analysis in the primary headache group and 17 in the secondary headache group. The S100B protein assay was significantly higher in secondary headaches than primary headaches, with an AUC of the ROC curve of 0.67. The optimal threshold of 0.06 ug.L-1 allowed to obtain those diagnostic charac- teristics: sensitivity 75% [48; 93], specificity 62% [48; 74], PPV 35% [20; 54] and NPV 90% [76; 97]. The association between the S100B protein level and the onset of pain was significantly higher for patients with headaches <3 h. Conclusion: The assay of the S100B protein could be useful in the management of this pathology in emergencies. Future studies taking into account dosing time and etiologies could be conducted in order to refine its use in practice.

(C) 2023

  1. Introduction

About 15% of the population has had at least one Severe headache in the past 3 months [1]. Headaches represent between 1 and 3% of the reason for consultation in emergency units [2]. More than 70% of these patients return home with a diagnosis of benign primary headache [3]. Secondary headaches, defined by the International Head Society as cephalgias symptomatic of a local or general cause (subarachnoid hem- orrhage, meningitis, tumors, etc.), may affect the patient’s life or function prognosis. The challenge in emergency units is therefore to be able to identify a secondary headache in order to quickly implement

* Corresponding author.

E-mail address: [email protected] (L. Grau-Mercier).

an adapted treatment. It requires additional examinations such as com- puted tomography, MRI or lumbar puncture [4,5].

Protein S100B is an intracellular protein secreted essentially by glial cells of the central nervous system [6]. Its plasma half-life is estimated to be less than 2 h [7]. The interest of the S100B serum level in Mild head trauma has been widely demonstrated in both children [8] and adults [9], showing an association between elevated serum levels and the pres- ence of cerebral damage in computed tomography, if sampling timeframe was less than 3 h [10]. For vascular pathologies, studies sug- gest that higher serum levels of S100B protein may reveal more severe intracerebral damage in stroke patients [11,12]. This also applies in patients with subarachnoid bleeding [13]. Recent studies suggest a link between post-cardiac arrest S100B protein levels in the blood and the Neurological prognosis of patients [14,15]. There is also a direct

https://doi.org/10.1016/j.ajem.2023.03.036

0735-6757/(C) 2023

link between trisomy 21 and the S100B protein since two of its coding genes are on chromosome 21 [16]. In Alzheimer’s disease, its expression seems to differ according to the stages of the disease [17]. High blood levels of S100B protein also seemed to be associated with malignant melanoma [18], some brain tumors [19] or brain metastases for non- small cell Lung cancer [20]. For migraine headaches, studies have been conducted on small samples and their results are discordant [21,22].

Based on the current state of knowledge, we hypothesize that the S100B protein level would be significantly higher in patients with a sec- ondary headache than in patients with a primary headache. Confirma- tion of this hypothesis would improve the diagnostic strategy by reducing the use of imaging for patients with a non-traumatic headache in emergency units.

The main objective of this study was to study the diagnostic perfor- mance of S100B serum assay to distinguish primary and secondary headaches among patients with non-traumatic headaches in the emer- gency department.

  1. Materials and methods
    1. Study design

This was a prospective monocentric diagnostic pilot study. In order to determine the discriminating capacity of the serum assay of the S100B protein in the differential diagnosis between primary and sec- ondary headache, its concentration (in ug.L-1) was compared between two groups of headache patients presenting to the emergency depart- ment: secondary headache group and primary headache group.

The serum assay of the S100B protein was performed in a blood sam- ple in addition to the usual management in the emergency unit. Symp- toms and characteristics of the patient, including phototype following the Fitzpatrick Skin Phototype Classification [23], were collected. Phototype was recorded because some phototypes have higher serum S\\100B concentrations than others [24]. This is physiologically explained by an expression of this protein by the melanocytes [25].

If not performed in the emergency unit, an MRI was performed be- tween 48 and 96 h after inclusion.

At 1 month the information concerning medical data, modalities of management, procedures practiced (lumbar puncture, brain imaging), length of stays in the emergency unit and hospital were collected in the patient record. The patient was then contacted by the investigator by telephone in order to collect the patient’s health status (onset of sec- ondary headache), events during the month, and mortality.

      1. Participants of the study

Patients consulting in the emergency unit for a non-traumatic head- ache with a pain rated over 3/10 on a visual analog scale were included. Signature of the subject’s free and informed consent was collected.

Exclusion criteria were: age under 18 or over 60 years old, preg- nancy, absence of written consent, inclusion in another current study, contraindication to MRI, Alzheimer’s disease, Multiple sclerosis, trisomy 21, melanoma, Creutzfledt-Jakob disease.

      1. standard protocol approvals, regi”>Standard protocol approvals, registrations, and patient consents

The study was approved by the local Ethics Comittee (Comite de Pro- tection des Personnes, CPP Sud Mediterrannee III, Nimes, France, RCB: 2016-A00013-48). All experimental methods were carried out in accor- dance with the ethical guidelines determined by the French National Ministry of Health, Labour and Welfare and by the Declaration of Helsinki. All participants provided written informed consent before en- tering the study. The protocol was registered on clinicaltrials.gov (NCT 02683304).

      1. Outcomes

The main objective of this study is to study the diagnostic perfor- mance of the serum assay of the S100B protein in order to estimate

the discriminating value of the S100B protein in the differential diagno- sis of primary and secondary headaches, among patients with non- traumatic headaches in the emergency unit.

The secondary objectives of this study were to:

– assess the association between S100B protein level and the time from symptom onset to the blood sampling (less or greater than 3 h).

– assess the association between S100B protein level and 28-days mortality.

– evaluate the association between the S100B protein level and: aver- age length of stay in the emergency department, performing a lum- bar puncture, performing brain imaging, average length of stay in the hospital.

      1. S100B protein analysis

The primary outcome in this study is the serum concentration of the S100B protein. A 10 mL blood sample was collected in addition to the usual management after informed consent of patients. The samples were centrifuged for 10 min at 2000g at Room temperature and then aliquoted in 500 uL fractions, stored at -80 ?C until use. This assay was performed using an enzyme-linked immunosorbent assay (ELISA) method using the Liaison reagent (DiaSorin(R), Saluggia(R), Italy(R)).

      1. Blinding

The reference diagnosis of primary or secondary headache was es- tablished subsequently according to clinical and imaging data, blindly from the S100B protein level, by an expert committee composed of a neurologist, a radiologist and an emergency physician.

      1. Sample size

In the study by Bouvier et al. [26], patients with minor head trauma had an average S100B protein level of 0.28 ug.L-1 for those without CT lesions and 0.88 ug.L-1 for those with CT lesions, with a common stan- dard deviation of 0.74 ug.L-1. To show a similar size of effect, a popula- tion of 80 subjects was calculated as necessary for our study, assuming an equal distribution between the two groups, to reach a power of 95% with a bilateral alpha risk of 0.05.

    1. Statistical analysis

Quantitative data were expressed as means with standard devia- tions or median with 25th and 75th percentiles according to distribu- tion. Qualitative variables were expressed as frequency with percentages. Comparisons of continuous variables between the groups were performed using the Student t-test or Wilcoxon-Mann-Whitney test according to the distribution. Categorical variables were compared between groups by ?2 or Fisher’s exact test.

The serum assay of the S100B protein was compared between the

two groups using a Wilcoxon-Mann-Whitney test. A logistic regression model was performed to assess the performance of S100B protein dos- age to predict a diagnosis of secondary headache using a receiver oper- ating characteristic (ROC) curve with an estimate of its area under the curve (AUC) and its 95% confidence interval (CI). An optimal threshold was sought with Youden index (sensitivity + specificity – 1) maximiza- tion. The robustness of the results was checked by bootstrapping (bootstrapping (500 resamples equal to the size of the original popula- tion were used). Univariate tests were performed for the co-variables of interest to determine whether potential factors were likely to bias the estimation of the relationship between the serum S100B assay results and the final diagnosis selected. A logistic model was used to consider confounding factors. The statistical significance threshold used to in- clude the variables in the model was a p < 0.20. The number of variables considered was limited to 1 variable per 10 patients per group based on current recommendations [27]. The various ROC curves obtained were compared using a non-parametric test [28].

The association between S100B protein level and qualitative vari- ables was tested by Mann-Whitney tests. The correlation between serum S100B protein concentration and quantitative variables was

Table 1

Population characteristics. BMI: body mass index. SAP = systolic arterial pressure; DAP = diastolic arterial pressure

assessed by the spearman correlation coefficient.

All statistical analyses were conducted using SAS (9.4; SAS Inc., Cary, NC, USA). All p values were two tailed and a p value <0.05 was consid-

Primary headache group (n = 63)

Secondary p headache group

(n = 17)

ered significant.

  1. Results
    1. Population

Eighty-one patients were included from March 2017 to March 2020 in the emergency department of Nimes university hospital. The patient flow chart is presented in Fig. 1. Patients were all successfully reached by phone for the one-month follow-up. The characteristics of the study population and their clinical presentation are summarized in Table 1. Final diagnosis established by the expert committee for patients of each group, according to the International Head Society classification

[29] are presented in Table 2.

    1. Primary outcome

The S100B median assay was 0.05 [0.04; 0.06] ug.L-1 for patients with primary headaches and 0.07 [0.05; 0.11] ug.L-1 for patients with secondary headaches (p = 0.0344) (Fig. 2). The area under the curve

Age (years) 35 [25; 43] 46 [34; 50] 0.03

Female 44 (70%) 9 (53%) 0.20

BMI (kg.m-2) 24 [21; 26] 25 [22; 27] 0.48

Phototype

type 1 3 (5%) 1(6%) 0.97

type 2 10 (16%) 3 (18%)

type 3 18 (29%) 4 (24%)

type 4 23 (37%) 6 (35%)

type 5

4 (6%)

2 (12%)

type 6

Patient history

5 (8%)

1 (6%)

Stroke

1 (2%)

3 (18%)

0.03

Vasculitis

0

3 (18%)

0.01

Migraine

21 (33%)

2 (12%)

0.13

Heart rate (bpm)

81 [75; 90]

78 [66; 85]

0.24

SAP (mmHg)

132 [119; 143]

132 [127; 140]

0.54

DAP (mmHg)

84 [73; 91]

79 [76; 95]

0.22

Temperature (?C)

36.7 [36.3; 37]

36.6 [36.1; 37]

0.61

Visual analog pain scale

8 [6; 9]

7 [5; 9]

0.31

Time to peak pain

<1 min

26 (41%)

8 (47%)

0.79

[1 min; 1 h]

30 (48%)

8 (47%)

>1 h

duration of headache when

7 (11%)

5 [2; 13]

1 (6%)

2 [1; 4]

0.03

consulting (hours)

Duration of headache when blood

8.5 [5.8; 16]

3.9 [2.2; 6.5]

sampling for S100B assay (hours)

threshold value of 0.06 ug.L-1(Youden index: 0.37). At this threshold, sensitivity was 75% [48; 93], specificity 62% [48; 74], PPV 35% [20; 54], NPV 90% [76; 97]. Using a Bootstrap method, the AUC obtained was

of ROC curve of S100B protein assay to predict the secondary character-

Stiff neck

10 (16%)

8 (47%)

0.02

istics of a headache was 0.68 [0.50; 0.85] (Fig. 3); with an optimal

Photo-phonophobia

34 (54%)

6 (35%)

0.17

0.67 [0.39; 0.85].

As age is a factor that could potentially influence blood levels of S100B protein, it was selected for multivariable logistic regression. The combination of age and S100B protein level minimized the Akaike Information Criterion (AIC, best model fitted dataset) with an estimated AUC of 0.73 [0.60; 0.87]. The parameters of the multi- variable logistic regression model allowed the calculation of a com- posite score combining the S100B result and the age were:

-4.5626 + 4.9756*S100B (ug.L-1) + 0.0722*Age. The ROC curve

of the composite score is shown in Fig. 4. Using a Bootstrap method, the AUC obtained was 0.73 [0.61; 0.86]. The Youden index for the composite score did not provide a single optimal threshold. At

Hemicranial 28 (44%) 4 (24%) 0.12

Temporal 34 (54%) 5 (29%) 0.07

prodromal symptoms

13 (21%)

0

0.06

Crush pain

30 (48%)

15 (94%)

0.01

Pain location

Occipital 15 (24%) 10 (59%) 0.01

optimum threshold values the composite score didn’t show an im- provement in the estimated sensitivity.

    1. Secondary outcomes

The median S100B assay for patients with headaches less than 3 h at the time of blood sampling was 0.06 [0.06; 0.09] ug.L-1, and 0.05 [0.04; 0.07] for patients with headaches for 3 h or more (p = 0.046).

Image of Fig. 1

Fig. 1. Flow chart.

Table 2

Final diagnosis established by the expert committee for patients of each group

Primary headache group (n = 63)

Migraine

22 (35%)

tension-type headache

40 (63%)

Cluster headache

1 (2%)

Secondary headache group (n = 17)

Transitory ischemic accident

2 (12%)

Hypertension

1 (6%)

Hemorrhagic stroke

6 (35%)

Ischemic stroke

3 (17%)

Sinusitis

2 (12%)

Epilepsy

1 (6%)

Meningitis

1 (6%)

Anxiety disorder

1 (6%)

No patient died during the study period, 28-day mortality based on S100B assay could not be assessed.

The correlation coefficient between S100B test result and length of stay in the emergency unit was estimated at -0.15 [-0.37; 0.08] (p = 0.19). The correlation coefficient between S100B test result and length of stay in the hospital was estimated at 0,01 [-0.22; 0.24] (p = 0.92).

The median S100B level in patients for which a lumbar puncture was performed, or not, was the exact same: 0.05 [0.04; 0.07] ug.L-1 (p = 0.68). The median S100B level in patients for which a brain imaging was performed, excepted the one planned for the protocol, was 0.05 [0.04; 0.07] ug.L-1 and 0.05 [0.03; 0.06] ug.L-1 for other patients

(p = 0.15).

  1. Discussion

To our knowledge, this pilot study is the first to have studied the di- agnostic performance of the serum assay of the S100B protein in order to discriminate, in patients presenting for non-traumatic headaches in the emergency department, primary and secondary etiologies.

Image of Fig. 3

Fig. 3. ROC curve of the diagnostic value of the S100B protein assay to predict the second- ary characteristics of a headache. AUC = area under the curve.

In this study population, the S100B concentration was significantly higher for patients with secondary headaches compared to patients with primary headaches (0.07 [0.05;0.11] ug.L-1 vs 0.05 [0.04;0.06] ug.L-1; p = 0.03). The AUC of the ROC curve from this sample was 0.67, which corresponds to a low diagnostic value [30]. The AUC ob- tained after resampling with Boostrap method was similar. The Youden index identified a threshold of 0.06 ug.L-1 for which the diagnostic per- formance of the plasma concentration of the S100B protein appeared to be maximum with a sensitivity of 75% [48%; 93%], a specificity of 62% [48%; 74%], a PPV of 35% [20%; 54%] and a NPV of 90% [76%; 97%]. Despite

Image of Fig. 2

Fig. 2. S100B protein assay in the primary headache and secondary headache groups.

Image of Fig. 4

Fig. 4. ROC curve for the diagnosis of secondary headache with the composite score including the age and the S100B protein assay.

a low diagnostic value, the assay of the S100B protein seems relevant to eliminate the diagnosis of secondary headache with a 90% NPV. Analysis of a composite score combining the S100B protein assay with age did not provide a better diagnostic performance.

The threshold of 0.06 ug.L-1 is significantly lower than that used in studies for mild to moderate head injuries, ranging from 0.10 ug.L-1

[16] to 0.20 ug.L-1 [31], for superior diagnostic performance with sensi- tivity >90%. This difference may be explained by the lack of power of our study induced by the difference between the number of patients in the primary and secondary headache group, due to the lower preva- lence of secondary headache in the emergency department [3]. One study suggests that the median Blood concentration of S100B protein in Healthy adults is 0.05 ug.L-1, regardless of age and sex [32]. Other studies find that there is a correlation between blood levels of S100B protein and age up to 20 years, and then it finally disappears between 20 and 70 years [33]. The levels of S100B protein observed in our study are close to normal and could be interpreted as physiological variations.

If we consider secondary non-traumatic headaches according to their etiologies (Table 2), we find a diversity of data in the literature on the correlation with serum levels of S100B protein. There was an association between S100B protein levels and the following patholo- gies: ischemic stroke [12], subarachnoid hemorrhage [13], Horton vascular episodic headache [33], certain primary tumors [19] or metas- tatic [20], and carbon monoxide poisoning [34]. In the case of meningi- tis, one study found an association between the S100B protein concentration in cerebrospinal fluid and Bacterial meningitis [35].). A recent meta-analysis including nine studies showed that S100B levels were indeed significantly increased in migraineurs than in the healthy

control group, although the results of previous clinical studies on this topic were not consistent [36]. The interpretation of the s100b assay therefore seems difficult in these patients. There is insufficient data in the literature regarding serum S100B protein assay for other etiologies. Data from the literature suggest that serum S100B protein levels in- crease as a result of the passage of the blood-brain barrier following a lysis of astrocyte cells in the central nervous system [17]. The Interna- tional Headache Society also classifies headache attributed to disorder of homoeostasis and headache attributed to psychiatric disorder as sec- ondary headaches [29], which are not brain damaging processes. It is therefore likely that the level of S100B protein is influenced by the eti- ology of secondary headache, which complicates the interpretation of its dosage for the diagnosis of all causes of secondary headaches.

The association between the S100B protein level and the onset of pain is significantly higher for patients with headaches that last less than 3 h at the time of collection compared to patients with headaches that last more than 3 h (0.06 [0.06;0.09] ug.L-1 vs 0.05 [0.04;0.07] ug. L-1; p = 0.0462). This result is consistent with the maximum time of 3 h to perform the serum dose found in studies already conducted in mild to moderate head injuries [37]. This is probably due to the short half-life of this molecule [7]. In this study, the S100B protein was assayed more than 3 h after the onset of pain in 63 (85%) subjects, which may have affected the study results. This delay should be taken into account in future studies in the interpretation of the S100B protein blood test results. However, this delay will probably vary depending on the pathological mechanism involved.

The main limitation of this study is the small size of the secondary headache group. The calculation of the number of subjects required sug- gested the inclusion of 80 subjects assuming an equal distribution

between the two groups. As the diagnosis of primary or secondary headache was established a posteriori of inclusion, the distribution be- tween the two groups was not balanced. This is also explained by the higher prevalence of primary headaches in emergency units [3]. The analysis of only 16 patients with secondary headaches did not allow suf- ficient power to estimate a discriminating value of the S100B protein in their diagnosis. This was a monocentric study, although the population appeared to be representative of the general population [3]. Finally, the recruitment time for this study was rather long (3 years from January 2017 to January 2020) for a yet frequent pathology in the emergency department [2], which could have induced a selection bias. This under- lines the difficulty of carrying out studies in emergency structures.

  1. Conclusion

The level of S100B protein is significantly higher in secondary head- aches than in primary headaches, but its diagnostic value remains low, with an AUC of the ROC curve from this sample of 0.67. However, the optimal threshold of 0.06 ug.L-1 allowed to obtain a VPN around 90%. The assay of this protein could therefore be a relevant biomarker to ex- clude secondary causes of headache. The search for an optimal thresh- old will have to be evaluated in other studies with more power in order to avoid a selection bias due to the low prevalence of secondary headaches in emergency units.

The association between the S100B protein level and the onset of pain was significantly higher for patients with headaches less than 3 h. Future studies should therefore take into account this delay in the determination of the S100B protein for the interpretation of the results. Finally, further studies to investigate an association between S100B protein levels and secondary headaches according to specific etiologies are necessary to develop the use of this protein assay in emergency

units.

CRediT authorship contribution statement

Laura Grau-Mercier: Writing – original draft. Romain Genre Grandpierre: Writing – review & editing, Supervision, Methodology, Investigation, Funding acquisition, Conceptualization. Sandrine Alonso: Methodology, Formal analysis. Alice Savey: Writing – review & editing. Anne Le Floch: Writing – review & editing. Fabien de Oliveira: Writing – review & editing. Thibaud Masia: Investigation. Nathalie Jory: Investigation. Fabien Coisy: Writing – review & editing. Pierre-Geraud Claret: Writing – review & editing, Validation, Funding acquisition.

Declaration of Competing Interest

None.

References

  1. Burch RC, Loder S, Loder E, Smitherman TA. The prevalence and burden of migraine and severe headache in the United States: updated statistics from government health surveillance studies. Headache J Head Face Pain. 2015;55(1):21-34.
  2. Goldstein J, Camargo C, Pelletier A, Edlow J. Headache in United States emergency departments: demographics, work-up and frequency of pathological diagnoses. Cephalalgia. 2006;26(6):684-90.
  3. Pari E, Rinaldi F, Gipponi S, Venturelli E, Liberini P, Rao R, et al. Management of head- ache disorders in the emergency department setting. Neurol Sci. 2015;36(7): 1153-60.
  4. Moisset X, Mawet J, Guegan-Massardier E, Bozzolo E, Gilard V, Tollard E, et al. Recommandations pour la prise en charge d’une cephalee en urgence. Douleurs. Eval Diagn Trait. 2018;19(1):4-16.
  5. Bo SH, Davidsen EM, Gulbrandsen P, Dietrichs E. acute headache: a prospective diagnostic work-up of patients admitted to a general hospital. Eur J Neurol. 2008; 15(12):1293-9.
  6. Benabdesselam O, Ziani S, Beaudeux J. La proteine S-100B : biomarqueur de diagnos- tic et de suivi de lesions cerebrales aigues. Spectra Anal. 2007;26:40-5.
  7. Beaudeux JL, Soler C, Foglietti MJ. Physiopathologie de la proteine S-100 ? : interet de son dosage en biologie clinique. Immuno-Anal Biol Spec. 2002;17(5):280-6.
  8. Oris C, Pereira B, Durif J, Simon-Pimmel J, Castellani C, Manzano S, et al. The bio- marker S100B and mild traumatic brain injury: a meta-analysis. Pediatrics. 2018; 141(6):e20180037.
  9. Leidel BA, Bogner V, Zock M, Kanz K-G. Das serologische Protein S100B: Stellenwert in der Notfalldiagnostik Erwachsener mit Verdacht auf leichtes Schadel-Hirn- Trauma – eine Metaanalyse. Unfallchirurg. 2012;115(10):903-12.
  10. Grandpierre RG, Bobbia X, de La Coussaye J-E, Claret P-G. Interet clinique des con- centrations seriques de la proteine S100? dans l’evaluation des patients traumatises craniens. Ann Fr Medecine D’urgence. 2018;8(3):172-7.
  11. Ye H, Wang L, Yang X-K, Fan L-P, Wang Y-G, Guo L. serum S100B levels may be as- sociated with cerebral infarction: a meta-analysis. J Neurol Sci. 2015;348(1-2):81-8.
  12. Missler U, Wiesmann M, Friedrich C, Kaps M. S-100 protein and neuron-specific eno- lase concentrations in blood as indicators of infarction volume and prognosis in acute ischemic stroke. Stroke. 1997;28(10):1956-60.
  13. Chong ZZ. S100B raises the alert in subarachnoid hemorrhage. Rev Neurosci. 2016; 27(7):745-59.
  14. Deye N, Nguyen P, Vodovar N, Sadoune M, Collet C, Voicu S, et al. Protein S100B as a reliable tool for early prognostication after cardiac arrest. Resuscitation. 2020;156: 251-9.
  15. Wang C-H, Chang W-T, Su K-I, Huang C-H, Tsai M-S, Chou E, et al. Neuroprognostic accuracy of blood biomarkers for post-cardiac arrest patients: a systematic review and meta-analysis. Resuscitation. 2020;148:108-17.
  16. Mrak RE, Griffin WST. Trisomy 21 and the brain. J Neuropathol Exp Neurol. 2004;63 (7):679-85.
  17. Rothermundt M, Peters M, Prehn JHM, Arolt V. S100B in brain damage and neurode- generation. Microsc Res Tech. 2003;60(6):614-32.
  18. Nakajima T, Watanabe S, Sato Y, Kameya T, Shimosato Y, Ishihara K. Immunohisto- chemical demonstration of S 100 protein in malignant melanoma and pigmented nevus, and its diagnostic application. Cancer. 1982;50(5):912-8.
  19. Syeda T, Rizvi HA, Hadi SM, Hashim M, AS, Khan MH.. Interrelationship of serum S100B and P53 in brain tumour patients. JPMA J Pak Med Assoc. 2016;66(6):748-50.
  20. Pang X, Min J, Liu L, Liu Y, Ma N, Zhang H. S100B protein as a possible participant in the brain metastasis of NSCLC. Med Oncol. 2012;29(4):2626-32.
  21. Teepker M, Munk K, Mylius V, Haag A, Moller JC, Oertel WH, et al. Serum concentra- tions of s100b and NSE in migraine. Headache J Head Face Pain. 2009;49(2):245-52.
  22. Yilmaz N, Karaali K, Ozdem S, Turkay M, Unal A, Dora B. Elevated S100B and neuron specific enolase levels in patients with migraine-without Aura: evidence for neuro- degeneration? Cell Mol Neurobiol. 2011;31(4):579-85.
  23. Roberts WE. Skin type Classification systems old and new. Dermatol Clin. 2009;27 (4):529-33.
  24. Gannon JM, Kelly DL, Besch A, Thakur T, Khurana N, Shurin MR, et al. Racial differ- ences in S100b levels in persons with schizophrenia. Psychiatry Q. 2020;91(1): 137-45. https://doi.org/10.1007/s11126-019-09687-4. [PMID: 31786729].
  25. Cheong KA, Noh M, Kim CH, Lee AY. S100B as a potential biomarker for the detection of cytotoxicity of melanocytes. Exp Dermatol. 2014;23(3):165-71. https://doi.org/ 10.1111/exd.12332. [PMID: 24451020].
  26. Bouvier D, Oddoze C, Ben Haim D, Moustafa F, Legrand A, Alazia M, et al. Interet du dosage serique de la proteine S100B dans la prise en charge du patient apres traumatisme cranien leger. Ann Biol Clin (Paris). 2009;67(4):425-31.
  27. Harrell FE, Lee KL, Mark DB. Multivariable prognostic models: issues in developing models, evaluating assumptions and adequacy, and measuring and reducing errors. Statist Med. 1996;15(4):361-87.
  28. DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Bio- metrics sept. 1988;44(3):837-45.
  29. Headache Classification Committee of the International Headache Society (IHS) The International Classification of Headache Disorders. Cephalalgia3rd ed.. , 38(1); 2018; 1-211. https://doi.org/10.1177/0333102417738202 (PMID: 29368949).
  30. Swets JA. Measuring the accuracy of diagnostic systems. Science. 1988;240(4857): 1285-93.
  31. Heidari K, Vafaee A, Rastekenari AM, Taghizadeh M, Shad EG, Eley R, et al. S100B protein as a screening tool for computed tomography findings after mild traumatic brain injury: systematic review and meta-analysis. Brain Inj. 2015;29(10):1146-57.
  32. Wiesmann M, Missler U, Gottmann D, Gehring S. Plasma S-100b protein concentra- tion in healthy adults is age- and sex-independent. Clin Chem Mai. 1998;44(5): 1056-8.
  33. Portela LV, Tort AB, Schaf DV, Ribeiro L, Nora DB, Walz R, et al. The serum S100B con- centration is age dependent. Clin Chem. 2002;48(6):950-2.
  34. Liu H, Zhang Y, Ren Y-B, Kang J, Xing J, Qi Q-H, et al. Serum S100B level may be cor- related with carbon monoxide poisoning. Int Immunopharmacol juill. 2015;27(1): 69-75.
  35. Jung K, Goerdt C, Lange P, Blocher J, Djukic M, Gerber J, et al. The use of S100B and tau protein concentrations in the cerebrospinal fluid for the differential diagnosis of bacterial meningitis: a retrospective analysis. Eur Neurol. 2011;66(3):128-32.
  36. Chu C, Zhong R, Cai M, Li N, Lin W. Elevated blood S100B levels in patients with mi- graine: a systematic review and Meta-analysis. Front Neurol. 2022;14(13):914051.
  37. Unden J, Romner B. Can low serum levels of S100B predict normal CT findings after Minor head injury in adults?: an evidence-based review and meta-analysis. J Head Trauma Rehabil juill. 2010;25(4):228-40.