Clinica Chimica Acta 453 (2016) 62–66
Contents lists available at ScienceDirect
Clinica Chimica Acta journal homepage: www.elsevier.com/locate/clinchim
Change of serum levels of thioredoxin in patients with severe traumatic brain injury De-Sheng Pan a,⁎, Hai-Wei Le b, Min Yan a, Muhammad Hassan a, Jiang-Biao Gong a, Hao Wang a a
Department of Neurosurgery, The First Affiliated Hospital, School of Medicine, Zhejiang University, 79 Qingchun Road, Hangzhou 310003, China Department of Neurosurgery, The People's Hospital of Beilun District, Beilun Branch Hospital of The First Affiliated Hospital of Medical School of Zhejiang University, 1288 Lushan East Road, Beilun District, Ningbo 315800, China b
a r t i c l e
i n f o
Article history: Received 14 November 2015 Received in revised form 29 November 2015 Accepted 30 November 2015 Available online 2 December 2015 Keywords: Traumatic brain injury Thioredoxin Mortality Severity Function outcome
a b s t r a c t Background: Thioredoxin (TRX), a potent anti-oxidant, is released during inflammation and oxidative stress. The purpose of this study was to establish the relationship between serum TRX concentrations and trauma severity and outcome in severe traumatic brain injury (STBI). Methods: We determined serum TRX concentrations in 112 patients and 112 controls. Multivariate analyses were performed to analyze the predictive factors of 1-week mortality, 6-month mortality and 6-month unfavorable outcome. The predictive values were investigated under receiver operating characteristic curves. Results: Serum TRX concentrations were markedly higher in patients than in controls (19.1 ± 7.8 ng/ml vs. 8.0 ± 2.3 ng/ml, P b 0.001). There was a significant negative association between serum TRX concentrations and Glasgow coma scale (GCS) scores (r = −0.543, P b 0.001). Increased TRX was identified as an independent prognostic marker of 1-week mortality [Odds ratio (OR), 1.220; 95% confidence interval (CI), 1.101–1.367; P b 0.001], 6-month mortality (OR, 1.201; 95% CI, 1.097–1.324; P b 0.001) and 6-month unfavorable outcome (OR, 1.189; 95% CI, 1.090–1.311; P b 0.001). TRX concentrations improved area under curve of GCS scores for 6-month unfavorable outcome, but not for 1-week mortality and 6-month mortality. Conclusions: Increased serum TRX concentration, associated highly with trauma severity and poor outcome, might be a novel prognostic marker in patients with STBI. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Oxidative stress reaction is one of the most popular mechanisms in the secondary brain injury after traumatic brain injury (TBI) [1–3]. Oxidative stress, resulting from an imbalance in the production of reactive oxygen species and the anti-oxidative defenses that maintain a cellular redox state, is proposed to be an important factor leading to oxidative damage in brain tissues and consequently contributing to mortality or neurological dysfunction after TBI [4–6]. The determination of oxidative stress in TBI would be very significant for a better understanding of TBI pathophysiology and for identifying subgroups of patients at the risk of poor prognosis. Blood biomarkers of oxidative stress have been investigated broadly in recent years [7–10]. Thioredoxin (TRX), a ubiquitous, 12.5 kDa intracellular thiol protein, is identified as a potent anti-oxidant which regulates inflammation, cell growth and apoptosis [11,12]. TRX is released during oxidative stress and inflammation, and subsequently its circulating concentrations are increased in many diseases including sepsis, post-cardiac arrest syndrome, acute myocardial infarction and malignant neoplasms, in which TRX has close relation to the severity and prognosis [13–17], ⁎ Corresponding author. E-mail address: [email protected] (D.-S. Pan).
http://dx.doi.org/10.1016/j.cca.2015.11.030 0009-8981/© 2015 Elsevier B.V. All rights reserved.
suggesting that TRX might be an oxidative stress marker. Recently, the close association of increased serum TRX with the clinical severity, infarct volume and short-term functional outcome were demonstrated in patients with acute ischemic stroke [18], indicating TRX as a possible novel prognostic marker in neurological diseases. 2. Methods 2.1. Design and subjects A prospective, observational study of a consecutive series of patients with severe TBI at our hospital between January 2011 and January 2014 was performed. Severe TBI was defined as Glasgow Coma Scale (GCS) score lower than 9 points. Patients with Injury Severity Score in noncranial aspects higher than 9 points, time from trauma to admission more than 6 h, pregnancy, infectious diseases, immunological diseases, use of immunosuppressant, fever within recent 1 month before head trauma, an increased white blood cell count, positive chest X-ray, b18 y, previous severe head trauma, neurological disease including ischemic or hemorrhagic stroke, use of antiplatelet or anticoagulant medication and presence of other prior systemic diseases such as diabetes mellitus, hypertension, uremia, liver cirrhosis, malignancy and chronic heart or lung disease were excluded from the study.
D.-S. Pan et al. / Clinica Chimica Acta 453 (2016) 62–66
The control group consisted of healthy volunteers who came to our hospital for healthy examination during the period of July 2013 to January 2014 and were checked for the absence of chronic or acute illness by questionnaire and medical examination. The study followed the tenets of the Declaration of Helsinki and was approved by the Institutional Review Board of our hospital. Written informed consent from the subjects or from their legal guardians was obtained. 2.2. Assessment The following information was recorded for each patient: age, gender, blood pressure, time from trauma to admission, GCS score, traumatic subarachnoid hemorrhage, abnormal cisterns, midline shift N5 mm, pupil dilation and brain lesion according to Marshall computed tomography classification [19]. Radiological procedures were completed according to the neuroradiology department protocol. The investigative group comprised a neurosurgeon and a radiologist and they were blinded to clinical information. Clinical outcomes were evaluated with 1-week mortality, 6-month mortality and 6-month unfavorable outcome. Unfavorable outcome was defined as Glasgow outcome scale score of 1–3 [20–22]. For follow-up, we used structure telephone interviews performed by 1 doctor, blinded to clinical information. 2.3. Immunological analysis The blood was drawn from cubital vein at admission in the patients and at study entry in the controls. After centrifugation, aliquots of the samples were immediately stored −80 °C before assay. Measurements of TRX were performed in duplicate samples with a commercially available sensitive enzyme-linked immunosorbent assay (Redox Biosciences, Kyoto, Japan). Patients exhibiting hemolysis were excluded due to the high intracellular concentration of TRX, which will bias assessment [11]. All determinations were performed by laboratory technicians blinded to all clinical data.
63
non-cranial aspects N9 points; 4 patients, time from trauma to admission N6 h; 1 patient, pregnancy; 2 patients, infectious diseases; 2 patients, immunological diseases; 1 patient, use of immunosuppressant; 2 patients, fever within recent 1 month before head trauma; 3 patients, an increased white blood cell count; 2 patients, positive chest X-ray; 1 patient, b 18 y; 2 patients, previous severe head trauma; 4 patients, neurological disease including ischemic or hemorrhagic stroke; 3 patients, use of antiplatelet or anticoagulant medication; 7 patients, presence of other prior systemic diseases such as diabetes mellitus, hypertension, uremia, liver cirrhosis, malignancy and chronic heart or lung disease. This study finally included 112 STBI patients and 112 healthy controls. There were not statistical significances in intergroup differences of gender and age. In this group of STBI patients, median initial postresuscitation GCS scores were 5 (4–7) (range, 3–8); 52 patients (46.4%) had unreactive pupils on admission; 55 patients (49.1%), CT classification 5 or 6; 53 patients (47.3%), abnormal cisterns on initial CT scan; 57 patients (50.9%), midline shift N5 mm on initial CT scan; 65 patients (58.0%), presence of traumatic subarachnoid hemorrhage on initial CT scan; 61 patients (54.5%), intracranial surgery in 1st 24 h. 3.2. The change of serum TRX concentrations and their association with GCS scores The admission serum TRX concentrations were significantly increased in all patients (19.1 ± 7.8 ng/ml), compared with healthy controls (8.0 ± 2.3 ng/ml, P b 0.001). We found that the serum TRX concentrations reflected the head trauma severity. Concentrations of TRX increased with increasing severity of head trauma as defined by the GCS score. There was a significant negative association between serum TRX concentrations and GCS scores (r = −0.543, P b 0.001) in Fig. 1. 3.3. 1-week mortality prediction
2.4. Statistical methods The normality of data distribution was assessed by the Kolmogorovor–Smirnov test or Shapiro–Wilk test. The results were reported as counts (percentage) for the categorical variables, mean ± SD if normally distributed and median (the upper and lower quartiles) if not normally distributed for the continuous variables. Linear relationships between GCS scores and TRX concentrations were examined using Spearman's correlation coefficient. Initial univariate analysis was done to assess the statistical significance of the intergroup observed difference with Student t test or Mann–Whitney U-test for the continuous variables and χ2 test or Fisher exact test for the categorical variables as appropriate. All parameters that were found to be significant in the univariate analysis were further analyzed using multivariate regression to identify those parameters that retained significant while accounting for all relevant variables. The odds ratio values and 95% confidence intervals were calculated and reported. Receiver operating curves (ROCs) were generated to determine cutoff values for optimal prognostic predictive sensitivities and specificities. The area under curves (AUCs) and 95% CI were calculated and reported based on the ROC curves. In a combined logistic-regression model, the additive benefit of TRX concentrations to GCS scores was estimated. All data were analyzed using SPSS 19. and MedCalc ver 9.6.4.0 Statistical significance was defined as a P b 0.05.
Fifteen patients (13.4%) deceased within 1 week after head trauma. In Table 1, non-survivors had higher serum TRX concentrations than survivors. When the variables the univariate analysis found significant were introduced into the logistic model, GCS scores (OR, 0.299; 95% CI, 0.102–0.774; P b 0.001) and serum TRX concentrations (OR, 1.220; 95% CI, 1.101–1.367; P b 0.001) were identified as the independent predictors for 1-week mortality of patients. A ROC curve identified that an admission serum TRX concentration N 21.7 ng/ml predicted 1-week mortality of patients
3. Results 3.1. Subjects' characteristics Initially, 149 patients were assessed. 37 patients were excluded because of the following reasons. 3 patients had Injury Severity Score in
Fig. 1. The association of serum thioredoxin (TRX) levels and Glasgow coma scale (GCS) scores.
64
D.-S. Pan et al. / Clinica Chimica Acta 453 (2016) 62–66
Table 1 Factors associated with 1-week mortality.
Table 2 Factors associated with 6-month mortality.
Characteristics
Non-survivors (n = 15)
Survivors (n = 97)
P value
Characteristics
Non-survivors (n = 32)
Survivors (n = 80)
P value
Male Age (y) GCS score Unreactive pupils CT classification 5 or 6 Abnormal cisterns Midline shift N5 mm Traumatic subarachnoid hemorrhage Intracranial surgery in 1st 24 h Admission time (h) Plasma-sampling time (h) Systolic arterial pressure (mmHg) Diastolic arterial pressure (mmHg) Mean arterial pressure (mmHg) Blood glucose level (mmol/l) Plasma C-reactive protein level (mg/l) Blood white blood cell count (109/l) Blood hemoglobin level (g/l) Blood platelet count (109/l) Serum thioredoxin level (ng/ml)
10 (66.7%) 40.4 ± 10.5 3(3–4) 14 (93.3%) 12 (80.0%) 14 (93.3%) 13 (86.7%) 14 (93.3%) 10 (66.7%) 1.7 ± 0.9 4.6 ± 1.8 127.5 ± 14.0 74.2 ± 8.0 92.0 ± 5.8 14.7 ± 4.8 10.6 ± 2.9 8.1 ± 4.1 119.0 ± 22.1 133.7 ± 40.6 28.2 ± 9.4
62 (63.9%) 39.1 ± 14.0 5(4–7) 38 (39.2%) 43 (44.3%) 39 (40.2%) 44 (45.4%) 51 (52.6%) 51 (52.6%) 2.0 ± 1.2 4.1 ± 2.1 119.8 ± 22.8 71.7 ± 11.1 87.7 ± 12.8 11.5 ± 3.8 8.9 ± 2.3 8.5 ± 2.7 126.7 ± 28.2 142.3 ± 44.6 17.7 ± 6.5
NS NS b0.001 b0.001 0.010 b0.001 0.003 0.003 NS NS NS NS NS NS 0.004 0.015 NS NS NS b0.001
Male Age (y) GCS score Unreactive pupils CT classification 5 or 6 Abnormal cisterns Midline shift N5 mm Traumatic subarachnoid hemorrhage Intracranial surgery in 1st 24 h Admission time (h) Plasma-sampling time (h) Systolic arterial pressure (mmHg) Diastolic arterial pressure (mmHg) Mean arterial pressure (mmHg) Blood glucose level (mmol/l) Plasma C-reactive protein level (mg/l) Blood white blood cell count (109/l) Blood hemoglobin level (g/l) Blood platelet count (109/l) Serum thioredoxin level (ng/ml)
20 (62.5%) 40.1 ± 12.0 4(3–4) 26 (81.3%) 23 (71.9%) 25 (78.1%) 24 (75.0%) 25 (78.1%) 20 (62.5%) 1.8 ± 0.9 4.1 ± 1.6 120.1 ± 28.3 71.6 ± 9.1 87.8 ± 11.6 14.0 ± 4.7 10.3 ± 3.2 8.5 ± 3.8 122.6 ± 33.0 138.0 ± 40.3 25.9 ± 8.0
52 (65.0%) 39.0 ± 14.2 6(4–7) 26 (32.5%) 32 (40.0%) 28 (35.0%) 33 (41.3%) 40 (50.0%) 41 (51.3%) 2.0 ± 1.3 4.2 ± 2.3 121.1 ± 19.1 72.2 ± 11.4 88.5 ± 12.4 11.0 ± 3.5 8.7 ± 2.0 8.4 ± 2.6 126.9 ± 25.0 142.4 ± 45.5 16.4 ± 5.8
NS NS b0.001 b0.001 0.002 b0.001 0.001 0.006 NS NS NS NS NS NS 0.002 0.002 NS NS NS b0.001
Numerical variables were presented as mean ± SD or median (the upper and lower quartiles) and analyzed by unpaired student's t test or Mann–Whitney U-test. Categorical variables were expressed as counts (percentage) and analyzed by χ2 test or Fisher exact test. GCS indicates Glasgow Coma Scale; CT, computerized tomography.
Numerical variables were presented as mean ± SD or median (the upper and lower quartiles) and analyzed by unpaired student's t test or Mann–Whitney U-test. Categorical variables were expressed as counts (percentage) and analyzed by χ2 test or Fisher exact test. GCS indicates Glasgow Coma Scale; CT, computerized tomography.
with 86.7% sensitivity and 79.4% specificity, with AUC at 0.863 (95% CI, 0.785–0.921) (Fig. 2). Based on AUC, its predictive value was similar to GCS score's (AUC, 0.885; 95% CI, 0.811–0.937; P = 0.734). In Fig. 1, TRX numerically improved AUC of GCS score to 0.915 (95% CI, 0.848–0.960; P = 0.348).
Thirty-two patients (28.6%) died within 6-month after head trauma. Higher serum TRX concentrations and other variables in Table 2 were associated highly with 6-month mortality. When all parameters that
were found to be significant in the univariate analysis were further analyzed using multivariate regression, GCS scores (OR, 0.328; 95% CI, 0.114–0.798; P b 0.001) and serum TRX concentrations (OR, 1.201; 95% CI, 1.097–1.324; P b 0.001) retained significant for the prediction of 6-month mortality of patients. Under ROC curve, an admission serum TRX concentration N 21.7 ng/ml predicted 6-month mortality of patients with 75.0% sensitivity and 88.8% specificity, with AUC at 0.859 (95% CI, 0.781–0.918) (Fig. 3). Based on AUC, its predictive value was similar to GCS score's (AUC, 0.873; 95% CI, 0.797–0.928; P = 0.778). When a combined logistic-regression model was constructed, TRX numerically
Fig. 2. Receiver operating characteristic (ROC) curve analysis of Glasgow coma scale (GCS) scores, serum thioredoxin (TRX) levels and GCS scores combined with serum TRX levels for identifying severe traumatic brain injury patients with 1-week mortality.
Fig. 3. Receiver operating characteristic (ROC) curve analysis of Glasgow coma scale (GCS) scores, serum thioredoxin (TRX) levels and GCS scores combined with serum TRX levels for identifying severe traumatic brain injury patients with 6-month mortality.
3.4. 6-month mortality prediction
D.-S. Pan et al. / Clinica Chimica Acta 453 (2016) 62–66
65
improved AUC of GCS score to 0.909 (95% CI, 0.840–0.955; P = 0.098), shown in Fig. 2. 3.5. 6-month unfavorable outcome prediction Fifty-five patients (49.1%) had unfavorable outcome within 6-month after head trauma. Higher serum TRX concentrations and other variables in relation to 6-month unfavorable outcome were shown in Table 3. When all parameters appearing significant in the univariate analysis were further assessed using multivariate regression, (OR, 0.344; 95% CI, 0.120–0.842; P b 0.001) and serum TRX concentrations (OR, 1.189; 95% CI, 1.090–1.311; P b 0.001) emerged statistically significant for the prediction of 6-month unfavorable outcome of patients. Based on the ROC curve, the optimal cutoff value of serum TRX concentrations as an indicator for prognosis of 6-month unfavorable outcome was projected to be 20.3 ng/ml, which yielded a sensitivity of 72.7% and a specificity of 87.7%, with AUC at 0.846 (95% CI, 0.766– 0.907) (Fig. 4). Based on AUC, its predictive value was similar to GCS score's (AUC, 0.852; 95% CI, 0.772–0.912; P = 0.896). We configured a combined logistic-regression model, and therefore, found that TRX statistically significantly improved AUC of GCS score to 0.908 (95% CI, 0.838–0.954; P = 0.032) in Fig. 3. 4. Discussion A recent study has investigated serum TRX concentrations in ischemic stroke and found elevation of this biomarker's concentrations in the peripheral blood [18]. Interestingly, serum TRX concentrations were associated negatively with GCS scores at admission. Importantly, we demonstrated that TRX emerge as a strong and independent prognostic biomarker of 1-week mortality, 6-month mortality and unfavorable outcome, suggesting that TBI patients were in high concentrations of oxidative stress and TRX have the potential to be a good prognostic biomarker after TBI. The pathophysiology of head trauma is dominated by primary brain injury and secondary brain injury associated with activation of inflammatory process and oxidative stress [23–26]. Oxidative stress increases quickly after brain injury, subsequently contributes widely to neuronal death including apoptosis and necrosis, and eventually leading to neurological dysfunction [4–6]. TRX reflects both inflammation and oxidative stress [11,12]. Intracellular TRX is released from cells on oxidative
Fig. 4. Receiver operating characteristic (ROC) curve analysis of Glasgow coma scale (GCS) scores, serum thioredoxin (TRX) levels and GCS scores combined with serum TRX levels for identifying severe traumatic brain injury patients with 6-month unfavorable outcome (defined as the Glasgow outcome scale scores of 1–3).
stress, leading to high extracellular concentrations in numerous situations relevant to critical care, including severe burn injury, acute lung injury, ischemia-reperfusion injury, heart disease and sepsis [13–17]. Our data showed that increased serum TRX concentrations had close relation to GCS scores, indicating TRX might be associated with trauma severity. To date, several biological substances related to oxidative stress, namely, malondialdehyde, F2-isoprostanes, glutathione, and NADPH oxidase have been investigated over the years as peripheral markers of oxidative stress in neurological diseases [27–31]. Disappointingly, these biomarkers have not shown high prognostic predictive values
Table 3 Factors associated with 6-month functional outcome. Characteristics
Unfavorable outcome (n = 55)
Favorable outcome (n = 57)
P value
Male Age (y) GCS score Unreactive pupils CT classification 5 or 6 Abnormal cisterns Midline shift N5 mm Traumatic subarachnoid hemorrhage Intracranial surgery in 1st 24 h Admission time (h) Plasma-sampling time (h) Systolic arterial pressure (mmHg) Diastolic arterial pressure (mmHg) Mean arterial pressure (mmHg) Blood glucose level (mmol/l) Plasma C-reactive protein level (mg/l) Blood white blood cell count (109/l) Blood hemoglobin level (g/l) Blood platelet count (109/l) Serum thioredoxin level (ng/ml)
35 (63.6%) 40.2 ± 12.7 4(3–5) 40 (72.2%) 34 (61.8%) 38 (69.1%) 37 (67.3%) 41 (74.6%) 35 (63.6%) 1.8 ± 1.1 4.1 ± 2.1 120.5 ± 25.4 71.6 ± 11.5 87.9 ± 13.4 13.2 ± 4.3 9.9 ± 2.9 8.2 ± 3.2 122.8 ± 29.3 139.1 ± 37.3 23.6 ± 7.7
37 (64.9%) 38.5 ± 14.4 7(5–8) 12 (21.1%) 21 (36.8%) 15 (26.3%) 20 (35.1%) 24 (42.1%) 26 (45.6%) 2.0 ± 1.2 4.3 ± 2.1 121.1 ± 18.3 72.4 ± 10.1 88.7 ± 10.9 10.7 ± 3.4 8.5 ± 1.8 8.6 ± 2.7 128.5 ± 25.5 143.2 ± 49.8 14.8 ± 4.9
NS NS b0.001 b0.001 0.008 b0.001 0.001 0.001 NS NS NS NS NS NS 0.001 0.002 NS NS NS b0.001
Numerical variables were presented as mean ± SD or median (the upper and lower quartiles) and analyzed by unpaired student's t test or Mann–Whitney U-test. Categorical variables were expressed as counts (percentage) and analyzed by χ2test or Fisher exact test. GCS indicates Glasgow Coma Scale; CT, computerized tomography.
66
D.-S. Pan et al. / Clinica Chimica Acta 453 (2016) 62–66
compared with other clinical grade systems like GCS [7–10]. Recently, Qi et al. found that TRX was an independent prognostic maker of 3-month functional outcome indicated by modified Rankin scale; Importantly, TRX improved the AUC of National Institutes of Health Stroke Scale score for functional outcome [18]. In the current study, besides 6month function outcome, 1-week mortality and 6-month mortality were also regarded as the parameters of clinical outcome. Multivariate analysis demonstrated that serum TRX was an independent prognostic predictor of these short-term or long-term clinical outcomes. Furthermore, we found that predictive value of serum TRX concentrations were similar to that of GCS scores; meanwhile, TRX statistically significantly improved the predictive performance of GCS scores for 6month unfavorable outcome, but not for 1-week mortality and 6mortality. Hence, there was a little difference in the prognostic predictive value of TRX between mortality and function outcome. However, the difference in the prediction of 6-month mortality had trend to statistical significance (P value = 0.098). Therefore, it is assumed that this difference might be derived from a small sample size. As a whole, TRX might be a good prognostic predictive biomarker of TBI. Our study had certain limitations. Firstly, the present work was a single center study. The results remained to be verified in a multicenter study. Secondly, we did not perform an analysis of serum TRX concentrations at different time points during follow-up to investigate the evolution of this parameter. Thus, this study might be regarded as a pilot one. 5. Conclusions The novel finding of our study was that serum TRX concentrations are associated with trauma severity and clinical outcomes including 1week mortality, 6-month mortality and 6-month unfavorable outcome. Moreover, TRX can improve the predictive value of GCS scores. Therefore, TRX could be used as a prognostic biomarker in patients with severe TBI. Abbreviations GCS TRX TBI
Glasgow coma scale thioredoxin traumatic brain injury
Acknowledgments The authors thank all staffs in Department of Neurosurgery, The First Affiliated Hospital, School of Medicine, Zhejiang University for their technical support. References [1] J. Sahuquillo, M.A. Poca, S. Amoros, Current aspects of pathophysiology and cell dysfunction after severe head injury, Curr. Pharm. Des. 7 (2001) 1475–1503. [2] P. Fernandez, D. Diaz, Experimental animal models of traumatic brain injury: medical and biomechanical mechanism, Crit. Rev. Neurosurg. 9 (1999) 44–52. [3] C. Cornelius, R. Crupi, V. Calabrese, et al., Traumatic brain injury (TBI): oxidative stress and neuroprotection, Antioxid. Redox Signal. 19 (2013) 836–853. [4] N. Şenol, M. Nazıroğlu, Melatonin reduces traumatic brain injury-induced oxidative stress in the cerebral cortex and blood of rats, Neural Regen. Res. 9 (2014) 1112–1116.
[5] W. Zhang, B. Li, Y. Guo, et al., Rhamnetin attenuates cognitive deficit and inhibits hippocampal inflammatory response and oxidative stress in rats with traumatic brain injury, Cent. Eur. J. Immunol. 40 (2015) 35–41. [6] K.S. Panickar, R.A. Anderson, Effect of polyphenols on oxidative stress and mitochondrial dysfunction in neuronal death and brain edema in cerebral ischemia, Int. J. Mol. Sci. 12 (2011) 8181–8207. [7] C. Nayak, D. Nayak, S. Bhat, A. Raja, A. Rao, Relationship between neurological outcome and early oxidative changes in erythrocytes in head injury patients, Clin. Chem. Lab. Med. 45 (2007) 629–633. [8] A. Paolin, L. Nardin, P. Gaetani, et al., Oxidative damage after severe head injury and its relationship to neurological outcome, Neurosurgery 51 (2002) 949–954. [9] L. Lorente, M.M. Martín, P. Abreu-González, et al., Association between serum malondialdehyde levels and mortality in patients with severe brain trauma injury, J. Neurotrauma 32 (2015) 1–6. [10] A. Hohl, S. Gullo Jda, C.C. Silva, et al., Plasma levels of oxidative stress biomarkers and hospital mortality in severe head injury: a multivariate analysis, J. Crit. Care 27 (2012) 523. [11] A. Burke-Gaffney, M.E. Callister, H. Nakamura, Thioredoxin: friend or foe in human disease? Trends Pharmacol. Sci. 26 (2005) 398–404. [12] Thioredoxin in sepsis: just another biomarker or a plausible therapeutic target? Crit. Care Med. 37 (2009) 2304–2305. [13] A. Burke-Gaffney, T.W. Evans, G.J. Quinlan, A. Abdiu, H. Nakamura, B. Sahaf, J. Yodoi, A. Holmgren, A. Rosen, Thioredoxin blood level increases after severe burn injury, Antioxid. Redox Signal. 2 (2000) 707–716. [14] M.E. Callister, A. Burke-Gaffney, G.J. Quinlan, et al., Extracellular thioredoxin levels are increased in patients with acute lung injury, Thorax 61 (2006) 521–527. [15] M.E. Callister, A. Burke-Gaffney, G.J. Quinlan, H. Betts, S. Nadel, T.W. Evans, Persistently low plasma thioredoxin is associated with meningococcal septic shock in children, Intensive Care Med. 33 (2007) 364–367. [16] A. Jekell, A. Hossain, U. Alehagen, U. Dahlstrom, A. Rosen, Elevated circulating levels of thioredoxin and stress in chronic heart failure, Eur. J. Heart Fail. 6 (2004) 883–890. [17] S.K. Leaver, N.S. MacCallum, V. Pingle, et al., Increased plasma thioredoxin levels in patients with sepsis: positive association with macrophage migration inhibitory factor, Intensive Care Med. 36 (2010) 336–341. [18] A.Q. Qi, Y. Li, Q. Liu, et al., Thioredoxin is a novel diagnostic and prognostic marker in patients with ischemic stroke, Free Radic. Biol. Med. 80 (2015) 129–135. [19] L.F. Marshall, S.B. Marshall, M.R. Klauber, M.V. Clark, A new classification of head injury based on computerized tomography, J. Neurosurg. 75 (1991) S14–S20. [20] B. Jennett, M. Bond, Assessment of outcome after severe brain damage, Lancet 1 (1975) 480–484. [21] Z.Y. Zhang, L.X. Zhang, X.Q. Dong, et al., Comparison of the performances of copeptin and multiple biomarkers in long-term prognosis of severe traumatic brain injury, Peptides 60 (2014) 13–17. [22] K.Y. Wang, G.F. Yu, Z.Y. Zhang, Q. Huang, X.Q. Dong, Plasma high-mobility group box 1 levels and prediction of outcome in patients with traumatic brain injury, Clin. Chim. Acta 413 (2012) 1737–1741. [23] B. Halliwell, Reactive oxygen species and the central nervous system, J. Neurochem. 59 (1992) 1609–1623. [24] P.M. Kochanek, C.E. Dixon, D.K. Shellington, et al., Screening of biochemical and molecular mechanisms of secondary injury and repair in the brain after experimental blast-induced traumatic brain injury in rats, J. Neurotrauma 30 (2013) 920–937. [25] D. Brough, A. Denes, Interleukin-1α and brain inflammation, IUBMB Life 67 (2015) 323–330. [26] K.N. Corps, T.L. Roth, D.B. McGavern, Inflammation and neuroprotection in traumatic brain injury, JAMA Neurol. 72 (2015) 355–362. [27] M. Seneş, N. Kazan, O. Coşkun, O. Zengi, L. Inan, D. Yücel, Oxidative and nitrosative stress in acute ischaemic stroke, Ann. Clin. Biochem. 44 (2007) 43–47. [28] F. Içme, Ö. Erel, A. Avci, S. Satar, M. Gülen, S. Acehan, The relation between oxidative stress parameters, ischemic stroke, and hemorrhagic stroke, Turk. J. Med. Sci. 45 (2015) 947–953. [29] J. Kotur-Stevuljevic, N. Bogavac-Stanojevic, Z. Jelic-Ivanovic, et al., Oxidative stress and paraoxonase 1 status in acute ischemic stroke patients, Atherosclerosis 241 (2015) 192–198. [30] D. Paspalj, P. Nikic, M. Savic, et al., Redox status in acute ischemic stroke: correlation with clinical outcome, Mol. Cell. Biochem. 406 (2015) 75–81. [31] N.W. Tsai, Y.T. Chang, C.R. Huang, et al., Association between oxidative stress and outcome in different subtypes of acute ischemic stroke, Biomed. Res. Int. 2014 (2014) 256879.
Contents lists available at ScienceDirect
Clinica Chimica Acta journal homepage: www.elsevier.com/locate/clinchim
Change of serum levels of thioredoxin in patients with severe traumatic brain injury De-Sheng Pan a,⁎, Hai-Wei Le b, Min Yan a, Muhammad Hassan a, Jiang-Biao Gong a, Hao Wang a a
Department of Neurosurgery, The First Affiliated Hospital, School of Medicine, Zhejiang University, 79 Qingchun Road, Hangzhou 310003, China Department of Neurosurgery, The People's Hospital of Beilun District, Beilun Branch Hospital of The First Affiliated Hospital of Medical School of Zhejiang University, 1288 Lushan East Road, Beilun District, Ningbo 315800, China b
a r t i c l e
i n f o
Article history: Received 14 November 2015 Received in revised form 29 November 2015 Accepted 30 November 2015 Available online 2 December 2015 Keywords: Traumatic brain injury Thioredoxin Mortality Severity Function outcome
a b s t r a c t Background: Thioredoxin (TRX), a potent anti-oxidant, is released during inflammation and oxidative stress. The purpose of this study was to establish the relationship between serum TRX concentrations and trauma severity and outcome in severe traumatic brain injury (STBI). Methods: We determined serum TRX concentrations in 112 patients and 112 controls. Multivariate analyses were performed to analyze the predictive factors of 1-week mortality, 6-month mortality and 6-month unfavorable outcome. The predictive values were investigated under receiver operating characteristic curves. Results: Serum TRX concentrations were markedly higher in patients than in controls (19.1 ± 7.8 ng/ml vs. 8.0 ± 2.3 ng/ml, P b 0.001). There was a significant negative association between serum TRX concentrations and Glasgow coma scale (GCS) scores (r = −0.543, P b 0.001). Increased TRX was identified as an independent prognostic marker of 1-week mortality [Odds ratio (OR), 1.220; 95% confidence interval (CI), 1.101–1.367; P b 0.001], 6-month mortality (OR, 1.201; 95% CI, 1.097–1.324; P b 0.001) and 6-month unfavorable outcome (OR, 1.189; 95% CI, 1.090–1.311; P b 0.001). TRX concentrations improved area under curve of GCS scores for 6-month unfavorable outcome, but not for 1-week mortality and 6-month mortality. Conclusions: Increased serum TRX concentration, associated highly with trauma severity and poor outcome, might be a novel prognostic marker in patients with STBI. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Oxidative stress reaction is one of the most popular mechanisms in the secondary brain injury after traumatic brain injury (TBI) [1–3]. Oxidative stress, resulting from an imbalance in the production of reactive oxygen species and the anti-oxidative defenses that maintain a cellular redox state, is proposed to be an important factor leading to oxidative damage in brain tissues and consequently contributing to mortality or neurological dysfunction after TBI [4–6]. The determination of oxidative stress in TBI would be very significant for a better understanding of TBI pathophysiology and for identifying subgroups of patients at the risk of poor prognosis. Blood biomarkers of oxidative stress have been investigated broadly in recent years [7–10]. Thioredoxin (TRX), a ubiquitous, 12.5 kDa intracellular thiol protein, is identified as a potent anti-oxidant which regulates inflammation, cell growth and apoptosis [11,12]. TRX is released during oxidative stress and inflammation, and subsequently its circulating concentrations are increased in many diseases including sepsis, post-cardiac arrest syndrome, acute myocardial infarction and malignant neoplasms, in which TRX has close relation to the severity and prognosis [13–17], ⁎ Corresponding author. E-mail address: [email protected] (D.-S. Pan).
http://dx.doi.org/10.1016/j.cca.2015.11.030 0009-8981/© 2015 Elsevier B.V. All rights reserved.
suggesting that TRX might be an oxidative stress marker. Recently, the close association of increased serum TRX with the clinical severity, infarct volume and short-term functional outcome were demonstrated in patients with acute ischemic stroke [18], indicating TRX as a possible novel prognostic marker in neurological diseases. 2. Methods 2.1. Design and subjects A prospective, observational study of a consecutive series of patients with severe TBI at our hospital between January 2011 and January 2014 was performed. Severe TBI was defined as Glasgow Coma Scale (GCS) score lower than 9 points. Patients with Injury Severity Score in noncranial aspects higher than 9 points, time from trauma to admission more than 6 h, pregnancy, infectious diseases, immunological diseases, use of immunosuppressant, fever within recent 1 month before head trauma, an increased white blood cell count, positive chest X-ray, b18 y, previous severe head trauma, neurological disease including ischemic or hemorrhagic stroke, use of antiplatelet or anticoagulant medication and presence of other prior systemic diseases such as diabetes mellitus, hypertension, uremia, liver cirrhosis, malignancy and chronic heart or lung disease were excluded from the study.
D.-S. Pan et al. / Clinica Chimica Acta 453 (2016) 62–66
The control group consisted of healthy volunteers who came to our hospital for healthy examination during the period of July 2013 to January 2014 and were checked for the absence of chronic or acute illness by questionnaire and medical examination. The study followed the tenets of the Declaration of Helsinki and was approved by the Institutional Review Board of our hospital. Written informed consent from the subjects or from their legal guardians was obtained. 2.2. Assessment The following information was recorded for each patient: age, gender, blood pressure, time from trauma to admission, GCS score, traumatic subarachnoid hemorrhage, abnormal cisterns, midline shift N5 mm, pupil dilation and brain lesion according to Marshall computed tomography classification [19]. Radiological procedures were completed according to the neuroradiology department protocol. The investigative group comprised a neurosurgeon and a radiologist and they were blinded to clinical information. Clinical outcomes were evaluated with 1-week mortality, 6-month mortality and 6-month unfavorable outcome. Unfavorable outcome was defined as Glasgow outcome scale score of 1–3 [20–22]. For follow-up, we used structure telephone interviews performed by 1 doctor, blinded to clinical information. 2.3. Immunological analysis The blood was drawn from cubital vein at admission in the patients and at study entry in the controls. After centrifugation, aliquots of the samples were immediately stored −80 °C before assay. Measurements of TRX were performed in duplicate samples with a commercially available sensitive enzyme-linked immunosorbent assay (Redox Biosciences, Kyoto, Japan). Patients exhibiting hemolysis were excluded due to the high intracellular concentration of TRX, which will bias assessment [11]. All determinations were performed by laboratory technicians blinded to all clinical data.
63
non-cranial aspects N9 points; 4 patients, time from trauma to admission N6 h; 1 patient, pregnancy; 2 patients, infectious diseases; 2 patients, immunological diseases; 1 patient, use of immunosuppressant; 2 patients, fever within recent 1 month before head trauma; 3 patients, an increased white blood cell count; 2 patients, positive chest X-ray; 1 patient, b 18 y; 2 patients, previous severe head trauma; 4 patients, neurological disease including ischemic or hemorrhagic stroke; 3 patients, use of antiplatelet or anticoagulant medication; 7 patients, presence of other prior systemic diseases such as diabetes mellitus, hypertension, uremia, liver cirrhosis, malignancy and chronic heart or lung disease. This study finally included 112 STBI patients and 112 healthy controls. There were not statistical significances in intergroup differences of gender and age. In this group of STBI patients, median initial postresuscitation GCS scores were 5 (4–7) (range, 3–8); 52 patients (46.4%) had unreactive pupils on admission; 55 patients (49.1%), CT classification 5 or 6; 53 patients (47.3%), abnormal cisterns on initial CT scan; 57 patients (50.9%), midline shift N5 mm on initial CT scan; 65 patients (58.0%), presence of traumatic subarachnoid hemorrhage on initial CT scan; 61 patients (54.5%), intracranial surgery in 1st 24 h. 3.2. The change of serum TRX concentrations and their association with GCS scores The admission serum TRX concentrations were significantly increased in all patients (19.1 ± 7.8 ng/ml), compared with healthy controls (8.0 ± 2.3 ng/ml, P b 0.001). We found that the serum TRX concentrations reflected the head trauma severity. Concentrations of TRX increased with increasing severity of head trauma as defined by the GCS score. There was a significant negative association between serum TRX concentrations and GCS scores (r = −0.543, P b 0.001) in Fig. 1. 3.3. 1-week mortality prediction
2.4. Statistical methods The normality of data distribution was assessed by the Kolmogorovor–Smirnov test or Shapiro–Wilk test. The results were reported as counts (percentage) for the categorical variables, mean ± SD if normally distributed and median (the upper and lower quartiles) if not normally distributed for the continuous variables. Linear relationships between GCS scores and TRX concentrations were examined using Spearman's correlation coefficient. Initial univariate analysis was done to assess the statistical significance of the intergroup observed difference with Student t test or Mann–Whitney U-test for the continuous variables and χ2 test or Fisher exact test for the categorical variables as appropriate. All parameters that were found to be significant in the univariate analysis were further analyzed using multivariate regression to identify those parameters that retained significant while accounting for all relevant variables. The odds ratio values and 95% confidence intervals were calculated and reported. Receiver operating curves (ROCs) were generated to determine cutoff values for optimal prognostic predictive sensitivities and specificities. The area under curves (AUCs) and 95% CI were calculated and reported based on the ROC curves. In a combined logistic-regression model, the additive benefit of TRX concentrations to GCS scores was estimated. All data were analyzed using SPSS 19. and MedCalc ver 9.6.4.0 Statistical significance was defined as a P b 0.05.
Fifteen patients (13.4%) deceased within 1 week after head trauma. In Table 1, non-survivors had higher serum TRX concentrations than survivors. When the variables the univariate analysis found significant were introduced into the logistic model, GCS scores (OR, 0.299; 95% CI, 0.102–0.774; P b 0.001) and serum TRX concentrations (OR, 1.220; 95% CI, 1.101–1.367; P b 0.001) were identified as the independent predictors for 1-week mortality of patients. A ROC curve identified that an admission serum TRX concentration N 21.7 ng/ml predicted 1-week mortality of patients
3. Results 3.1. Subjects' characteristics Initially, 149 patients were assessed. 37 patients were excluded because of the following reasons. 3 patients had Injury Severity Score in
Fig. 1. The association of serum thioredoxin (TRX) levels and Glasgow coma scale (GCS) scores.
64
D.-S. Pan et al. / Clinica Chimica Acta 453 (2016) 62–66
Table 1 Factors associated with 1-week mortality.
Table 2 Factors associated with 6-month mortality.
Characteristics
Non-survivors (n = 15)
Survivors (n = 97)
P value
Characteristics
Non-survivors (n = 32)
Survivors (n = 80)
P value
Male Age (y) GCS score Unreactive pupils CT classification 5 or 6 Abnormal cisterns Midline shift N5 mm Traumatic subarachnoid hemorrhage Intracranial surgery in 1st 24 h Admission time (h) Plasma-sampling time (h) Systolic arterial pressure (mmHg) Diastolic arterial pressure (mmHg) Mean arterial pressure (mmHg) Blood glucose level (mmol/l) Plasma C-reactive protein level (mg/l) Blood white blood cell count (109/l) Blood hemoglobin level (g/l) Blood platelet count (109/l) Serum thioredoxin level (ng/ml)
10 (66.7%) 40.4 ± 10.5 3(3–4) 14 (93.3%) 12 (80.0%) 14 (93.3%) 13 (86.7%) 14 (93.3%) 10 (66.7%) 1.7 ± 0.9 4.6 ± 1.8 127.5 ± 14.0 74.2 ± 8.0 92.0 ± 5.8 14.7 ± 4.8 10.6 ± 2.9 8.1 ± 4.1 119.0 ± 22.1 133.7 ± 40.6 28.2 ± 9.4
62 (63.9%) 39.1 ± 14.0 5(4–7) 38 (39.2%) 43 (44.3%) 39 (40.2%) 44 (45.4%) 51 (52.6%) 51 (52.6%) 2.0 ± 1.2 4.1 ± 2.1 119.8 ± 22.8 71.7 ± 11.1 87.7 ± 12.8 11.5 ± 3.8 8.9 ± 2.3 8.5 ± 2.7 126.7 ± 28.2 142.3 ± 44.6 17.7 ± 6.5
NS NS b0.001 b0.001 0.010 b0.001 0.003 0.003 NS NS NS NS NS NS 0.004 0.015 NS NS NS b0.001
Male Age (y) GCS score Unreactive pupils CT classification 5 or 6 Abnormal cisterns Midline shift N5 mm Traumatic subarachnoid hemorrhage Intracranial surgery in 1st 24 h Admission time (h) Plasma-sampling time (h) Systolic arterial pressure (mmHg) Diastolic arterial pressure (mmHg) Mean arterial pressure (mmHg) Blood glucose level (mmol/l) Plasma C-reactive protein level (mg/l) Blood white blood cell count (109/l) Blood hemoglobin level (g/l) Blood platelet count (109/l) Serum thioredoxin level (ng/ml)
20 (62.5%) 40.1 ± 12.0 4(3–4) 26 (81.3%) 23 (71.9%) 25 (78.1%) 24 (75.0%) 25 (78.1%) 20 (62.5%) 1.8 ± 0.9 4.1 ± 1.6 120.1 ± 28.3 71.6 ± 9.1 87.8 ± 11.6 14.0 ± 4.7 10.3 ± 3.2 8.5 ± 3.8 122.6 ± 33.0 138.0 ± 40.3 25.9 ± 8.0
52 (65.0%) 39.0 ± 14.2 6(4–7) 26 (32.5%) 32 (40.0%) 28 (35.0%) 33 (41.3%) 40 (50.0%) 41 (51.3%) 2.0 ± 1.3 4.2 ± 2.3 121.1 ± 19.1 72.2 ± 11.4 88.5 ± 12.4 11.0 ± 3.5 8.7 ± 2.0 8.4 ± 2.6 126.9 ± 25.0 142.4 ± 45.5 16.4 ± 5.8
NS NS b0.001 b0.001 0.002 b0.001 0.001 0.006 NS NS NS NS NS NS 0.002 0.002 NS NS NS b0.001
Numerical variables were presented as mean ± SD or median (the upper and lower quartiles) and analyzed by unpaired student's t test or Mann–Whitney U-test. Categorical variables were expressed as counts (percentage) and analyzed by χ2 test or Fisher exact test. GCS indicates Glasgow Coma Scale; CT, computerized tomography.
Numerical variables were presented as mean ± SD or median (the upper and lower quartiles) and analyzed by unpaired student's t test or Mann–Whitney U-test. Categorical variables were expressed as counts (percentage) and analyzed by χ2 test or Fisher exact test. GCS indicates Glasgow Coma Scale; CT, computerized tomography.
with 86.7% sensitivity and 79.4% specificity, with AUC at 0.863 (95% CI, 0.785–0.921) (Fig. 2). Based on AUC, its predictive value was similar to GCS score's (AUC, 0.885; 95% CI, 0.811–0.937; P = 0.734). In Fig. 1, TRX numerically improved AUC of GCS score to 0.915 (95% CI, 0.848–0.960; P = 0.348).
Thirty-two patients (28.6%) died within 6-month after head trauma. Higher serum TRX concentrations and other variables in Table 2 were associated highly with 6-month mortality. When all parameters that
were found to be significant in the univariate analysis were further analyzed using multivariate regression, GCS scores (OR, 0.328; 95% CI, 0.114–0.798; P b 0.001) and serum TRX concentrations (OR, 1.201; 95% CI, 1.097–1.324; P b 0.001) retained significant for the prediction of 6-month mortality of patients. Under ROC curve, an admission serum TRX concentration N 21.7 ng/ml predicted 6-month mortality of patients with 75.0% sensitivity and 88.8% specificity, with AUC at 0.859 (95% CI, 0.781–0.918) (Fig. 3). Based on AUC, its predictive value was similar to GCS score's (AUC, 0.873; 95% CI, 0.797–0.928; P = 0.778). When a combined logistic-regression model was constructed, TRX numerically
Fig. 2. Receiver operating characteristic (ROC) curve analysis of Glasgow coma scale (GCS) scores, serum thioredoxin (TRX) levels and GCS scores combined with serum TRX levels for identifying severe traumatic brain injury patients with 1-week mortality.
Fig. 3. Receiver operating characteristic (ROC) curve analysis of Glasgow coma scale (GCS) scores, serum thioredoxin (TRX) levels and GCS scores combined with serum TRX levels for identifying severe traumatic brain injury patients with 6-month mortality.
3.4. 6-month mortality prediction
D.-S. Pan et al. / Clinica Chimica Acta 453 (2016) 62–66
65
improved AUC of GCS score to 0.909 (95% CI, 0.840–0.955; P = 0.098), shown in Fig. 2. 3.5. 6-month unfavorable outcome prediction Fifty-five patients (49.1%) had unfavorable outcome within 6-month after head trauma. Higher serum TRX concentrations and other variables in relation to 6-month unfavorable outcome were shown in Table 3. When all parameters appearing significant in the univariate analysis were further assessed using multivariate regression, (OR, 0.344; 95% CI, 0.120–0.842; P b 0.001) and serum TRX concentrations (OR, 1.189; 95% CI, 1.090–1.311; P b 0.001) emerged statistically significant for the prediction of 6-month unfavorable outcome of patients. Based on the ROC curve, the optimal cutoff value of serum TRX concentrations as an indicator for prognosis of 6-month unfavorable outcome was projected to be 20.3 ng/ml, which yielded a sensitivity of 72.7% and a specificity of 87.7%, with AUC at 0.846 (95% CI, 0.766– 0.907) (Fig. 4). Based on AUC, its predictive value was similar to GCS score's (AUC, 0.852; 95% CI, 0.772–0.912; P = 0.896). We configured a combined logistic-regression model, and therefore, found that TRX statistically significantly improved AUC of GCS score to 0.908 (95% CI, 0.838–0.954; P = 0.032) in Fig. 3. 4. Discussion A recent study has investigated serum TRX concentrations in ischemic stroke and found elevation of this biomarker's concentrations in the peripheral blood [18]. Interestingly, serum TRX concentrations were associated negatively with GCS scores at admission. Importantly, we demonstrated that TRX emerge as a strong and independent prognostic biomarker of 1-week mortality, 6-month mortality and unfavorable outcome, suggesting that TBI patients were in high concentrations of oxidative stress and TRX have the potential to be a good prognostic biomarker after TBI. The pathophysiology of head trauma is dominated by primary brain injury and secondary brain injury associated with activation of inflammatory process and oxidative stress [23–26]. Oxidative stress increases quickly after brain injury, subsequently contributes widely to neuronal death including apoptosis and necrosis, and eventually leading to neurological dysfunction [4–6]. TRX reflects both inflammation and oxidative stress [11,12]. Intracellular TRX is released from cells on oxidative
Fig. 4. Receiver operating characteristic (ROC) curve analysis of Glasgow coma scale (GCS) scores, serum thioredoxin (TRX) levels and GCS scores combined with serum TRX levels for identifying severe traumatic brain injury patients with 6-month unfavorable outcome (defined as the Glasgow outcome scale scores of 1–3).
stress, leading to high extracellular concentrations in numerous situations relevant to critical care, including severe burn injury, acute lung injury, ischemia-reperfusion injury, heart disease and sepsis [13–17]. Our data showed that increased serum TRX concentrations had close relation to GCS scores, indicating TRX might be associated with trauma severity. To date, several biological substances related to oxidative stress, namely, malondialdehyde, F2-isoprostanes, glutathione, and NADPH oxidase have been investigated over the years as peripheral markers of oxidative stress in neurological diseases [27–31]. Disappointingly, these biomarkers have not shown high prognostic predictive values
Table 3 Factors associated with 6-month functional outcome. Characteristics
Unfavorable outcome (n = 55)
Favorable outcome (n = 57)
P value
Male Age (y) GCS score Unreactive pupils CT classification 5 or 6 Abnormal cisterns Midline shift N5 mm Traumatic subarachnoid hemorrhage Intracranial surgery in 1st 24 h Admission time (h) Plasma-sampling time (h) Systolic arterial pressure (mmHg) Diastolic arterial pressure (mmHg) Mean arterial pressure (mmHg) Blood glucose level (mmol/l) Plasma C-reactive protein level (mg/l) Blood white blood cell count (109/l) Blood hemoglobin level (g/l) Blood platelet count (109/l) Serum thioredoxin level (ng/ml)
35 (63.6%) 40.2 ± 12.7 4(3–5) 40 (72.2%) 34 (61.8%) 38 (69.1%) 37 (67.3%) 41 (74.6%) 35 (63.6%) 1.8 ± 1.1 4.1 ± 2.1 120.5 ± 25.4 71.6 ± 11.5 87.9 ± 13.4 13.2 ± 4.3 9.9 ± 2.9 8.2 ± 3.2 122.8 ± 29.3 139.1 ± 37.3 23.6 ± 7.7
37 (64.9%) 38.5 ± 14.4 7(5–8) 12 (21.1%) 21 (36.8%) 15 (26.3%) 20 (35.1%) 24 (42.1%) 26 (45.6%) 2.0 ± 1.2 4.3 ± 2.1 121.1 ± 18.3 72.4 ± 10.1 88.7 ± 10.9 10.7 ± 3.4 8.5 ± 1.8 8.6 ± 2.7 128.5 ± 25.5 143.2 ± 49.8 14.8 ± 4.9
NS NS b0.001 b0.001 0.008 b0.001 0.001 0.001 NS NS NS NS NS NS 0.001 0.002 NS NS NS b0.001
Numerical variables were presented as mean ± SD or median (the upper and lower quartiles) and analyzed by unpaired student's t test or Mann–Whitney U-test. Categorical variables were expressed as counts (percentage) and analyzed by χ2test or Fisher exact test. GCS indicates Glasgow Coma Scale; CT, computerized tomography.
66
D.-S. Pan et al. / Clinica Chimica Acta 453 (2016) 62–66
compared with other clinical grade systems like GCS [7–10]. Recently, Qi et al. found that TRX was an independent prognostic maker of 3-month functional outcome indicated by modified Rankin scale; Importantly, TRX improved the AUC of National Institutes of Health Stroke Scale score for functional outcome [18]. In the current study, besides 6month function outcome, 1-week mortality and 6-month mortality were also regarded as the parameters of clinical outcome. Multivariate analysis demonstrated that serum TRX was an independent prognostic predictor of these short-term or long-term clinical outcomes. Furthermore, we found that predictive value of serum TRX concentrations were similar to that of GCS scores; meanwhile, TRX statistically significantly improved the predictive performance of GCS scores for 6month unfavorable outcome, but not for 1-week mortality and 6mortality. Hence, there was a little difference in the prognostic predictive value of TRX between mortality and function outcome. However, the difference in the prediction of 6-month mortality had trend to statistical significance (P value = 0.098). Therefore, it is assumed that this difference might be derived from a small sample size. As a whole, TRX might be a good prognostic predictive biomarker of TBI. Our study had certain limitations. Firstly, the present work was a single center study. The results remained to be verified in a multicenter study. Secondly, we did not perform an analysis of serum TRX concentrations at different time points during follow-up to investigate the evolution of this parameter. Thus, this study might be regarded as a pilot one. 5. Conclusions The novel finding of our study was that serum TRX concentrations are associated with trauma severity and clinical outcomes including 1week mortality, 6-month mortality and 6-month unfavorable outcome. Moreover, TRX can improve the predictive value of GCS scores. Therefore, TRX could be used as a prognostic biomarker in patients with severe TBI. Abbreviations GCS TRX TBI
Glasgow coma scale thioredoxin traumatic brain injury
Acknowledgments The authors thank all staffs in Department of Neurosurgery, The First Affiliated Hospital, School of Medicine, Zhejiang University for their technical support. References [1] J. Sahuquillo, M.A. Poca, S. Amoros, Current aspects of pathophysiology and cell dysfunction after severe head injury, Curr. Pharm. Des. 7 (2001) 1475–1503. [2] P. Fernandez, D. Diaz, Experimental animal models of traumatic brain injury: medical and biomechanical mechanism, Crit. Rev. Neurosurg. 9 (1999) 44–52. [3] C. Cornelius, R. Crupi, V. Calabrese, et al., Traumatic brain injury (TBI): oxidative stress and neuroprotection, Antioxid. Redox Signal. 19 (2013) 836–853. [4] N. Şenol, M. Nazıroğlu, Melatonin reduces traumatic brain injury-induced oxidative stress in the cerebral cortex and blood of rats, Neural Regen. Res. 9 (2014) 1112–1116.
[5] W. Zhang, B. Li, Y. Guo, et al., Rhamnetin attenuates cognitive deficit and inhibits hippocampal inflammatory response and oxidative stress in rats with traumatic brain injury, Cent. Eur. J. Immunol. 40 (2015) 35–41. [6] K.S. Panickar, R.A. Anderson, Effect of polyphenols on oxidative stress and mitochondrial dysfunction in neuronal death and brain edema in cerebral ischemia, Int. J. Mol. Sci. 12 (2011) 8181–8207. [7] C. Nayak, D. Nayak, S. Bhat, A. Raja, A. Rao, Relationship between neurological outcome and early oxidative changes in erythrocytes in head injury patients, Clin. Chem. Lab. Med. 45 (2007) 629–633. [8] A. Paolin, L. Nardin, P. Gaetani, et al., Oxidative damage after severe head injury and its relationship to neurological outcome, Neurosurgery 51 (2002) 949–954. [9] L. Lorente, M.M. Martín, P. Abreu-González, et al., Association between serum malondialdehyde levels and mortality in patients with severe brain trauma injury, J. Neurotrauma 32 (2015) 1–6. [10] A. Hohl, S. Gullo Jda, C.C. Silva, et al., Plasma levels of oxidative stress biomarkers and hospital mortality in severe head injury: a multivariate analysis, J. Crit. Care 27 (2012) 523. [11] A. Burke-Gaffney, M.E. Callister, H. Nakamura, Thioredoxin: friend or foe in human disease? Trends Pharmacol. Sci. 26 (2005) 398–404. [12] Thioredoxin in sepsis: just another biomarker or a plausible therapeutic target? Crit. Care Med. 37 (2009) 2304–2305. [13] A. Burke-Gaffney, T.W. Evans, G.J. Quinlan, A. Abdiu, H. Nakamura, B. Sahaf, J. Yodoi, A. Holmgren, A. Rosen, Thioredoxin blood level increases after severe burn injury, Antioxid. Redox Signal. 2 (2000) 707–716. [14] M.E. Callister, A. Burke-Gaffney, G.J. Quinlan, et al., Extracellular thioredoxin levels are increased in patients with acute lung injury, Thorax 61 (2006) 521–527. [15] M.E. Callister, A. Burke-Gaffney, G.J. Quinlan, H. Betts, S. Nadel, T.W. Evans, Persistently low plasma thioredoxin is associated with meningococcal septic shock in children, Intensive Care Med. 33 (2007) 364–367. [16] A. Jekell, A. Hossain, U. Alehagen, U. Dahlstrom, A. Rosen, Elevated circulating levels of thioredoxin and stress in chronic heart failure, Eur. J. Heart Fail. 6 (2004) 883–890. [17] S.K. Leaver, N.S. MacCallum, V. Pingle, et al., Increased plasma thioredoxin levels in patients with sepsis: positive association with macrophage migration inhibitory factor, Intensive Care Med. 36 (2010) 336–341. [18] A.Q. Qi, Y. Li, Q. Liu, et al., Thioredoxin is a novel diagnostic and prognostic marker in patients with ischemic stroke, Free Radic. Biol. Med. 80 (2015) 129–135. [19] L.F. Marshall, S.B. Marshall, M.R. Klauber, M.V. Clark, A new classification of head injury based on computerized tomography, J. Neurosurg. 75 (1991) S14–S20. [20] B. Jennett, M. Bond, Assessment of outcome after severe brain damage, Lancet 1 (1975) 480–484. [21] Z.Y. Zhang, L.X. Zhang, X.Q. Dong, et al., Comparison of the performances of copeptin and multiple biomarkers in long-term prognosis of severe traumatic brain injury, Peptides 60 (2014) 13–17. [22] K.Y. Wang, G.F. Yu, Z.Y. Zhang, Q. Huang, X.Q. Dong, Plasma high-mobility group box 1 levels and prediction of outcome in patients with traumatic brain injury, Clin. Chim. Acta 413 (2012) 1737–1741. [23] B. Halliwell, Reactive oxygen species and the central nervous system, J. Neurochem. 59 (1992) 1609–1623. [24] P.M. Kochanek, C.E. Dixon, D.K. Shellington, et al., Screening of biochemical and molecular mechanisms of secondary injury and repair in the brain after experimental blast-induced traumatic brain injury in rats, J. Neurotrauma 30 (2013) 920–937. [25] D. Brough, A. Denes, Interleukin-1α and brain inflammation, IUBMB Life 67 (2015) 323–330. [26] K.N. Corps, T.L. Roth, D.B. McGavern, Inflammation and neuroprotection in traumatic brain injury, JAMA Neurol. 72 (2015) 355–362. [27] M. Seneş, N. Kazan, O. Coşkun, O. Zengi, L. Inan, D. Yücel, Oxidative and nitrosative stress in acute ischaemic stroke, Ann. Clin. Biochem. 44 (2007) 43–47. [28] F. Içme, Ö. Erel, A. Avci, S. Satar, M. Gülen, S. Acehan, The relation between oxidative stress parameters, ischemic stroke, and hemorrhagic stroke, Turk. J. Med. Sci. 45 (2015) 947–953. [29] J. Kotur-Stevuljevic, N. Bogavac-Stanojevic, Z. Jelic-Ivanovic, et al., Oxidative stress and paraoxonase 1 status in acute ischemic stroke patients, Atherosclerosis 241 (2015) 192–198. [30] D. Paspalj, P. Nikic, M. Savic, et al., Redox status in acute ischemic stroke: correlation with clinical outcome, Mol. Cell. Biochem. 406 (2015) 75–81. [31] N.W. Tsai, Y.T. Chang, C.R. Huang, et al., Association between oxidative stress and outcome in different subtypes of acute ischemic stroke, Biomed. Res. Int. 2014 (2014) 256879.
