Article, Neurology

Investigation of UCH-L1 levels in ischemic stroke, intracranial hemorrhage and metabolic disorder induced impaired consciousness

a b s t r a c t

Objective: We aimed to determine the levels of ubiquitin C-terminal hydrolase-L1 (UCH-L1) in patients admitted to the emergency department with Impaired consciousness due to metabolic or neurological reasons.

Materials – methods: The study included 80 patients with ischemic stroke (IS), 40 patients with intracranial hem- orrhage (ICH), 80 patients with Metabolic disorder induced impaired consciousness (MDIC) and 40 healthy con- trols.

Results: The levels of UCH-L1 [median (IQR)] were as follows: 5.59 ng/mL (3.90-9.37) in IS, 5.44 ng/ml (4.01- 13.98) in ICH, 3.34 ng/ml (2.29-5.88) in MDIC and 3.94 ng/ml (3.31-7.95) in healthy volunteers. Significantly higher levels were detected in IS and ICH than in MDIC and healthy volunteers. In ROC curve analysis, we detect- ed 63.75% sensitivity and 62.5% specificity (AUC = 0.626, p b 0.0199, 95% CI: 0.533-0.713) with a cutoff value of

4.336 ng/ml for IS and 75% sensitivity and 55% specificity (AUC = 0.664, p b 0.0071, 95% CI: 0.549-0.766) with a

cut-off value of 4.036 ng/ml for ICH. However, the sensitivity and specificity for MDIC was 36.25% and 77.5%, re- spectively, with a cut-off value of 3.256 ng/ml (AUC = 0.525, p = 0.6521, 95% CI: 0.432-0.617). UCH-L1 levels were found to increase significantly with increasing time between the onset of symptoms and blood sampling (r = 0.345, p b 0.001). However, no correlation was found between UCH-L1 levels and age (r = 0.014, p = 0.833), GCS (r = -0.115, p = 0.074), mRS (r = 0.063, p = 0.475) and NIHSS (r = 0.056, p = 0.520).

Conclusion: In this study, we detected significantly higher levels of UCH-L1 in patients with IS and ICH compared to patients with MDIC and healthy volunteers.

Introduction

The impaired consciousness is defined as alteration in the perception and evaluation of a stimulus, and reaction to that environmental stimulus. It constitutes approximately 4-10% of annual emergency department visits [1]. Due to greater number of patients and being a life-threatening situation, the rapid discrimination of structural and non-structural causes is very important in terms of determining the appropriate therapy and patient outcomes [2]. In particular, potential biomarker studies have been the focus of attention in human and animal studies in determining the prognosis and diagnosis of intracrani- al diseases and metabolic disorders.

Ubiquitin C-terminal hydrolase-L1 (UCH-L1) is also known as neu- ron-specific protein gene product 9.5 (PGP 9.5) and was previously used as a histological marker for neurons due to being present in excess

* Corresponding author at: Firat University, School of Medicine, Department of Emergency Medicine, 23200 Elazig, Turkey.

E-mail address: [email protected] (M. Yilmaz).

amounts in neurons and having a specific release. UCH-L1 is found al- most in all neurons and comprises approximately 1-5% of total soluble brain proteins. In addition to the brain, UCH-L1 is released from the pe- ripheral nervous system, including sensory and motor nerves, at high rates [3-5]. UCH-L1 is involved in the ubiquitination of proteins, thus play an important role in the removal of oxidized or excess proteins in normal or pathological conditions [6]. The levels of UCH-L1 have been investigated particularly in intracranial diseases [7]. In animal studies, UCH-L1 was shown to play an important role in maintaining the struc- ture and function of neuromuscular junction, and a significant reduction in neurotransmitter release from motor nerve endings and progressive degeneration in the presynaptic nerves have been demonstrated in the absence of UCH-L1 [8]. UCH-L1 levels can be detected both in the ce- rebrospinal fluid (CSF) and in the serum [9].

The UCH-L1 level determination studies in the literature include

chronic degenerative diseases [10] such as Parkinson’s, Alzheimer’s, Huntington disease, and acute intracranial diseases such as trauma, stroke, Epileptic seizures [9], carbon monoxide poisoning [11]. In animal and human studies on intracranial hemorrhages (ICH), levels of UCH-L1

http://dx.doi.org/10.1016/j.ajem.2017.06.032

1896 I. Yigit et al. / American Journal of Emergency Medicine 35 (2017) 1895–1898

Table 1

The characteristics and Vital parameters of study participants.

	IS	ICH	MDIC	Control
n (male/female)	80 (37/43)	40 (22/18)	80 (41/39)	40 (22/18)
Age (mean +- SD)	72.06 +- 13.01	59.25 +- 20.31	67.72 +- 17.60	61.50 +- 13.68
GCS median (minimum-maximum)	13 (6-14)	13 (3-14)	14 (4-14)	15 (15-15)
Time (mean +- SD)^a	5.82 +- 1.78	4.85 +- 1.12	4.16 +- 1.72	–
UCHL1 (ng/mL) median (IQR)	5.59 (3.90-9.37)abc	5.44 (4.01-13.98)de	3.34 (2.29-5.88)f	3.94 (3.31-7.95)

IS: ischemic stroke, ICH: intracranial hemorrhage, MDIC: metabolic disorder induced impaired consciousness, SD: standard deviation, GCS: Glasgow coma scale, UCH-L1: ubiquitin C-ter- minal hydrolase-L1, IQR: interquartile range.

IS comparison with (a) ICH p = 0.468, (b) MDIC p = 0.004, (c) Control p = 0.025. ICH comparison with (d) MDIC p = 0.002, (e) Control p = 0.012.

MDIC comparison with (f) Control p = 0.656.

^a Time: the period between the onset of symptoms and blood sampling.

were reported to be increased [12] or unchanged [13]. However, al- though there are studies [13,14] showing significantly increased UCH- L1 levels in experimental animal models of ischemic stroke (IS), no sig- nificant increase was found in IS in humans [15].

Computed tomography (CT) and diffusion-weighted MR imaging are commonly used Imaging techniques in clinical practice in the differ- ential diagnosis of impaired consciousness in the emergency depart- ment [16]. cranial CT scans are used in addition to clinical and laboratory findings in the diagnosis of patients with metabolic disorder induced impaired consciousness (MDIC) for the exclusion of possible in- tracranial pathology. In our study, we aimed to determine the availabil- ity of UCH-L1 levels in the discrimination of metabolic and primary neurological causes of impaired consciousness in the emergency departments.

Materials and methods

Patient selection

The study was initiated following approval from the local ethics committee. Eighty patients with IS, 40 patients with ICH, 80 patients with MDIC that were admitted to emergency department in a 9- month period and 40 healthy controls were enrolled in our study. Pa- tients who were over 18 years of age, who consented to participate in the study and who were admitted to the emergency department within

the first 24 hours after the onset of symptoms were included in the study, however, patients who were under 18 years of age, who did not consent to participate and who had Parkinson’s or Alzheimer’s disease were excluded from the study.

Determination of serum levels of human UCH-L1

Blood samples were collected into plain tubes by venipuncture. Sera were obtained by centrifugation at 1000g for 5 min at room tempera- ture and stored at -70 ?C until analysis. Serum UCHL1 level (ng/mL) was measured with double-antibody sandwich enzyme-linked immu- nosorbent assay (ELISA) using a human UCH-L1 ELISA kit (YH Biosearch Laboratory, Shanghai, China, Catalogue No: YHB3139Hu). Minimum de- tectable concentration of this assay was estimated at 0.1 ng/mL. All samples were assayed in duplicate. Absorbance was read spectrophoto- metrically via ELx800(TM) Absorbance Microplate Reader (BioTek Instru- ments, Inc., Winooski, VT, USA) at 450 nm. The intra- and inter-assay coefficients of variance of this kit were b 10% and b 12%, respectively.

Statistical analysis

Data were analyzed by Statistical Package for the Social Sciences (SPSS 21, Chicago, IL, USA). The numerical and categorical data were expressed as mean +- standard deviation and percentage, respectively. Kolmogorov-Smirnov and Shapiro-Wilk tests were used as tests of

Fig. 1. Schematic demonstration of UCH-L1 levels in study participants.

I. Yigit et al. / American Journal of Emergency Medicine 35 (2017) 1895–1898 1897

Table 2

The characteristics of patients with metabolic disorder induced impaired consciousness.

	DKA	HE	Hypoglycemia	Hyponatremia	Sepsis	Other
n (male/female)	14 (8/6)	5 (2/3)	34 (15/19)	9 (7/2)	6 (3/3)	12 (3/3)
Age (mean +- SD)	56.07 +- 18.85	73.80 +- 15.53	69.82 +- 15.87	72.11 +- 13.45	71.33 +- 24.22	69.33 +- 14.50
GCS median (minimum-maximum)	14 (13-14)	13 (10-14)	14 (6-14)	13 (11-14)	9.5 (4-14)	13 (3-14)
Time (mean +- SD)^a	3.79 +- 1.76	3.60 +- 1.34	3.32 +- 1.32	5.89 +- 1.27	4.50 +- 0.55	5.75 +- 1.60
UCHL1 (ng/mL) median (IQR)	3.54 (2.89-7.12)	4.68 (0.80-5.08)	3.54 (2.72-5.98)	7.90 (4.70-11.44)	4.14 (2.99-11.89)	4.98 (3.51-9.60)

DKA: diabetic ketoacidosis, HE: Hepatic encephalopathy, GCS: Glasgow coma scale, UCH-L1: ubiquitin C-terminal hydrolase-L1.

^a Time: the period between the onset of symptoms and blood sampling.

normality for continuous variables. Kruskal-Wallis test was used for the comparison of more than two independent groups that do not meet the normal distribution. Mann-Whitney U test was used for determining the relationship between two groups. Spearman’s correlation test was used for the assessment of the correlation between scales, which are or- dinal variables. p b 0.05 was considered significant.

Results

Eighty patients with IS, 40 patients with ICH, 80 patients with MDIC and 40 healthy volunteers participated in our study. The characteristics and vital parameters of patients during blood sampling were given in Table 1.

In our study, the levels of UCH-L1 in patients with IS and ICH were found to be significantly increased compared to in patients with MDIC and in healthy volunteers (Fig. 1).

The causes of MDIC were as follows: hypoglycemia (n = 34), diabet- ic ketoacidosis (n = 14), hyponatremia (n = 9), sepsis (n = 6), hepatic encephalopathy (n = 5) and other (n = 12). The characteristics of pa- tients with MDIC were given in Table 2.

UCH-L1 levels were found not to differ between IS and ICH groups (p = 0.468), however, significantly increased (p = 0.025) levels were detected in patients with IS than in patients with MDIC. Levels were sig- nificantly higher in patients with ICH compared with the healthy con- trol group (p = 0.012) and patients with MDIC (p = 0.002). The pairwise comparison of UCH-L1 levels in study groups were given in Table 1.

In ROC curve analysis, we detected 63.75% sensitivity and 62.5% specificity (AUC = 0.626, p b 0.0199, 95% CI: 0.533-0.713) with a cut- off value of 4.336 ng/ml for IS and 75% sensitivity and 55% specificity (AUC = 0.664, p b 0.0071, 95% CI: 0.549-0.766) with a cut-off value of

4.036 ng/ml for ICH. However, the sensitivity and specificity for MDIC was 36.25% and 77.5%, respectively, with a cut-off value of

3.256 ng/ml (AUC = 0.525, p = 0.6521, 95% CI: 0.432-0.617)

(Table 3). UCH-L1 levels were found to increase significantly with increasing time between the onset of symptoms and blood sampling (r = 0.345, p b 0.001). However, no correlation was found between UCH-L1 levels and age (r = 0.014, p = 0.833), GCS (r = -0.115, p = 0.074), mRS (r = 0.063, p = 0.475) and NIHSS (r = 0.056, p = 0.520).

Discussion

Impaired consciousness may occur due to metabolic or primary structural reasons that are related to the brain tissue. UCH-L1 levels

Table 3

ROC curve analysis results of Serum UCH-L1 levels for ischemic stroke, intracranial hemor- rhage and metabolic disorder induced impaired consciousness.

Cut-off	AUC	Sensitivity (%)	Specificity (%)	p	95% CI
IS N 4.336	0.626	63.75	62.50	0.0199	0.533-0.713
ICH N 4.036	0.664	75.00	55.00	0.0071	0.549-0.766
MDIC <= 3.256	0.525	36.25	77.50	0.6521	0.432-0.617

AUC: area under curve, CI: confidence interval, IS: ischemic stroke, ICH: intracranial hem- orrhage, MDIC: metabolic disorder induced impaired consciousness.

were significantly increased in patients with primary structural reasons compared with patients with MDIC.

UCH-L1 can be detected both in CSF and in the serum, and is released in the brain and Peripheral nervous system at a high rate. UCH-L1 plays an important role in the removal of oxidized or excess proteins by being involved in the ubiquitination of proteins [3-5,7,9]. Regarding these rea- sons, UCH-L1 levels have been investigated in acute and chronic intra- cranial pathologies.

The studies in the literature on the determination of UCH-L1 levels in acute intracranial diseases generally include trauma, seizures and stroke. The levels of UC-HL1 have been reported to be increased in par- allel with the brain injury especially in pediatric and adult patients with head trauma. In a biomarker study in 39 pediatric patients with differ- ent levels of Traumatic brain injury and 10 healthy controls, Berger et al. [17] reported that UCH-L1 levels were increased with increased TBI severity and decreased GCS. Similarly, in a study comparing UCH- L1 and GFAB levels in 206 traumatic brain injury patients with a mean age of 47 years and 175 healthy controls with a mean age of 37 years, Diaz-Arrastia et al. [18] reported significantly increased UCH-L1 levels with GCS. UCH-L1 levels increase after a certain period of time after the onset of symptoms and then decrease. In our study, consistent with the literature, significantly increased UCH-L1 levels were found in patients with ICH compared with the healthy control group.

Different results have been achieved in animal and human studies on UCH-L1 levels in stroke. Liu et al. [14] reported increased serum and CSF UCH-L1 levels in experimental animal models of traumatic brain injury and Middle cerebral artery occlusion induced ischemia. In a study on UCH-L1 and ?II-spectrin levels in experimental rat models of IS and non-traumatic hemorrhagic stroke, Ren et al. [13] detected sig- nificantly increased serum and CSF levels of UCH-L1 via ELISA method 3 and 6 h after the onset of IS and only increased serum UCH-L1 levels via Westernblot method 6 h later. They reported that ELISA method was more sensitive than Westernblot method in the determination of UCH-L1 levels. They also reported no significant increase in UCH-L1 levels after hemorrhagic stroke. In a study comparing healthy controls, ICH, IS, subarachnoid hemorrhage and Transient ischemic attack groups with Kruskal-Wallis test, Ren et al. [15] reported that the UCH-L1 levels between the groups were significantly different, mainly due to ICH and SAH groups. They reported that there were no significant differences between IS, TIA and healthy control groups. They reported an average of 4.5 h for blood sampling following the onset of the symp- toms. In our study, we detected significantly increased serum UCH-L1 levels also in patients with IS compared with the healthy control group (p = 0.025). We believe that this difference was due to an aver- age of 5.82 h for blood sampling in our IS group. Because it was shown that the increase in UCH-L1 levels started within 2-6 h and that UCH- L1 levels could be detected in serum or CSF at 24-48 h in animal studies [13,14]. Liu et al. [14] also reported prolonged higher UCH-L1 levels in experimental ischemia with longer duration.

MDIC occurs as a result of an accumulation of Toxic substances or a decrease in substances used by the brain. Unlike intracranial disorders, the brain is diffusely affected in case of MDIC. In our study, UCH-L1 levels were evaluated in some metabolic diseases that caused MDIC and the levels were found to remain at the same level with healthy in- dividuals. We believe that the reasons for this proximity in UCH-L1

1898 I. Yigit et al. / American Journal of Emergency Medicine 35 (2017) 1895–1898

levels are rapid development of impaired consciousness in metabolic disorders and faster admissions of these patients to hospitals. But higher levels of UCH-L1 in diseases like IS may be due to both delayed admis- sions to hospitals and directly affected brain cells in these diseases. We detected higher levels of UCH-L1 in hyponatremia, and this might be due to slow progression of clinical situation and longer duration of blood sampling after the onset of the symptoms. For these reasons, brain injury may occur in metabolic disorders at later stages and the levels of biomarkers associated with brain cells may possibly increase during this period.

There was a significant difference in UCH-L1 levels between patients with metabolic and intracranial impairment of consciousness. But be- cause of close cut-off values between the two groups, we think that a single measurement cannot clearly distinguish these disease groups, and that using other parameters affecting the UCH-L1 level, such as the time of sampling, along with clinical evaluation may be effective in clinical decision-making.

In conclusion, as seen in intracranial degenerative diseases, UCH-L1 levels increase in primary acute intracranial disorders (IS, ICH etc.) de- pending on the time from the onset of symptoms to blood sampling. UCH-L1 levels in rapidly developing MDICs were found not to be signif- icantly increased compared with healthy individuals. However, UCH-L1 levels also increase in slowly progressive MDICs, such as hyponatremia.

Study limitations

The main limitations of our study were as follows: (1) As this study was intended for diagnosis in the emergency department, sin- gle measurements were performed and serial measurements at cer- tain time intervals could not be performed. (2) Although there are many diseases causing MDIC, only a few of these conditions were in- cluded in this study. (3) The appropriate time standardization could not be made due to late admissions of patients enrolled in the study and therefore, the times of blood sampling were different between study groups.

Financial support and sponsorship

This study was supported by Firat University Scientific Research Unit for Funding (FUBAP TF.15.01).

Acknowledgment

This study was presented as oral presentation in the 3rd Interconti- nental Emergency Medicine Congress (May 19-22, 2016, Antalya, Turkey).

References

AJEM