Effect of using fluoxetine at different time windows on neurological functional prognosis after ischemic stroke

Abstract

Purpose: To evaluate the effect of using fluoxetine at different time intervals after ischemic stroke on neurological functional prognosis in China.

Methods: The patients enrolled were randomly allocated to three groups. Group A received fluoxetine 20 mg/day immediately; group B received fluoxetine 20 mg/day 7 days after enrollment; and group C did not receive fluoxetine. The therapeutic duration of fluoxetine was 90 days and the follow-up period was 180 days.

Results: The mean NIHSS score at day 90 was significantly lower in group A than group C (P = 0.005), while at day 180, the mean score in group A was significantly lower than groups B and C (P = 0.035, P = 0.000), respectively. The mean BI score at day 90 was significantly higher in group A than group C (P = 0.001), while at day 180, the mean score in group A was significantly higher than groups B and C (P = 0.036, P = 0.000), respectively. Regression analysis indicated that lower NIHSS score and higher BI score at day 180 were attributed to the early administration of fluoxetine.

Conclusions: In patients with ischemic stroke, early administration of fluoxetine may improve the neurological functional prognosis.

Keywords

Fluoxetine ischemic stroke neurological functional prognosis time intervals

1 Introduction

According to epidemiological data, ischemic stroke accounts for more than 70% of all strokes, and over the years, there has been a growing prevalence of ischemic stroke (PROGRESS Collaborative Group, 2001). Ischemic stroke is associated with high morbidity (Miniño et al., 2011), severely affecting the quality of life (Zhang et al., 2010). Thus, it is critical to explore the novel therapeutic strategies in order to improve the neurological functional recovery after stroke.

Fluoxetine is a selective serotonin reuptake inhibitor (SSRI) that could increase the expression of brain-derived neurotrophic factor (De Foubert et al., 2004) and facilitate the regeneration of hippocampal neurons (Li et al., 2009). Thus fluoxetine was thought to improve the recovery of motor function after ischemic stroke. A multi-center double-blind randomized controlled clinical study conducted in France demonstrated that the usage of fluoxetine could improve the recovery of motor function after 90 days in patients with ischemic stroke, with or without post-stroke depression (Chollet et al., 2011). Similar studies had prompted that SSRIs could improve partial or integral motor function in patients with post-stroke depression and other comorbidities (Acler et al., 2009; Zittel et al., 2008; Pariente et al., 2001). However, the above studies failed to provide any statistical data and discussion on the different therapeutic effects of SSRIs in improving the neurological functional recovery at different time windows. In view of these limitations, our study used fluoxetine for improving the neurological outcomes at different time windows in patients with ischemic stroke, and extended the period of follow-up compared with the previous studies to make a detailed assessment on different curative effects.

2 Methods

2.1 Participants

2.1.1 Inclusion criteria

Patients who fulfilled the following criteria were included in the study: (1) Adherence to the diagnostic criteria of ischemic stroke as defined by World Health Organization (WHO) (Kunitz et al., 1984). (2) 18– 80 years of age. (3) First onset of stroke within 1 week. (4) Symptoms of defective nervous system and NIHSS score >2. (5) Informed consent from patients or from their legal relatives.

2.1.2 Exclusion criteria

Patients with the following conditions were excluded from the study: (1) Coma. (2) hemorrhagic stroke. (3) Symptoms of defective nervous system before the onset of stroke. (4) Use of antidepressants within 3 months or benzodiazepines within 2 weeks before the onset of stroke. (5) Self-injurious or having suicidal tendency, and urgency of antidepressants. (6) Serious heart diseases, respiratory failure, malignant tumors. (7) Abnormal liver enzymes or creatinine levels. (8) Active peptic ulcer or other gastrointestinal diseases, which may affect the absorption of drugs. (9) Allergic constitution. (10) Severe mental disorders, such as schizophrenia. (11) Pregnant or lactating women. (12) Participated in other clinical studies within 3 months before the onset of the study. (13) Planned endovascular treatment. (14) Patients or their legal relatives who refused to participate in the study.

2.1.3 Withdrawal criteria

Patients were withdrawn from the study in the event of any of the following: (1) Violation of randomization or blinding rules during the follow-up. (2) Serious adverse reactions, such as anaphylactic shock. (3) Urgency of surgery due to the aggravation of ischemic stroke. (4) Serious infections or other medical complications. (5) Usage of antidepressants. (6) Tendency of self-injury or suicide and serious depression, urgency of antidepressants. (7) Patients or their legal relatives requested to quit the study.

2.2 Randomization method

The patients enrolled in the study were divided using the randomized number table method. This procedure was conducted by the randomized code personnel who did not participate in the process of treatment and follow-up. According to the random codes, the patients were divided into three groups: group A (fluoxetine was used immediately when enrolled), group B (fluoxetine was used 7 days after enrollment), and group C (did not use fluoxetine) based on the ratio of 1:1:1 initially. If the patient failed to return for follow-up, its random code was voided. The random codes and information on sectionalization were sealed and saved by the randomized code personnel until the end of the study.

2.3 Sample size calculation

We conducted preliminary sampling, in which 20 patients were assigned to each of the groups. The mean NIHSS scores after 180 days of enrollment were 1.10±1.48, 1.20±2.07, and 2.05±2.44 for groups A, B, and C, respectively. The type I error, α, was set at 0.05, while the type II error, β, was set at 0.15, thus the power was 85% (1– 0.15). Power Analysis andSample Size (PASS 11) software was used for thesample size calculation. The total sample size was 253 cases calculated by PASS 11. Since the dropout rate in the study was about 15% , the final sample size was chosen to be 298 cases. Moreover, the ratio of the sample size for the three groups was 1:1:1, hence each group was expected to enroll 100 cases, and the total sample size was 300 cases.

2.4 Blinding

The evaluator was single-blinded. During the enrollment and follow-up, the investigator evaluated the NIHSS score at days 15, 90, and 180 and BI score at days 90 and 180. The investigator would inquire and record the information on age, gender, present illness, past medical history, blood pressure, blood tests, and adverse events during follow-up. The investigator did not administer the treatment and could not query the information on randomization. The investigator was an experienced clinical physician of neurology, an expert in the assessment of neurological scales as well as in the recognition of the psychiatric symptoms post stroke. The personnel randomizing the patients was banned to participate in the process of treatment and follow-up.The data administrators in charge of the input and analysis of the research data were also banned to participate in and query any other process of our study. However, both the patients and the medical team in charge of treatments were aware of the treatment beingadministered.

2.5 Therapeutic interventions

The enrolled patients were selected by the same medical team of neurology in the hospital where the study was performed. The conventional therapies for stroke and related risk factors were referred to the Guidelines for the Prevention of Stroke in Patients with Stroke or Transient Ischemic Attack revised by the American Heart Association and America Stroke Association (Furie et al., 2011). The inflammatory response after ischemic stroke may influence the long-term functional prognosis (Smith et al., 2004). Hence we hypothesized that earlier use of fluoxetine immediately and 7 days after stroke are critical in improving the neurological functional recovery. In our study, we used day as the unit of time for starting treatment of fluoxetine. The dosage of fluoxetine was 20 mg per day, the therapeutic duration was 90 days and the follow-up period was 180 days. Fluoxetine dispersible tablets used in our study were manufactured by Eli lilly (Indianapolis, IN). The drug production code, approved by the state Food and Drug Administration, was J20120001. All patients received the same health education and the same rehabilitation therapies during the follow-up period. After enrollment, all the patients received baseline examinations, including cranial computed X-ray tomography (CT) or magnetic resonance imaging (MRI), carotid artery color Doppler ultrasound, electrocardiogram (ECG), monitoring of blood pressure (BP), serum white blood cell (WBC) count, serum high-sensitivity C-reactive protein (hs-CRP), serum lipid, serum glutamic-oxaloacetic transaminase (GOT), serum creatinine (Cr), and glucose and glycosylated hemoglobin (HbA1c). The examinationof serum lipid included total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), total triglyceride (TG), and high-density lipoprotein cholesterol (HDL-C). All the patients who completed the follow-up were reviewed for the serum WBC count and hs-CRP at day 30 and serum glutamic-oxaloacetic transaminase, creatinine, TG, TC, LDL-C, HDL-C, glucose and HbA1c, and BP at day 180.

2.6 Compliance assessment

The rate of hospital revisit was used to evaluate the therapeutic compliance of the patients. The patients were required to return to the hospital for revisit every 15 days and every revisit would be recorded by the same evaluator for follow-up. During the follow-up, each patient should have revisited the hospital 11 times.

Revisit rate = total number of revisits / (number of enrolled cases×11)

2.7 The relevant assessment indexes

(1) The National Institutes of Health Stroke Scale (NIHSS) score was used to evaluate the neurological deficits in patients with stroke, which was reflective of all the aspects of neural function reliability, validity,and sensitivity (Tilley et al., 1996). (2) The Barthel index (BI) scores, which can be evaluated telephonically, was widely used in large clinical trials for quantitative assessment of daily living disability. The higher the BI score, the more independent the patient is (Uyttenboogaart et al., 2005; Hsueh et al., 2002). (3) Secondary inflammatory injury was presented after acute ischemic stroke, and serum hs-CRP levels were reflective of the severity of these symptoms (Chotik-anuchit et al., 2011; Smith et al., 2004). (4) WBC count is a major player in the inflammatory response and also in the early inflammatory reaction after ischemic stroke (Kochanek et al., 1987).

2.8 Endpoint events

2.8.1 Primary endpoint events

The NIHSS score at days 15, 90, and 180 afterenrollment.

2.8.2 Secondary endpoint events

The BI score at days 90 and 180 after enrollment.

2.9 Statistical analysis

Statistical analysis of the data was conducted using Statistical Package for the Social Sciences software (SPSS 17.0). The mean±standard deviation was used to represent measurement data, including normal distribution, and was compared between the groups or within the groups by t-test. One-way analysis of variance (ANOVA) was conducted to compare the means from multiple samples. The Student-Newman-Keuls test would be conducted between groups if statistically significant values were observed in ANOVA. The measurement of data was evaluated using skewness distribution, median was used for representation and data were compared between the groups or within the groups by Wilcoxon rank sum test or Kruskal-Wallis H test. Data were compared between the groups by chi-square test or Fisher’s exact test. Chi-square tests between any two sample rates would be conducted if statistically significant values were observed when comparing multiple sample rates. Finally, we assessed the influence of the time at which fluoxetine was used after ischemic stroke for the NIHSS and BI scores at day 180 by multiple linear regression analysis. Statistical significance was defined by P < 0.05.

2.10 Ethics standard and registration

This study was approved by the ethics committee of Shenzhen People’s Hospital. The members of the medical team in this study informed the patients about the purpose, significance, possible treatment, period of follow-up, and possible adverse events associated with the study. The patients were enrolled after signing the informed consent form, and they were given the authority to quit after the beginning of the study. The study was registered in the Chinese clinical trial registry center, a registered institution in the the World Health Organization international clinical trial registry platform. The registration web address is http://www.chictr.org, and the registration number is ChiCTR-IPR-15007658.

3 Results

Three hundred patients, who fulfilled the criteria,were enrolled and allocated into three groups from January 2012 to June 2014. 7 patients were lost to follow-up in group A, 14 in group B, and 9 in group C. The dropout rates between the groups were not significantly different (7.0% vs 14.0% vs 9.0% , P = 0.236). The revisit rates in the three groups were 97.2% , 97.7% , and 96.8% , respectively, and the compliances between the groups were not significantly different (P = 0.502). During the follow-up period, intracranial hemorrhage was observed in one patient in group A, 2 patients in both group B and group C, respectively; however, occupied hematoma was not observed in any of the patients. The incidence of intracranial hemorrhage in the three groups were not significantly different (1.1% vs 2.3% vs 2.2% , P = 0.772). During the period of follow-up, 1 patient died among the three groups. The mortality rates in the three groups were not significantly different (1.1% vs 1.2% vs 1.1% , P = 0.998). The dead patients were excluded while conducting the statistics of endpoint events. Thus, a total of 267 patients were enrolled for the evaluation of endpoints. Among these, 92 patients were from group A, 85 patients from group B, and 90 patients from group C (Fig. 1).

There were no significant differences in the baseline data between the three groups, including mean age, proportion of men, and anterior or posterior circulation infarct, proportion of smoking and alcoholism, proportion of lung or urinary tract infection, proportion of hypertension, diabetes, erratic carotid plaque and atrial fibrillation, mean of fasting plasma glucose, HbA1c and serum lipid, and time for starting conventional treatment. Additionally, no significant differences were observed in the incidence of adverse events during follow-up between the three groups (Table 1).

There were no significant differences in the mean NIHSS scores at baseline and day 15 and the mean BI scores at baseline and between the three groups. There was no significant difference in the mean NIHSS score at day 90 between the groups; however, the mean score in group A was significantly lower than group C (P = 0.005). The mean NIHSS score at day 180 was significantly lower in group A than group B or group C (P = 0.035, P = 0.000), and the mean score in group B was significantly lower than group C (P = 0.049). The mean NIHSS scores in each group at day 15 were significantly lower than baseline (P < 0.05), while the scores at day 90 were significantly lower than day 15 (P < 0.05) and the scores at day 180 was significantly lower than day 90 (P < 0.05) (Table 2, Fig. 2). No significant difference was observed in the mean BI scores at day 90 between group A and group B, and between group B and group C. However, the scores in group A was significantly higher than group C (P = 0.001). The mean BI score at day 180 was significantly higher in group A than group B or group C (P = 0.036, P = 0.000), and the score in group B was significantly higher than group C (P = 0.043). The mean BI scores in each group at day 90 were significantly higher than baseline (P < 0.05), and the scores at day 180 were significantly higher than day 90 (P < 0.05) (Table 3, Fig. 2).

There were no significant differences between the mean WBC counts at baseline and at day 30 between the three groups. Moreover, no significant difference was observed in the mean serum hs-CRP at baseline between the three groups; however, the value at day 30 was significantly lower in group A than group B or group C (P = 0.045, P = 0.000), the value at day 30 was significantly lower in group B than group C (P = 0.044). The mean WBC counts and serum hs-CRP at day 30 were significantly lower than baseline (P < 0.05) (Table 4).

To evaluate whether the time for starting fluoxetine treatment is an independent factor for long-term neurological functional prognosis, fluoxetine was used as the statistical object in group A and B cases. The NIHSS and BI scores at day 180 were used as dependent variables. The value of systolic blood pressure, diastolic blood pressure, fasting plasma glucose, TC, TG, HDL-C, LDL-C at day 180, and age, male gender, time for starting fluoxetine treatment were used as independent variables. A multiple linear regression analysis was used in which the independent variables were included in the regression equation, when the p-value was <0.05. The results showed that the time for starting the fluoxetine treatment was the independent factor for the NIHSS and BI scores at day 180 (Table 5). Thus after the ischemic stroke, early start of the fluoxetine treatment may lower the NIHSS score, and the BI score was higher at day 180.

4 Discussion

4.1 Possible relative mechanism

Some studies have suggested that fluoxetine, an SSRI, could improve the recovery of neurological function by increasing the expression of brain-derived neurotrophic factor (Liu et al., 2014; Karpova et al., 2011; De Foubert et al., 2004), facilitating the regeneration of hippocampal neurons (Li et al., 2009; Greene et al., 2009; Lee et al., 2009; Kubera et al., 2009), anti-inflammatory effect (Lim et al., 2009), and antioxidant effect (Kim et al., 2007).

Chollet F, in a double-blind randomized-controlled study, demonstrated that the treatment of fluoxetine could improve patient’s motor function in 90 days after ischemic stroke (Chollet et al., 2011). In our study, under the principle of expanding the sample size and extending the follow-up period, we showed that long-term neurological prognosis was better in patients who received fluoxetine treatment compared with those who did not, and the findings supported the conclusions. Fluoxetine does not exhibit thrombolytic activity at acute phase and its effect on improving the neurological recovery is based on long-term application. However, currently, there is a lack of clinical studies to validate the influence of fluoxetine usage at different time windows after ischemic stroke on the neurological functional prognosis.

Under the normal physiological state, although hippocampal dentate gyrus can continuously produce new neurocytes, they will be cleared within 4 weeks, and consequently apoptosis occurs in large number of new neurocytes before maturity, thereby maintaining the stable function of hippocampal neurological network (Snyder et al., 2009). When compared to the normal physiological state, under the pathological state, such as ischemic stroke, fluoxetine can promote the survival of newborn neurocytes through its antiapoptotic effect and facilitate the recovery of neurological function. Thus, it can be implied that the therapeutic time window of fluoxetine for neurological functional recovery after ischemic stroke may be relatively broad. However, SSRIs can enhance the level of serotonin in primates and rodents rapidly (Frazer & Benmansour, 2002), thus whether the early application of fluoxetine can further promote the recovery of neurological function in ischemic stroke is still unresolved.

The study conducted by Chollet F confirmed that the application of fluoxetine within 10 days was conducive to the recovery of motor function at 90 days after ischemic stroke. In light of this, our study used fluoxetine within 1 week after ischemic stroke for some patients; however, the results of the two studiescould not be horizontally compared because of the differences in baseline characteristics. In our study, the neurological deficits and daily living disability at day 180 were significantly reduced in the patients receivingfluoxetine immediately within 1 week after the onset of ischemic stroke compared with patients who received fluoxetine after 7 days. The risk factors of ischemic stroke, such as hypertension, diabetes, and lipid metabolism disorder may affect the neurological functional prognosis (Vemmos et al., 2004; Ahmed et al., 2000; Karsito & Soeatmadji, 2008; Vauthey et al., 2000; Weir et al., 2003). We demonstrated that the time of starting the fluoxetine treatment is the independent risk factor for NIHSS and BI scores at day 180 after adjusting age, sex, and the previously mentionedfactors. These results suggested that using fluoxetine as early as possible after ischemic stroke could further improve the long-term neurological functionalprognosis.

We considered that the early anti-inflammatory effect of fluoxetine on central nervous system (CNS) was critical for improving the neurological functional prognosis. The CNS generates inflammatory response within several hours to several days after ischemic stroke, and strong inflammatory response can accelerate the spread of ischemia penumbra, which can lead to further ischemic neurological damage (Dirnagl et al., 1999). Lim CM confirmed that intravenous application of fluoxetine on a rat model of middle cerebral artery occlusion at 2– 9 h after cerebral ischemia could inhibit microglia to produce pro-inflammatory cytokines, and sequentially reduce the damage of neurocytes generated by inflammatory response after cerebral ischemia. Moreover, fluoxetine inhibits microglia activation and neutrophil infiltration in the cerebral ischemic area, and reduces the risk of cerebral infarction and improves motor function in rats (Lim et al., 2009). The study of Lim CM prompted that using fluoxetine during early stages of cerebral ischemia exerts an anti-inflammatory effect on CNS. Dhami also showed that application of SSRIs (citalopram or fluoxetine) reduces the inflammatory response of cortical neurons by inhibiting the production of pro-inflammatory cytokines (Dhami et al., 2013).

In our study, the median of serum hs-CRP at day 30 after enrollment was significantly lower in patients who received fluoxetine immediately within 1 week after the onset of ischemic stroke than those who received fluoxetine after 7 days, although the mean of WBC counts was not significantly different between the two. Comparing the effect in serum hs-CRP and WBC counts, which are commonly associated with inflammatory effects and commonly used in clinical trials, it was confirmed that early administration of fluoxetine after ischemic stroke showed better anti-inflammatory effect. Thus, we opined that early usage of fluoxetine might save more neurons by strengthening anti-inflammatory effect at the early stages of cerebral ischemia and further improve the neurological functional prognosis.

Reperfusion at early stages of ischemic stroke may increase neuronal damage, with oxidative stress as one of the main pathological mechanisms (Gandhi & Abramov, 2012). A study showed that using SSRIs at 2 h after ischemic reperfusion could render catalase and superoxide dismutase increasing significantly in animal model by exerting antioxidant stress (Gaur &Kumar, 2010). Therefore, early use of fluoxetine improves the neurological functional prognosis by exerting antioxidant stress effect. In addition, many studies have shown that generation and migration of neurocytes, generated in the hindbrain after stroke, was reduced (Jin et al., 2006; Martí-Fábregas et al., 2010). Thus, early use of fluoxetine might exert neurotrophic effect and improves neurological functional prognosis. However, there is a lack of conventional clinical-related examinations to further demonstrate the differences in antioxidant stress and neurotrophic effects when fluoxetine is used at different timewindows.

4.2 Safety evaluation

In our study, the possible related adverse events during the follow-up period were not significantly different between the three groups; thereby highlighting that fluoxetine was well tolerated in all the patients. The incidences of intracranial hemorrhage in the three groups were relatively low and without any significant difference.

4.3 Limitation

This study demonstrated that the early use of fluoxetine after ischemic stroke might exert anti-inflammatory, antioxidant stress, and neurotrophic effects, which can further improve the neurological functional outcomes. However, there are limited studies to demonstrate these mechanisms. Thus, a combination of clinical and basic research is required to address the issues. This study was originally designed as a multi-center clinical study; however, owing to the highly innovative nature of the study and the medical environment in China, the study objective was not met appropriately. As a result, there was a certain extent of bias for the results in this study. Finally, whether extending the therapeutic duration of fluoxetine to 180 days would improve the neurological functional prognosis or increase the drug adverse effect is still unclear, and further well-designed studies are warranted to gain more clarity.

4.4 Conclusion

Our study suggested that early use of fluoxetine after ischemic stroke could reduce the neurological deficits and daily living disability at day 180. This conclusion will provide new referenced evidence for the use of fluoxetine in ischemic stroke as a drug for improving neurological functional recovery.

4.5 Outlook

Although a preliminary demonstration was made on the difference in neurological functional prognosis using fluoxetine at different time windows in patients with ischemic stroke, there was no relative report to demonstrate the differences of neurological functional prognosis when fluoxetine was used at different doses or duration of time currently. Further research is required in the experimental design to address these shortcomings.

Footnotes

Acknowledgments

Our study was funded by Guangdong province health department of the People’s Republic of China (Item number: 2011B031800130) and the innovation of science and technology committee of Shenzhen City, Guangdong province of the People’s Republic of China (Item number: 201101020).

References

Acler

, Robol

, Fiaschi

, Manganotti

(2009). A double blind placebo RCT to investigate the effects of serotonergic modulation on brain excitability and motor recovery in stroke patients. Journal of Neurology, 256, 1152–1158.

Ahmed

, Näsman

, Wahlgren

N.G.

(2000). Effect of intravenous nimodipine on blood pressure and outcome after acute stroke. Stroke, 31, 1250–1255.

Chollet

, Tardy

, Albucher

J.F.

, Thalamas

, Berard

, Lamy

, Bejot

, Deltour

, Jaillard

, Niclot

, Guillon

, Moulin

, Marque

, Pariente

, Arnaud

, Loubinoux

(2011). Fluoxetine for motor recovery after acute ischaemic stroke (FLAME): A randomised placebo-controlled trial. The Lancet Neurology, 10, 123–130.

Chotik-anuchit

, Nilanont

, Poungvarin

(2011). Relationship between ankle brachial index (ABI), high sensitivity C-reactive protein (hs-CRP) and initial disability level in acute ischemic stroke. Journal of the Medical Association ofThailand, 94, S94–S98.

De Foubert

, Carney

S.L.

, Robinson

C.S.

, Destexhe

E.J.

, Tomlinson

, Hicks

C.A.

, Murray

T.K.

, Gaillard

J.P.

, Deville

, Xhenseval

, Thomas

C.E.

, O’Neill

M.J.

, Zetterström

T.S.

(2004). Fluoxetine-induced change in rat brain expression of brain-derived neurotrophic factor varies depending on length of treatment. Neuroscience, 128, 597–604.

Dhami

K.S.

, Churchward

M.A.

, Baker

G.B.

, Todd

K.G.

(2013). Fluoxetine and citalopram decrease microglial release of glutamate and D-serine to promote cortical neuronal viability following ischemic insult. Molecular and Cellular Neurosciences, 56, 365–374.

Dirnagl

, Iadecola

, Moskowitz

M.A.

(1999). Pathobiology of ischaemic stroke: An integrated view. Trends in Neurosciences, 22, 391–397.

Frazer

, Benmansour

(2002). Delayed pharmacological effects of antidepressants. Molecular Psychiatry, 7, S23–S28.

Furie

K.L.

, Kasner

S.E.

, Adams

R.J.

, Albers

G.W.

, Bush

R.L.

, Fagan

S.C.

, Halperin

J.L.

, Johnston

S.C.

, Katzan

, Kernan

W.N.

, Mitchell

P.H.

, Ovbiagele

, Palesch

Y.Y.

, Sacco

R.L.

, Schwamm

L.H.

, Wassertheil-Smoller

, Turan

T.N.

, Wentworth

, American Heart Association, Council on Cardiovascular Nursing, Council on Clinical Cardiology, Interdisciplinary Council on Quality of Care, Outcomes, Research, (2011). Guidelines for the Prevention of Stroke in Patients With Stroke or Transient Ischemic Attack: A Guideline for Healthcare Professionals From the American Heart Association/American Stroke Association. Stroke, 42, 227–276.

10.

Gandhi

, Abramov

A.Y.

(2012). Mechanism of oxidative stress in neurodegeneration. Oxidative Medicine and Cellular Longevity, 2012, 428010.

11.

Gaur

, Kumar

(2010). Behavioral, biochemical and cellular correlates in the protective effect of sertraline against transient global ischemia induced behoral despair: Possible involvement of nitric oxide-cyclic guanosine monophosphate study pathway. Brain Research Bulletin, 82, 57–64.

12.

Greene

, Banasr

, Lee

, Warner-Schmidt

, Duman

R.S.

(2009). Vascular endothelial growth factor signaling is required for the behavioral actions of antidepressant treatment: Pharmacological and cellularcharacterization. Neuropsychopharmacology, 34, 2459–2468.

13.

Hsueh

I.P.

, Lin

J.H.

, Jeng

J.S.

, Hsieh

C.L.

(2002). Comparison of the psychometric characteristics of the functional independence measure, 5 item Barthel index, and 10 item Barthel index in patients with stroke. Journal of Neurology, Neurosurgery, Psychiatry, 73, 188–190.

14.

Jin

, Wang

, Xie

, Mao

X.O.

, Zhu

, Wang

, Shen

, Mao

, Banwait

, Greenberg

D.A.

(2006). Evidence for stroke-induced neurogenesis in the human brain. Proceedings of the National Academy of Sciences of the United States of America, 103, 13198–13202.

15.

Karpova

N.N.

, Pickenhagen

, Lindholm

, Tiraboschi

, Kulesskaya

, Agústsdóttir

, Antila

, Popova

, Akamine

, Bahi

, Sullivan

, Hen

, Drew

L.J.

, Castrén

(2011). Fear erasure in mice requires synergy between antidepressant drugs and extinction training. Science, 334, 1731–1734.

16.

Karsito Soeatmadji

D.W.

(2008). Diabetes and stroke. Acta Medica Indonesiana, 40, 151–158.

17.

Kim do

, Li

, Yoo

K.Y.

, Lee

B.H.

, Hwang

I.K.

, Won

M.H.

(2007). Effects of fluoxetine on ischemic cells and expressions in BDNF and some antioxidants in the gerbil hippocampal CA1 region induced by transient ischemia. Experimental Neurology, 204, 748–758.

18.

Kochanek

P.M.

, Dutka

A.J.

, Hallenbeck

J.M.

(1987). Indomethacin, prostacyclin, and heparin improve postischemic cerebral blood flow without affecting early postischemic granulocyte accumulation. Stroke, 18, 634–637.

19.

Kubera

, Grygier

, Arteta

, Urbańska

, Basta-Kaim

, Budziszewska

, Leśkiewicz

, Kołaczkowska

, Maes

, Szczepanik

, Majewska

, Lasoń

(2009). Age-dependent stimulatory effect of desipramine and fluoxetine pretreatment on metastasis formation by B16F10 melanoma in male C57BL/6 mice. Pharmacological Reports, 61, 1113–1126.

20.

Kunitz

S.C.

, Gross

C.R.

, Heyman

, Kase

C.S.

, Mohr

J.P.

, Price

T.R.

, Wolf

P.A.

(1984). The pilot Stroke Data Bank: Definition, design, and data. Stroke, 15, 740–746.

21.

Lee

J.S.

, Jang

D.J.

, Lee

, Ko

H.G.

, Kim

Y.S.

, Kim

, Son

, Kim

S.H.

, Chung

, Lee

M.Y.

, Kim

W.R.

, Sun

, Zhuo

, Abel

, Kaang

B.K.

, Son

(2009). Induction of neuronal vascular endothelial growth factor expression by cAMP in the dentate gyrus of the hippocampus is required for antidepressant-like behaviors. The Journal of Neuroscience, 29, 8493–8505.

22.

Lim

C.M.

, Kim

S.W.

, Park

J.Y.

, Kim

, Yoon

S.H.

, Lee

J.K.

(2009). Fluoxetine affords robust neuroprotection in the postischemic brain via its anti-inflammatory effect. Journal of Neuroscience Research, 87, 1037–1045.

23.

Liu

, Zhang

, Sun

, Fan

, Zhou

, Fu

(2014). Effects of fluoxetine on brain-derived neurotrophic factor serum concentration and cognition in patients with vascular dementia. Clinical Interventions in Aging, 9, 411–418.

24.

W.L.

, Cai

H.H.

, Wang

, Chen

, Zhou

Q.G.

, Luo

C.X.

, Liu

, Ding

X.S.

, Zhu

D.Y.

(2009). Chronic fluoxetine treatment improvesi schemia-induced spatial cognitive deficits through increasing hippocampal neurogenesis after stroke. Journal of Neuroscience Research, 87, 112–122.

25.

Martí-Fábregas

, Romaguera-Ros

, Gómez-Pinedo

, Martínez-Ramírez

, Jiménez-Xarrié

, Marín

, Martí-Vilalta

J.L.

, García-Verdugo

J.M.

(2010). Proliferation in the human ipsilateral subventricular zone after ischemic stroke. Neurology, 74, 357–365.

26.

Miniño

A.M.

, Murphy

S.L.

, Xu

, Kochanek

K.D.

(2011). Deaths: Final data for 2008. National Vital Statistics Reports, 59, 1–126.

27.

Pariente

, Loubinoux

, Carel

, Albucher

J.F.

, Leger

, Manelfe

, Rascol

, Chollet

(2001). Fluoxetine modulates motor performance and cerebral activation of patients recovering from stroke. Annals of Neurology, 50, 718–729.

28.

PROGRESS Collaborative Group. (2001). Randomised trial of a perindopril-based blood-pressure-lowering regimen among 6,105 individuals with previous stroke or transient ischaemic attack. Lancet, 358, 1033–1041.

29.

Snyder

J.S.

, Choe

J.S.

, Clifford

M.A.

, Jeurling

S.I.

, Hurley

, Brown

, Kamhi

J.F.

, Cameron

H.A.

(2009). Adult-born hippocampal neurons are more numerous, faster maturing, and more involved in behavior in rats than in mice. The Journal of Neuroscience, 29, 14484–14495.

30.

Smith

C.J.

, Emsley

H.C.

, Gavin

C.M.

, Georgiou

R.F.

, Vail

, Barberan

E.M.

, del Zoppo

G.J.

, Hallenbeck

J.M.

, Rothwell

N.J.

, Hopkins

S.J.

, Tyrrell

P.J.

(2004). Peak plasma interleukin-6 and other peripheral markers of inflammation in the first week of ischaemic stroke correlate with brain infarct volume, stroke severity and long-term outcome. BMC Neurology, 4, 2.

31.

Tilley

B.C.

, Marler

, Geller

N.L.

, Lu

, Legler

, Brott

, Lyden

, Grotta

(1996). Use of a global test for multiple outcomes in stroke trials with application to the National Institute of Neurological Disorders and Stroke t-PA Stroke Trial. Stroke, 27, 2136–2142.

32.

Uyttenboogaart

, Stewart

R.E.

, Vroomen

P.C.

, De Keyser

, Luijckx

G.J.

(2005). Optimizing cutoff scores for the Barthel index and the modified Rankin scale for defining outcome in acute stroke trials. Stroke, 36, 1984–1987.

33.

Vauthey

, de Freitas

G.R.

, van Melle

, Devuyst

, Bogousslavsky

(2000). Better outcome after stroke with higher serum cholesterol levels. Neurology, 54, 1941–1942.

34.

Vemmos

K.N.

, Tsivgoulis

, Spengos

, Zakopoulos

, Synetos

, Manios

, Konstantopoulou

, Mavrikakis

(2004). U-shaped relationship between mortality and admission blood pressure in patients with acute stroke. Journal of Internal Medicine, 255, 257–265.

35.

Weir

C.J.

, Sattarb

, Waltersa

M.R.

, Lees

K.R.

(2003). Low triglyceride, not low cholesterol concentration, independently predicts poor outcome following acute stroke. Cerebrovascular Diseases, 16, 76–82.

36.

Zhang

, Wang

, Liu

, Zhao

, Xue

, Zhou

, Wang

(2010). A prospective cohort study of the incidence and determinants of post-stroke depression among the mainland Chinese patients. Neurological Research, 32, 347–352.

37.

Zittel

, Weiller

, Liepert

(2008). Citalopram improves dexterity in chronic stroke patients. Neurorehabilitation and Neural Repair, 22, 311–314.