Prognostic Significance of Uric Acid Levels in Ischemic Stroke Patients

Transcription

1 Neurotox Res (2016) 29:10 20 DOI /s CLINICAL RESEARCH REPORT Prognostic Significance of Uric Acid Levels in Ischemic Stroke Patients Xia Zhang 1 Zhi-Chao Huang 1 Tao-Sheng Lu 2 Shou-Jiang You 1 Yong-Jun Cao 1 Chun-Feng Liu 1 Received: 13 May 2015 / Revised: 30 July 2015 / Accepted: 9 September 2015 / Published online: 16 September 2015 Springer Science+Business Media New York 2015 Abstract The importance and function of serum uric acid (UA) levels in patients with cardiovascular disease or stroke are unclear. We sought to evaluate the appropriate UA levels for stroke patients and the association between endogenous UA levels and clinical outcomes in acute ischemic stroke (AIS) patients, particularly regarding the possible interaction between gender and UA levels with respect to AIS prognosis. We examined 303 patients who had an onset of ischemic stroke within 48 h. Of those, 101 patients received thrombolytic treatment. Serum UA (lmol/l) levels were measured the second morning after admission. Patient prognosis was evaluated 90 days after clinical onset by modified Rankin Scale. Patients were divided into four groups according to serum UA quartiles. A binary multivariate logistic regression model was used to assess clinical relevance in regard to functional outcome and endogenous UA levels. Analysis of subgroups by gender and normal glomerular filtration rate were also been done. Poor functional outcome was associated with older age, history of atrial fibrillation, or higher baseline National Institutes of Health Stroke Scale scores. After adjustment for potential confounders, patients with higher UA levels Xia Zhang and Zhi-Chao Huang have contributed equally to this work. & Yong-Jun Cao yongjuncao@126.com & Chun-Feng Liu liucf@suda.edu.cn; liucf20@hotmail.com 1 2 Department of Neurology, The Second Affiliated Hospital of Soochow University, 1055 Sanxiang Road, Suzhou , China Department of Neurology, Changshu No. 1 People s Hospital, Changshu , China ([380 lmol/l) or lower UA levels (B250 lmol/l) were 2 3 times more likely to have a poor outcome (OR 2.95, 95 % CI ; OR 2.78, 95 % CI , respectively) compared to the baseline group (UA level lmol/l). The same results were observed in thrombolyzed patients. Patients with high and low UA levels were 9 18 times more likely to having poor outcomes compared to the baseline group (UA level: lmol/l; OR 18.50, 95 % CI: ; OR 9.66, 95 % CI , respectively). In men, patients with high UA levels were 6 times more likely to have poor outcomes compared to the baseline group (UA level: lmol/l; OR 6.10, 95 % CI ). However, female patients with UA level lmol/l were seven times more likely to perform badly compared to the baseline group (UA level[337 lmol/l, OR 7.06, 95 % CI ). Serum UA levels in an appropriate range were associated with better outcome in patients with AIS but may be harmful when too high or too low. The association of UA levels with AIS prognosis differed in male and female patients, which highlights the necessity of stratifying by gender in investigations of cerebrovascular risk factors. Keywords Acute ischemic stroke Uric acid level Thrombolysis Prognosis Functional outcome Gender Introduction Uric acid (UA) is the end product of the metabolism of purine nucleotides, which are principal constituents of DNA, RNA, and cellular energy stores. Until recently, its antioxidant properties had not been considered (Becker 1993). Many research findings have revealed that UA has

2 Neurotox Res (2016) 29: the antioxidant capacity to decrease oxidative stress and is considered neuroprotective (Waring et al. 2001; Chamorro et al. 2002). In patients with acute stroke, serum UA levels were decreased a few days after stroke onset (Hong et al. 2010). The administration of UA decreases lipid peroxidation and prevents an early fall of serum UA in patients treated with recombinant tissue plasminogen activator (rt- PA) within 3 h of stroke onset, which makes the administration of exogenous UA a possible choice for acute cerebral infarction treatment (Amaro et al. 2008). However, some studies showed an association between higher UA levels as an independent risk factor of arterial atherosclerosis and increased cardio-cerebrovascular events (Bos et al. 2006). One study showed increased levels of serum UA were associated with poorer stroke outcome (Weir et al. 2003). Thus, the nature of the relationship between UA and cardiovascular events or stroke remains controversial. We speculated that appropriate UA concentration could be beneficial for stroke patients and that too high or too low UA concentration might be harmful to stroke patients. Therefore, in this study, we aimed to examine the appropriate UA level and assess the association between endogenous UA levels and clinical outcomes in ischemic stroke patients, some of whom were treated with thrombolysis. We particularly regarded the possible interaction between gender and UA levels with respect to acute ischemic stroke (AIS) prognosis. Materials and Methods Study Objects This study included a total of 303 patients over 18 years old with AIS of the anterior circulation within 48 h onset at the Department of Neurology, the Second Affiliated Hospital of Soochow University, Suzhou, China, from June 2010 to December The diagnosis of ischemic stroke was accorded with diagnostic criteria of WHO: acute onset with focal defects of neurological function lasting more than 24 h ruled out non-vascular causes. Patients mainly manifested central facial tongue paralysis, hemiplegia and hemidysesthesia, aphasia, cerebral cortex dysfunction such as mental symptoms, etc., which were characteristics of anterior circulation eventually confirmed by head CT or MRI scans. Other inclusion criteria included first attack or without obvious sequelae after previous attacks of stroke (mrs B 1). Among these, 101 patients received rt-pa thrombolytic treatment. Patients were enrolled for thrombolytic treatment if they were between 18 and 80 years old; they were within 4.5 h onset; head CT scans showed no hemorrhage and no early signs of massive cerebral infarction; and informed consent forms were signed. Exclusion criteria are as follows: autoimmune diseases; thyroid dysfunction; tumors; acute or chronic infection; severe hepatic, renal, cardiac or respiratory diseases; obvious sequelae of previous stroke, and pregnancy. Severe hepatic, renal, cardiac, or respiratory diseases were defined as active liver diseases, alanine aminotransferase (ALT) or aspartate aminotransferase (AST) over three times of ULN, renal dysfunction, cardiac, or respiratory diseases which would influence patients prognosis. The research protocol was approved by the local Ethics Committee of Soochow University, and all patients or their family members signed informed consent forms. Data Collection and Outcome Evaluation Demographics, including the history of stroke, the prevalence of vascular risk factors, and stroke subtypes, were collected. Risk factors for stroke or vascular disease, such as hypertension, diabetes mellitus (DM), atrial fibrillation, hyperlipidemia, coronary artery disease (CAD), smoking, and previous stroke, were ascertained by standard criteria. History of medication was thoroughly investigated. Drugs such as diuretics, beta-blockers, aspirin, and calcium channel blockers (CCB) were recorded. Fasting blood samples were extracted in the next morning after admission for analysis of serum UA using standard laboratory procedures with urate oxidase reagent on a COBAS 8000 analyzer (Roche, Basel, Switzerland) with an inter-assay coefficient of variation \3 to 5 %. Hepatic and renal tests, fasting glucose, cholesterol, triglycerides, low-density, and high-density lipoprotein concentrations were also performed. Glomerular filtration rate (GFR) was calculated by simplified MDRD equation. The National Institutes of Health Stroke Scale (NIHSS) scores were used by stroke neurologists to assess neurological deficit when the patients were admitted. Stroke etiologic subtypes were classified according to ASCO Classification: A for atherosclerosis, S for small vessel disease, C for cardiac source, and O for other cause. All the patients had a follow-up of 90 days after discharge. The prognosis outcome was evaluated by modified Rankin Scale (mrs) at 90 days. mrs scores of \2 or C2 were defined as a good or poor outcome, respectively. Patients with thrombolytic treatment were given rt-pa (Boehringer Ingelheim, Ingelheim am Rhein, Germany) with a dosage of 0.9 mg/kg (the maximum dose was 90 mg). Brain CT scans were repeated at 24 h or whenever it was required in patients with worsening situations. Systematic hemorrhagic transformation (sht) was defined as at least a 4-point increment in the NIHSS scores.

3 12 Neurotox Res (2016) 29:10 20 Statistical Analysis The linearity of the relationship between urate and prognostic outcomes was assessed using the change in the -2log likelihoods between the linear and quadratic models. As the relationship was non-linear, UA concentration was divided into quartiles and assessed categorically. Results were expressed as a percentage, median with interquartile ranges, or mean ± SD. Differences in frequencies were compared with the v 2 test. Comparisons of median values among groups were done by the Student t test, one-way ANOVA, Mann Whitney, non-parametric tests, or Kruskal Wallis tests, as appropriate. We fitted a logistic model to estimate the association between serum UA levels and prognosis outcome with OR (odds ratio) and 95 % CI (confidence interval) after adjusting for potential confounders: sex (female = 0, male = 1), age, smoker (no = 0, yes = 1), atrial fibrillation (AF, no = 0, yes = 1), diabetes (no = 0, yes = 1), hypertension (no = 0, yes = 1), coronary artery disease (CAD, no = 0, yes = 1), previous stroke (no = 0, yes = 1), thrombolytic treatment (no = 0, yes = 1), baseline NIHSS scores, hyperlipidemia (no = 0, yes = 1), ASCO classification (A = 1, S = 2, C = 3, O = 4), GFR, and application of these drugs (no = 0, yes = 1). All statistical analyses were two tailed. A p value\0.05 was considered statistically significant. SPSS16.0 software (SPSS Inc., IBM, Chicago, IL, USA) was used for all analyses. Results Chief Characteristics of the Study Objects There were 303 stroke patients examined in this study. The mean age was 64.6 ± 12.8 years; 66.3 % of patients were male; and the median NIHSS score on admission was 7(4 12). There were 101 (33.3 %) patients receiving thrombolytic treatment. 11(3.6 %) patients were diagnosed gout in the medical history previously. Although a linear relationship was not observed between age and UA levels, males had higher levels of UA than females (338.5 ± 92.0 vs ± 86.0 lmol/l, p \ 0.001) and patients with a history of atrial fibrillation had higher levels than those without it (349.5 ± vs ± 89.8 lmol/l, p = 0.022). UA levels was higher in patients who were prescribed diuretics, beta-blockers, aspirin, or CCB (329.5 ± 96.8 vs ± 85.9 lmol/l, p = 0.015). Serum urate was apt to be higher among those with a history of hyperlipidemia and current smoking (p = 0.054, 0.052, respectively). Hypertension, diabetes, history of prior stroke, and CAD were not significantly relevant to UA levels (data not shown). A significant correlation was observed between GFR and UA levels (r = 0.299, p \ 0.001). A total of 107(35.3 %) patients had poor outcome. Of those 107 patients, there were 31, 24, 20, and 32 cases with poor outcomes in the four serum UA level groups, from the lowest to the highest UA levels, respectively. As shown in Table 1, poor outcome at 90 days was associated with older age, history of atrial fibrillation, or higher baseline NIHSS scores. Patients without thrombolysis had worse outcome at 90 days than those receiving it. However, UA levels in patients with good outcome were not significantly different from those in patients with poor outcome (317.6 ± 83.6 vs ± lmol/l, p = 0.606). Patients were stratified into four groups according to serum UA quartiles: UA level B250, , , and [380 lmol/l. Among the four groups, there were no significant differences in history of AF, hypertension, DM, hyperlipidemia, smoking, stroke, and CAD. Age, proportion of thrombolysis, medication history, poor outcomes, and baseline NIHSS scores did not differ among the four groups (p \ 0.05 for all). Sex and GFR differed among the four groups (p \ for both). Independent Clinical Outcome Predictors: Role of UA Patients with low (B250 lmol/l) and high ([380 lmol/l) UA levels had poorer functional outcomes (41.9 and 42.1 %, respectively). Patients with UA levels from 316 to 380 lmol/ L had the lowest poor outcome incidence (27.0 %), thus the group with UA levels from 316 to 380 lmol/l was used as a reference. As seen in Table 3, after adjusting for potential confounders, compared to the baseline group (UA level: lmol/l), patients who had higher UA levels ([380 lmol/l) or lower UA levels (B250 lmol/l) were 2 3 times more likely to having poor outcome risk (OR 2.95, 95 % CI ; OR 2.78, 95 % CI , respectively). Furthermore, older age, higher baseline NIHSS scores, history of hypertension, and no thrombolysis treatment were associated with the increased risk of poor outcome (Table 2). Gender Difference of the Role of UA Levels on Functional Outcome Definition of hyperuricemia between male and female is different and the upper end of normal range is lower in women than in men. In our study, males had higher levels of UA than females (p \ 0.001) and high UA quartiles had more percents of men. 201 male patients were divided into four groups according to their serum UA quartiles: UA level B278, , , and[402 lmol/l. Among the four groups, there were no significant differences in history of hypertension, DM, hyperlipidemia, smoking,

4 Neurotox Res (2016) 29: Table 1 Main characteristics of patients according to functional outcome at day 90 (n = 303) Total Good outcome n (%) Poor outcome n (%) p value B (58.1) 31 (41.9) (69.6) 24 (30.4) (73.0) 20 (27.0) [ (57.9) 32 (42.1) Mean (SD) (93.9) (83.6) (110.6) Sex Male (67.3) 69 (64.5) Female (32.7) 38 (35.5) Age, years, mean (SD) 64.6 (12.8) 63.2 (13.0) 67.2 (12.1) Smokers, no. (%) 109 (36.0) 72 (36.7) 37 (34.6) Atrial fibrillation, no. (%) 45 (14.8) 23 (11.7) 22 (20.6) Diabetes, no. (%) 68 (22.4) 39 (20.0) 29 (27.1) Hypertension, no. (%) 246 (81.2) 163 (83.2) 83 (77.6) CAD, no. (%) 29 (9.6) 17 (8.7) 12 (11.2) Previous stroke, no. (%) 53 (17.5) 30 (18.1) 23 (21.5) Hyperlipidemia, no. (%) 76 (25.1) 47 (24.0) 29 (27.1) Thrombolysis treatment, no. (%) 101 (33.3) 83 (42.3) 31 (29.0) Baseline NIHSS scores 7 (4 12) 5.8 (3.9) 13.4 (6.0) \0.001 Median (interquartiles) GFR, mean (SD) (30.8) (29.6) 98.2 (32.6) Drug, no. (%) 107 (35.3) 66 (33.7) 41 (38.3) ASCO classification A, no. (%) 241 (79.5) 159 (66.0) 82 (34.0) S, no. (%) 9 (2.6) 7 (77.8) 2 (22.2) C, no. (%) 38 (11.9) 21 (55.3) 17 (44.7) O, no. (%) 15 (5.0) 9 (60.0) 6 (40.0) CAD means coronary artery disease, GFR means glomerular filtration rate, calculated by simplified MDRD equation: GFR (ml/ min ) = 186*(Scr)-1.154*(age)-0.203*(0.742 female), Scr means serum creatinine (mg/dl), age (years) stroke, and CAD. Age, proportion of thrombolysis, medication history, and baseline NIHSS scores did not differ among the four groups (p \ 0.05 for all), but GFR and history of AF differed (p \ 0.001, p = 0.035, respectively). Functional prognosis among the four groups differed significantly (p = 0.003). Patients in UA level lmol/l group had the lowest poor outcome incidence (20.4 %); however, patients in high UA level group ([402 lmol/l) got the worst outcomes (54.3 %). The group with UA levels from 279 to 334 lmol/l were used as a reference. As seen in Table 4, after adjusting for potential confounders, compared to the baseline group (UA level: lmol/l), patients who had higher UA levels ([402 lmol/l) were six times more likely to having poor outcome risk (OR 6.10, 95 % CI ). In female subgroup, patients were also divided into four groups according to their serum UA levels: UA level B210, , , and [337 lmol/l. Among the four groups, there were no significant differences in history of AF, hypertension, DM, hyperlipidemia, smoking, stroke, and CAD. Age, proportion of thrombolysis, medication history, poor outcomes, and baseline NIHSS scores did not differ among the four groups (p \ 0.05 for all) except GFR (p = 0.012). In the four groups according to UA quartiles, patients with high ([337 lmol/l) UA levels had better functional outcomes (20.0 %); however, patients with UA levels lmol/l got the poorest outcomes (53.9 %). The group with high ([337 lmol/l) UA levels was used as a reference. As seen in Table 5, after adjusting for potential confounders, compared to the baseline group (UA level [337 lmol/l), patients who had UA levels ( lmol/l) were seven times more likely to having poor outcome risk (OR 7.06, 95 % CI ).

5 14 Neurotox Res (2016) 29:10 20 Table 2 Baseline characteristics of patients according to serum UA quartiles Characteristics Serum UA quartiles (lmol/l) p value B [380 No. (%) 74 (24.4) 74 (24.4) 79 (26.1) 76 (25.1) Sex [male no. (%)] 53 (71.6) 34 (45.9) 51 (64.6) 63 (82.9) \ Age (years, mean ± SD) 64.4 ± ± ± ± AF (no. (%)) 8 (10.8) 9 (12.2) 11 (13.9) 17 (22.4) HBP (mmhg, no. (%)) 59 (79.7) 60 (81.1) 66 (83.5) 61 (80.3) DM (mmol/l, no. (%)) 19 (25.7) 17 (23.0) 20 (25.3) 12 (15.8) Hyperlipidemia [no. (%)] 18 (24.3) 14 (18.9) 18 (22.8) 26 (34.2) Smoke [no. (%)] 28 (37.8) 20 (27.0) 28 (35.4) 33 (43.4) Previous stroke (no. (%)) 11 (14.9) 14 (18.9) 17 (21.5) 11 (14.5) CAD (no. (%)) 3 (4.1) 8 (10.8) 9 (11.4) 9 (11.8) ASCO [no. (%)] A 63 (85.1) 60 (81.1) 63 (77.8) 57 (75.0) S 3 (4.1) 1 (1.3) 3 (3.8) 2 (2.6) C 6 (8.1) 7 (9.5) 9 (11.4) 14 (18.4) O 2 (2.7) 6 (8.1) 4 (5.1) 3 (3.9) Thrombolysis [no. (%)] 21 (28.4) 29 (39.2) 27 (34.2) 24 (31.6) Baseline NIHSS scores [mean(quartiles)] 7 (4 12) 8 (5 12) 7 (4 10) 7 (4 13.7) Poor prognosis [no. (%)] 22 (29.7) 30 (40.5) 24 (30.4) 31 (40.8) Drugs [no. (%)] 26 (35.1) 29 (39.2) 34 (43.0) 18 (23.7) 0.07 GFR (mean ± SD) 99.1 ± ± ± ± 30.0 \ CAD means coronary artery disease, AF means atrial fibrillation, HBP means hypertension, DM means diabetes mellitus, A means atherosclerosis, S means small vessel disease, C means cardiac sources, O means other causes, GFR means glomerular filtration rate, calculated by simplified MDRD equation: GFR(ml/min ) = 186*(Scr)-1.154*(age)-0.203*(0.742 female), Scr means serum creatinine (mg/dl), age (years) The Role of UA Levels on Functional Outcome of Patients with Normal GFR Another major concern is the impact of GFR on the UA level. To get rid of its impact on outcomes, 208 patients with normal GFR were selected for further analysis, who were divided into four groups according to UA quartiles: UA level B244, , , and [372 lmol/l. Among the four groups, there were no significant differences in history of AF, hypertension, DM, hyperlipidemia, smoking, stroke, and CAD. Age, proportion of thrombolysis, medication history, poor outcomes, and baseline NIHSS scores did not differ among the four groups (p \ 0.05 for all). Sex and GFR differed (both p \ 0.001). In the four groups, patients with UA levels lmol/ L had better functional outcomes(25.0 %). The group was used as a reference. As seen in Table 6, after adjusting for potential confounders, compared to the baseline group (UA level: lmol/l), patients who had low UA levels (B244 lmol/l) were two times more likely to having poor outcome risk (OR 2.34, 95 % CI ). The Role of UA Levels on Functional Outcome of Patients Receiving Thrombolysis 101 patients with thrombolytic treatment were enrolled for further analysis. 15 patients (51.7 %) in the low UA group and 14 patients (58.3 %) in the high UA group had poor prognosis at 90 days. However, only 6 cases (28.6 %) in the UA level lmol/l group performed poorly at 90 days. Among the four groups, there were no significant differences in history of AF, hypertension, DM, hyperlipidemia, smoking, stroke, and CAD. Age, proportion of thrombolysis, medication history, poor outcomes, baseline NIHSS scores, and GFR did not differ among the four groups (p [ 0.05 for all) except sex (p = 0.001). In this thrombolysis-treated subgroup, after being adjusted for several potential confounding factors, patients with the highest UA level ([380 lmol/l) or the lowest UA level (B250 lmol/l) were 9 18 times more likely to having poor outcome risk compared to the baseline group (UA level: lmol/l; OR 9.66, 95 % CI ; OR 18.50, 95 % CI respectively,

6 Neurotox Res (2016) 29: see Table 7). Older age, higher baseline NIHSS scores, and history of hypertension were independently associated with increased risk of poor outcome (Table 3). Discussion Several studies have provided conflicting results about the clinical significance of elevated UA in patients with cardiovascular or cerebrovascular diseases. Previous studies reported patients with higher UA levels have an increased risk for stroke or excess mortality, especially in patients with non-insulin-dependent diabetes (Amaro et al. 2007), or isolated systolic hypertension (Wang et al. 1998). The addition of UA to thrombolytic therapy did not increase the proportion of patients achieving excellent outcome after stroke in a recent study (Lehto et al. 1998). However, some studies showed a better prognosis in stroke patients with high UA levels (Zhang et al. 2010; Amaro and Chamorro 2011). Therefore, to explore this question, we analyzed the prognosis in patients with or without thrombolytic treatment. The results showed patients with a UA level from 316 to 380 lmol/l had a better clinical outcome with or without thrombolytic treatment, whereas a worse outcome was associated with low or high UA levels. This demonstrated that having a UA level between 316 and 380 lmol/ L was beneficial to stroke patients. The strength in the study lie in use of a functional scale widely accepted in clinical therapeutic trials, enrollment of thrombolyzed patients, stratified analysis by gender and one of few such studies in Chinese population. A meta-analysis in 2015 including ten eligible studies with a total of 8131 cases reported high serum UA level was associated with better outcomes after AIS(Wang et al. 2015), which was contrary to our results. It included the trials in which the serum UA levels were measured within 24 h after the onset of ischemic stroke and the definition of poor outcomes was mostly mrs [ 2. These might explain the difference of results between our study and this metaanalysis. UA levels change in various stages of stroke. Hong et al. (Hong et al. 2010) reported that UA levels dropped at 48 h and gradually increased in a U-shaped pattern. The changes of UA during follow-up may have an important influence on the association between UA at the onset and patients prognosis. In our study, the serum UA levels were measured within h, so the prognostic role of UA levels might change. In our study, patients with thrombolysis were also enrolled. Besides, patients with a UA level from 316 to Table 3 The association between UA levels and the risk of poor outcome Variables ß S.E. Wald-v 2 p OR (95 % CI) (ref) B ( ) ( ) [ ( ) Sex ( ) Age ( ) AF ( ) Hypertension ( ) DM ( ) Smokers ( ) CAD ( ) Previous stroke ( ) Hyperlipidemia ( ) Thrombolysis treatment ( ) ASCO ( ) Baseline NIHSS scores ( ) Drugs ( ) GFR ( ) Binary logistic regression estimates as appropriate. The group with UA levels from 316 to 380 lmol/l was used as the reference group CAD indicates coronary artery disease, A means atherosclerosis, S means small vessel disease, C means cardiac sources, O means other causes, GFR means glomerular filtration rate, calculated by simplified MDRD equation: GFR(ml/min ) = 186*(Scr)-1.154*(age)-0.203*(0.742 female), Scr means serum creatinine (mg/dl), age (years)

7 16 Neurotox Res (2016) 29:10 20 Table 4 The association between UA levels and the risk of poor outcome in males (n = 201) Variables ß S.E. Wald-v 2 p OR (95 % CI) (ref) B ( ) ( ) [ ( ) Age ( ) AF ( ) Hypertension ( ) DM ( ) Smokers ( ) CAD ( ) Previous stroke ( ) Hyperlipidemia ( ) ASCO ( ) Baseline NIHSS scores ( ) Thrombolysis treatment ( ) Drugs ( ) GFR ( ) Binary logistic regression estimates as appropriate. The group with UA levels from 279 to 334 lmol/l was used as the reference group CAD indicates coronary artery disease, A means atherosclerosis, S means small vessel disease, C means cardiac sources, O means other causes, GFR means glomerular filtration rate, calculated by simplified MDRD equation: GFR(ml/min ) = 186*(Scr)-1.154*(age)-0.203*(0.742 female), Scr means serum creatinine (mg/dl), age (years) Table 5 The association between UA levels and the risk of poor outcome in females (n = 102) Variables ß S.E. Wald-v 2 p OR (95 % CI) [ (ref) B ( ) ( ) ( ) Age ( ) AF ( ) Hypertension ( ) DM ( ) CAD ( ) Previous stroke ( ) Hyperlipidemia ( ) ASCO ( ) Baseline NIHSS scores ( ) Thrombolysis treatment ( ) Drugs ( ) GFR ( ) Binary logistic regression estimates as appropriate. The group with UA levels[337 lmol/l was used as the reference group CAD indicates coronary artery disease, A means atherosclerosis, S means small vessel disease, C means cardiac sources, O means other causes, GFR means glomerular filtration rate, calculated by simplified MDRD equation: GFR(ml/min ) = 186*(Scr)-1.154*(age)-0.203*(0.742 female), Scr means serum creatinine (mg/dl), age(years)

8 Neurotox Res (2016) 29: Table 6 The association between UA levels and the risk of poor outcome in patients with normal GFR (n = 208) Variables ß S.E. Wald-v 2 p OR (95 % CI) (ref) B ( ) ( ) [ ( ) Sex ( ) Age ( ) AF ( ) Hypertension ( ) DM ( ) CAD ( ) Previous stroke ( ) Hyperlipidemia ( ) ASCO ( ) Baseline NIHSS scores ( ) Drugs ( ) Thrombolysis treatment ( ) Binary logistic regression estimates as appropriate. The group with UA levels lmol/l was used as the reference group CAD indicates coronary artery disease, A means atherosclerosis, S means small vessel disease, C means cardiac sources, O means other causes, GFR means glomerular filtration rate, calculated by simplified MDRD equation: GFR(ml/min ) = 186*(Scr)-1.154*(age)-0.203*(0.742 female), Scr means serum creatinine (mg/dl), age(years) Table 7 The association between UA levels and the risk of poor outcome in thrombolytic patients (n = 101) Variables ß S.E. Wald-v 2 p OR (95 % CI) (ref) B ( ) ( ) [ ( ) Sex ( ) Age ( ) AF ( ) Hypertension ( ) DM ( ) Smokers ( ) CAD ( ) Previous stroke ( ) Hyperlipidemia ( ) ASCO ( ) Baseline NIHSS scores ( ) Drugs ( ) GFR ( ) Binary logistic regression estimates as appropriate. The group with UA levels from 316 to 380 lmol/l was used as the reference group CAD indicates coronary artery disease, A means atherosclerosis, S means small vessel disease, C means cardiac sources, O means other causes, GFR means glomerular filtration rate, calculated by simplified MDRD equation: GFR(ml/min ) = 186*(Scr)-1.154*(age)-0.203*(0.742 female), Scr means serum creatinine (mg/dl), age(years)

9 18 Neurotox Res (2016) 29: lmol/l had a better clinical outcome, which was different from the results reported by Logallo et al They found UA level showed a positive correlation with clinical improvement and was an independent predictor for favorable stroke outcome. The differences of the results may be explained by the time window for serum UA measurements and subtypes of ischemic stroke. Logallo et al. focused their study on serum UA levels measured within the first 3 h after stroke onset. Prognostic role of UA measurements in different time windows may vary. Serum UA measurements within h in our study may play a dual role in neurological function of AIS patients. On one hand, UA is the most important endogenous antioxidant and has neuroprotective effects in preclinical studies (Yu et al. 1998; Romanos et al. 2007). A study has shown that administration of UA within hours of ictus in those with thrombolytic treatment for ischemic stroke yields lower lipid peroxidation and less activation of active matrix metalloproteinases (Muir et al. 2008). A UA level that is too low to scavenge hydroxyl radicals, hydrogen peroxide, and peroxynitrite, which promote lipid peroxidation, results in infarction enlargement, and poor prognosis. Similarly, in our study, patients in the low UA level group had a trend toward poor outcome. There are experimental data supporting these findings. For example, UAtreated neurons had suppressed oxyradical accumulation, stabilized calcium homeostasis, and preserved mitochondrial function (Yu et al. 1998). On the other hand, an association between elevated UA and worse outcome after stroke has been observed. A number of large, well-conducted epidemiological studies demonstrated that elevated serum UA is a powerful predictor of increased risk of a cardiovascular event and mortality (Dawson and Walters 2006). The high UA levels have been also associated with hypertension (Johnson et al. 2003), dyslipidemia, type 2 diabtes (Nakanishi et al. 2003), kidney disease, cardiovascular, and cerebrovascular events (Feig et al. 2008), which will worsen functional outcome of AIS patients. Small increments in UA are associated with reduced likelihood of favorable outcome at 90 days and increased risk of recurrent vascular events (Cherubini et al. 2000). Our results in the high UA level group support this. This may result from harmful effects of UA on platelet, smooth muscle, oxidation of low-density lipoprotein, and lipid peroxidation or endothelial function (Dawson et al. 2007), which all play important roles in the progression of arterial atherosclerosis. Community research has observed an association between UA levels and carotid intima-media thickness (IMT) (Iribarren et al. 1996). Human atherosclerotic plaques contain more urate crystals than normal arterial wall (Suarna et al. 1995). In some specific circumstances, such as low serum ascorbate levels, UA promotes oxidative stress (Glantzounis et al. 2005). It has been proposed the UA may show both anti- and pro-oxidant properties depending on levels of other antioxidants, levels of oxidative stress, and time of interaction with the target tissues, and that the balance between the anti- and pro-oxidant properties shifts in favor of neuroprotection in conditions of extraordinary oxidative stress such as AIS. Thus h after onset of ischemic stroke, high UA levels shift to pro-oxidant. The results from the present study are somewhat similar to that observed in a previous study (Seet et al. 2010). In the previous study, serum UA was also measured within 72 h from the onset of stroke. However, the present study has determined the use of those drugs such as diuretics, aspirin, beta-blockers, and CCBs and included the confounders such as GFR, which might probably be associated with UA level and outcome. Moreover, data were investigated in specific populations, such as patients receiving thrombolytic therapy. The same findings have been observed in the thrombolytic subgroup. Thus, UA may play a dual role in patients with AIS h after onset. There is not a simple linear correlation between serum UA levels and stroke prognosis, but an inverted U-shaped curve relationship. It is possible that elevated UA levels may signal harm as an indirect marker of greater pro-oxidative stress rather than being a direct toxic agent. More studies are needed to disentangle the causal mechanisms. This provides evidence for a stratified treatment regimen of exogenous UA. Kivity et al reported UA levels had stronger association with cardiovascular disease in women than in men, indicating the necessity of stratifying by gender in investigations of cardio-cerebrovascular risk factors. Moreover, UA levels are higher in men. In our study, it has been found that the association of UA levels with functional outcome differed in men and women. In women, the highest quartile of UA level was beneficial for good functional outcomes, and patients with the third quartile of UA level had the worst functional outcomes. In men, the second quartile of UA levels ( lmol/l) was an independent predictor of better outcome, and the highest UA level was associated with the worst outcome, which was similar to the results of the whole patients. These may result from the uricosuric effect (i.e., the role of gender steroids in UA regulation) (Adamopoulos et al. 1977) and the effect of the estrogen defense in premenopausal women. The association between UA levels and the prevalence of carotid plaques has been observed in men, however, not in women (Neogi et al. 2009). Moreover, the prevalence of carotid plaques was reportedly higher in postmenopausal women aged C55 years than in younger women (Neogi et al. 2009). On the other hand, the sample of women was about 50 % smaller than that of men, and thus, there was much less statistical observation power for

10 Neurotox Res (2016) 29: testing UA levels association with AIS prognosis in women. Our study has been one of few studies regarding the interaction between gender and UA levels with respect to AIS prognosis. In normal GFR subgroup, the similar findings to that of the whole patients have also been observed. The current study has some limitations. The UA levels were only determined once and were not re-checked during hospitalization or at discharge. The sample size was small; therefore, it is cautioned that our findings should be supported by future work in larger prospective studies. Conclusions Overall, this study provides new insight on the effects of UA on the clinical outcome of AIS and opens avenues of new therapeutic regimens of stratified treatment with exogenous UA. Our data suggest that UA levels should be within a proper range and that it is harmful when too high or too low. However, the definitive proof of causality between UA levels and stroke outcome requires larger, prospective studies and well-designed therapeutic controlled trials in the proper methodologic handling of its gender differences. Acknowledgments We thank Professor Xingshun Xu, Department of Neurology at the Second Affiliated Hospital of Soochow University, for the critical reading of this manuscript. This study was funded by the Suzhou Foundation of Science and Technology Development Plan (SYSD ) and National Natural Science Foundation of China ( , ). Compliance with Ethical Satandards Conflict of interest References The authors declare no conflict of interest. Adamopoulos D, Vlassopoulos C, Seitanides B, Contoyiannis P, Vassilopoulos P (1977) The relationship of sex steroids to uric acid levels in plasma and urine. Acta Endocrinol 85: Amaro S, Chamorro Á (2011) Translational stroke research of the combination of thrombolysis and antioxidant therapy. Stroke 42: Amaro S, Soy D, Obach V, Cervera A, Planas AM, Chamorro A (2007) A pilot study of dual treatment with recombinant tissue plasminogen activator and uric acid in acute ischemic stroke. Stroke 38: Amaro S, Planas AM, Chamorro A (2008) Uric acid administration in patients with acute stroke: a novel approach to neuroprotection. Expert Rev Neurother 8: Becker BF (1993) Towards the physiological function of uric acid. Free Radic Biol Med 14: Bos MJ, Koudstaal PJ, Hofman A, Witteman JC, Breteler MM (2006) Uric acid is a risk factor for myocardial infarction and stroke: the Rotterdam study. Stroke 37: Chamorro A, Obach V, Cervera A, Revilla M, Deulofeu R, Aponte JH (2002) Prognostic significance of uric acid serum concentration in patients with acute ischemic stroke. Stroke 33: Cherubini A, Polidori MC, Bregnocchi M, Pezzuto S, Cecchetti R, Ingegni T, di Iorio A, Senin U, Mecocci P (2000) Antioxidant profile and early outcome in stroke patients. Stroke 31: Dawson J, Walters MR (2006) Uric acid and xanthine oxidase: future therapeutic targets in the prevention of cardiovascular disease? Br J Clin Pharmacol 62: Dawson J, Quinn TQ, Walters MR (2007) Uric acid reduction: a new paradigm in the management of cardiovascular risk. Curr Med Chem 14: Feig DI, Kang DH, Johnson RJ (2008) Uric acid and cardiovascular risk. N Engl J Med 359: Glantzounis GK, Tsimoyiannis EC, Kappas AM, Galaris DA (2005) Uric acid and oxidative stress. Curr Pharm Des 11: Hong JM, Bang OY, Chung CS, Joo IS, Gwag BJ, Ovbiagele B (2010) Influence of recanalization on uric acid patterns in acute ischemic stroke. Cerebrovasc Dis 29: Iribarren C, Folsom AR, Eckfeldt JH, McGovern PG, Nieto FJ (1996) Correlates of uric acid and its association with asymptomatic carotid atherosclerosis: the ARICStudy/ Atherosclerosis Risk in Communities. Ann Epidemiol 6: Johnson RJ, Kang DH, Feig D, Kivlighn S, Kanellis J, Watanabe S, Tuttle KR, Rodriguez-Iturbe B, Herrera-Acosta J, Mazzali M (2003) Is there a pathogenetic role for uric acid in hypertension and cardiovascular and renal disease? Hypertension 41: Kivity S, Kopel E, Maor E, Abu-Bachar F, Segev S, Sidi Y, Olchovsky D (2013) Association of serum uric acid and cardiovascular disease in healthy adults. Am J Cardiol 111: Lehto S, Niskanen L, Rönnemaa T, Laakso M (1998) Serum uric acid is a strong predictor of stroke in patients with non-insulindependent diabetes mellitus. Stroke 29: Logallo N, Naess H, Idicula TT, Brogger J, Waje-Andreassen U, Thomassen L (2011) Serum uri acid: neuroprotection in thrombolysis. The Bergen NORSTROKE study. BMC Neurol 11: Muir SW, Harrow C, Dawson J, Lees KR, Weir CJ, Sattar N, Walters MR (2008) Allopurinol use yields potentially beneficial effects on inflammatory indices in those with recent ischemic stroke: a randomized, double-blind placebo-controlled trial. Stroke 39: Nakanishi N, Okamoto M, Yoshida H, Matsuo Y, Suzuki K, Tatara K (2003) Serum uric acid and risk for development of hypertension and impaired fasting glucose or Type II diabetes in Japanese male office workers. Eur J Epidemiol 18: Neogi T, Ellison RC, Hunt S, Terkeltaub R, Felson DT, Zhang Y (2009) Serum uric acid is associated with carotid plaques: the National Heart Lung, and Blood Institute Family Heart Study. J Rheumatol 36: Romanos E, Planas AM, Amaro S, Chamorro Á (2007) Uric acid reduces brain damage and improves the benefits of rt-pa in a rat model of thromboembolic stroke. J Cereb Blood Flow Metab 27:14 20 Seet RC, Kasiman K, Gruber J, Tang SY, Wong MC, Chang HM, Chan YH, Halliwell B, Chen CP (2010) Is uric acid protective or deleterious in acute ischemic stroke? A prospective cohort study. Atherosclerosis 209: Suarna C, Dean RT, May J, Stocker R (1995) Human atherosclerotic plaque contains both oxidized lipids and relatively large amounts of alpha-tocopherol and ascorbate. Arterioscler Thromb Vasc Biol 15:

11 20 Neurotox Res (2016) 29:10 20 Wang JG, Staessen JA, Fagard RH, Birkenhäger WH, Gong L, Liu L (1998) Prognostic significance of serum creatinine and uric acid in older Chinese patients with isolated systolic hypertension. Stroke 29: Wang Z, Lin Y, Liu Y, Chen Y, Wang B, Li C, Yan S, Wang Y, Zhao W (2015) Serum uric acid levels and outcomes after acute ischemic stroke. Mol Neurobiol. doi: /s Waring WS, Webb DJ, Maxwell SR (2001) Systemic uric acid administration increases serum antioxidant capacity in healthy volunteers. J Cardiovasc Pharmacol 38: Weir CJ, Muir SW, Walters MR, Lees KR (2003) Serum urate as an independent predictor of poor outcome and future vascular events after acute stroke. Stroke 34: Yu ZF, Bruce-Keller AJ, Goodman Y, Mattson MP (1998) Uric acid protects neurons against excitotoxic and metabolic insults in cell culture, and against focal ischemic brain injury in vivo. J Neurosci Res 53: Zhang B, Gao C, Yang N, Zhang W, Song X, Yin J, Pu S, Yi Y, Gao Q (2010) Is elevated SUA associated with a worse outcome in young Chinese patients with acute cerebral ischemic stroke. BMC Neurology 10:82