A cross-sectional study to assess the association between major depression and inflammatory markers in patients with acute ischemic stroke.

BACKGROUND: Increased interest in the relationship between affective disorder and long-term health consequences has generated recent examinations of depression and stroke. Observations suggest that depressive disorder is associated with abnormal physiological and immunological responses and a resultant increase in inflammatory markers. Given the high prevalence of stroke and associated costs for the community, it is important to understand the mechanisms that may impact on the outcome to achieve the best possible prognosis. AIMS: The view that inflammatory factors contribute to depression is predicated on findings that circulating cytokines and other inflammatory factors are increased in depressed patients. Therefore, it has been hypothesized that inflammation could be one of the mechanisms by which depression increases risk for ischemic stroke. Our aim was to determine whether there is any relationship between major depression and tumor necrosis factor-alpha (TNF-α), interleukin-1 beta (IL-1β), IL-18, brain-derived neurotrophic factor (BDNF), and neuron-specific enolase (NSE) in patients with acute ischemic stroke (AIS). STUDY
DESIGN: This was as a cross-sectional design.
MATERIALS AND METHODS: This study has a cross-sectional design, and it was conducted in Necmettin Erbakan University, the Meram Faculty of Medicine in Konya, Turkey, between 2014 and 2015. Fifty-three AIS patients admitted to the hospital within the first 24 h after stroke onset were recruited. Major depression was ascertained by means of the structured clinical interview for the diagnostic and statistical manual of mental disorders, Fourth Edition/Clinical Version. The enzyme-linked immunosorbent assay was used to measure the serum levels of TNF-α, IL-1 β, IL-18, BDNF, and NSE at admission.
RESULTS: A total of 53 patients with a mean age of 65.9 years were recruited. Of these patients, 17 (32.1%) had major depression. Depressive and nondepressive patients had similar demographical and clinical features. There was no significant statistical difference between depressive and nondepressive patients with AIS with respect to levels of TNF-α, IL-1 β, IL-18, BDNF, and NSE.
CONCLUSION: This study suggests that in patients who have experienced AIS, there is no significant relationship between major depression and basal proinflammatory cytokines (TNF-α, IL-1 β, IL-18), BDNF, and NSE.

INTRODUCTION

Stroke is a leading cause of death and permanent disability with significant economic losses due to functional impairments. Increased interest in the relationship between affective disorder and longterm health consequences has generated recent examinations of depression and stroke. Depression has been associated with increased rates of hypertension,[1] diabetes mellitus,[2] and cardiovascular disease and mortality,[3] and this raises the possibility that a similar relation may exist for cerebrovascular disease and mortality. Several epidemiological studies indicate a positive association between depressive symptoms and stroke.[4567] Salaycik et al. provide data that supports the association between mood disorders and risk of ischemic stroke.[7] In this study, depressive symptoms were an independent risk factor for an incidence of stroke. In addition, the risk of developing stroke was 4.21 times greater in those individuals with symptoms of depression. The Baltimore Study showed that individuals with a history of depressive disorder were 2.6 times more likely to report stroke.[5] Significant psychological distress was also a predictor of fatal ischemic stroke.[6] Previous studies[89] suggested that, in patient groups who have depressive symptoms, stroke morbidity increases by 2.3–2.7-fold, and baseline depression was found to be an independent predictor for total incidents of stroke and fatal stroke. Many subsequent studies showed similar results with respect to depression, morbidity, and mortality related to total incidence of stroke, fatal stroke, and ischemic stroke.[1011121314] Although the relationship between depression and cerebrovascular disease events has been consistent, the current understanding of the underlying mechanisms, especially the relationship between morbidity, mortality, and the existence of depression, remains speculative. It has been hypothesized that inflammation could be one of the mechanisms by which depression increases risk for ischemic stroke. Inflammatory factors have been linked to ischemic stroke.[1516] The view that inflammatory factors contribute to depression was initially predicated on findings that circulating cytokines and other inflammatory factors were found to be increased in depressed patients. A number of investigative clinical studies have found individuals with depression to have higher circulating levels of cytokines, for example, interleukin-1 beta (IL-1β) and tumor necrosis factor-alpha (TNF-α)[171819] and IL-18.[20] The administration of IL-1β administration after experimentally induced Ischemia increases infarct size and edema formation.[21] TNF-α is a depression-related proinflammatory cytokine that increases in response to ischemic conditions.[22] IL-18 can mediate delayed neuro-inflammatory processes in experimental hypoxic–ischemic brain injury.[23] To sum up, proinflammatory cytokines increases in acute ischemic stroke (AIS); it is also increased in psychiatric patients with depression and without stroke compared to controls. On the other hand, elevated blood levels of inflammatory markers are associated with an unfavorable functional outcome, a more severe neurological deficit, and an increased mortality after stroke.[242526] These studies have also shown that, in psychiatric patients with major depression[2728] and suicide victims, brain-derived neurotrophic factor (BDNF) levels are lower.[29] Pikula et al. found that a lower level of serum BDNF was associated with an increased risk of incident stroke,[30] and there are well-documented positive effects of BDNF on neuroplasticity after stroke.[31] neuron-specific enolase (NSE) is a glycolytic pathway enzyme, predominantly found in neuronal cells in the isoform γγ. The γ-subunit of the enolase originates mainly from the cytoplasm of neurons and neuroendocrine cells and plays a direct role in energy metabolism, neural differentiation, neuroplastic pathways, and cell survival.[32] NSE has been studied less frequently in both depressive psychiatric patients and neurological patients with AIS. One study found individuals with depression to have high levels of cerebrospinal fluid (CSF) and NSE.[33] Wunderlich et al. suggested that the relationship between the severity of neurological deficit in patients with ischemic stroke and infarction size with NSE.[34]

Within the scope of data mentioned above, in individual diagnosed with major depression before AIS, proinflammatory cytokines and NSE levels can be higher, and BDNF levels can be lower when compared to individuals without depression. These changes in inflammatory response could be related to the presence of depression before AIS and stroke-related neurological deficits can make sense in regard to morbidity and mortality. If the presence of major depression in patients just before AIS affects poststroke proinflammatory cytokines, BDNF, and NSE; due to the negative effects of these biological parameters on prognosis, the diagnosis and treatment of major depression, especially in patients with high AIS risk, must be a primary goal. However, whether the presence of major depression in patients just before AIS has an effect on poststroke proinflammatory cytokines (TNF-α, IL-1 β, IL-18), BDNF, and NSE has not yet been tested. In this study, we aim to examine these associations between these biological parameters and major depression in patients with AIS.

MATERIALS AND METHODS

Study population

This study has a cross-sectional design, and it was conducted at Necmettin Erbakan University in the Meram Faculty of Medicine in Konya, Turkey, between 2014 and 2015. Two hundred and thirty-six patients with a first episode of AIS were admitted to our hospital within the first 24 h of stroke onset during the study period, and we excluded patients according to the exclusion criteria at baseline. Ultimately, 53 patients were included in the study. Only patients without significant clinical factors that promote inflammation were selected. The inclusion criteria were the clinical presentation of the first-ever stroke in those age 18 years or older. Those who were admitted after 24 h of stroke whose cause of stroke was intracerebral and subarachnoid hemorrhage; who were unconscious; who had aphasia or dysphasia; who had a brain tumor or systemic malignancy; who recently or simultaneously had acute myocardial infarction and unstable angina pectoris; who had renal dysfunction; who had symptomatic peripheral artery disease; who had a history of infection within the last month, who had acquired an infection after admission and a simultaneous diagnosis of sepsis; who had autoimmune and immunosuppressive disease or used immunosuppressant drugs (corticosteroid etc.); who had trauma or a history of surgical intervention within the last month, who had acute/chronic inflammatory gastrointestinal disease; who had rheumatic disease and had been diagnosed with metabolic syndrome; who had been using antidepressant or other psychotropic drugs during the last month; who had anxiety disorder, schizophrenia or other psychotic disorders, bipolar disorder, or alcohol-substance abuse apart from major depression disorder were not recruited to the trial. Informed consent was obtained after providing verbal and written information to participants or nearest relatives when relevant.

Stroke case ascertainment

Stroke was defined as the presence of rapidly developing focal neurological signs or symptoms of vascular origin that persisted for 24 h.[7] All events were adjudicated by a panel of at least two neurologists. A general medical history was collected; and physical and neurological examinations, standard laboratory tests, and a 12-lead electrocardiogram were performed on all patients upon admission. The ischemic lesion and affected brain region were assessed and confirmed by both the clinical examination and computed tomography and/or brain magnetic resonance imaging during hospitalization for acute stroke. The location of the stroke was determined by radiologically findings. The stroke subtype was classified according to trial of ORG 10172 in acute stroke treatment (TOAST) criteria.[35] Stroke severity was evaluated at admission by trained neurologists using the National Institutes of Health Stroke Scale (NIHSS).[36]

Psychological measurement

After neurological assessments were completed, and the sociodemographic features were recorded, the patients were assessed by a psychiatrist in the same hospital. Psychiatric interviews were carried out by psychiatrists with at least 4 years of experience on psychiatric disorders and diagnostic instruments. The psychiatrists were also blinded to the neuro-psychiatric evaluations of the patients. Psychiatric disorders were screened using the Structured Clinical Interview for the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition/Clinical Version.[37] Preexisting major depression in the patients was diagnosed by the same clinical structured interview.

Blood collection and cytokine measurement

Blood samples were taken during the first 24 h after stroke onset. A total of 8 ml of peripheral venous blood was drawn and centrifuged at 1000 g for 10 min at room temperature to separate the blood components. Serum samples were kept at −80°C until assay. TNF-α, IL-1 β, IL-18, BDNF, NSE, and serum levels were measured with an enzyme-linked immunosorbent assay, using a commercial kit according to the manufacturer instructions (BDNF, IL-1 β and TNF-α: Boster Biological Technology, California, USA; NSE: Yehua Biological Technology, Shanghai, China; IL-18: eBioscience Bender MedSystems, Vienna, Austria).

Statistical analysis

Statistical analyses were performed with SPSS 16.0 for Windows (Statistical Package for the Social Sciences, Chicago, IL, USA). For comparisons between the study groups, a t-test (for normally distributed variables) and the Mann–Whitney U test (for abnormally distributed variables) were used for continuous variables and a Chi-square test (for 3 or more × 2 variables) or Fisher's exact test (for 2 × 2 variables) for categorical variables. All significant levels were two-tailed and set at a level of 0.05.

RESULTS

The mean age of the study sample (n = 53) was 65.96 ± 13.09 (range: 31–89) years. Of the patients, 26 (49.1%) were women, and 27 (50.9%) were men. 33 (62.3%) were living in the city center, 48 (90.6%) were living with their families, and 49 (92.5%) belonged to an ordinary income group. Hypertension (n = 40, 75.5%), diabetes mellitus (n = 19, 35.8%), coronary artery disease (n = 17, 32.1%), and HL (n = 15, 28.3%) were the most common diseases [Table 1]. The mean hospitalization time was reported as 11.92 ± 5.85/day. The initial mean NIHSS score was reported as 6.96 ± 4.75. Serum samples were collected within a mean of 11.88 ± 2.71/h after the onset of AIS symptoms (range = 7–18/h).

Table 1

Sociodemographic data and medical history of the patients

	Patient with AIS (n=36)	Patient with AIS and MD (n=17)	P
Sociodemographic data
Age, mean±SD	65.25±13.98	67.47±11.20	0.569
Sex, n (%)
Male	19 (52.8)	8 (47.1)	0.773
Female	17 (47.2)	9 (52.9)	0.773
Habitation, n (%)
City center	23 (63.9)	10 (58.8)	1.376
Town	11 (30.6)	7 (41.2)	1.376
Village	2 (5.6)	0 (0)	1.376
Income status, n (%)
High income	2 (5.6)	2 (11.8)	0.585
Ordinary income	34 (94.4)	15 (88.2)	0.585
Living condition, n (%)
Live with family	32 (88.9)	16 (94.1)	1.000
Live alone	4 (11.1)	1 (5.9)	1.000
Medical history, n (%)
HT	27 (75)	13 (76.5)	1.000
DM	11 (30.6)	8 (47.1)	0.358
CAD	11 (30.6)	6 (32.1)	0.760
HL	9 (25)	6 (35.3)	0.520
CHF	4 (11.1)	2 (11.8)	1.000
AF	5 (13.9)	6 (35.3)	0.143
HVD	2 (5.6)	3 (17.6)	0.313
Smoking	11 (30.6)	2 (11.8)	0.813
ASA use	13 (36.1)	7 (41.2)	0.768
Anticoagulant use	0 (0)	2 (11.8)	0.099
Antihypertensive use	23 (63.9)	12 (70.6)	0.760
OAD use	8 (22.2)	3 (17.6)	1.000
Insulin use	7 (19.4)	5 (29.4)	0.490

AIS – Acute ischemic stroke; HT – Hypertension; DM – Diabetes mellitus; CAD – Coronary artery disease; HL- Hyperlipidemia; CHF – Chronic heart failure; AF – Atrial fibrillation; HVD – Heart valve disease; OAD – Oral antidiabetic; ASA – Asetil Salisilic Acid; MD – Major depressive;; SD – Standard deviation

In the depressive patient group, serum samples were collected within a mean of 11.58 ± 3.04/h after the beginning of AIS symptoms, whereas in the nondepressive patient group, serum samples were collected within a mean of 12.02 ± 2.58/h after the onset of AIS symptoms. The difference was not statistically significant (P = 0.587).

As shown in Table 2, patients with AIS were classified as follows: 13 (24.5%) large artery disease; 13 (24.5%); cardioembolic; 7 (13.2%), small vessel disease and 19 (35.8%), unknown etiology. While 44 (83%) supratentorial space lesions were most commonly observed, the following were also observed: 14 (26.4%) parietal, 13 (24.5%) basal ganglion, and 10 (18.9%) temporal region lesions. Nearly 17 (32.1%) of the 53 patients in the study group were diagnosed with major depression.

Table 2

Etiological classification and stroke location of the patients

	Patient with AIS (n=36)	Patient with AIS and MD (n=17)	P
TOAST classification, n (%)
Large artery	9 (25)	4 (23.5)	1.000
Cardio-embolism	5 (13.9)	8 (47.1)	0.016
Small vessel disease	6 (16.7)	1 (5.9)	0.408
Other cause	1 (2.8)	0 (0)	1.000
Unknown	15 (41.7)	4 (23.5)	0.235
Stroke location
Supratentorial	29 (80.6)	15 (88.2)	0.701
İnfratentorial	7 (19.4)	2 (11.8)	0.701
Frontal	6 (16.7)	3 (17.6)	1.000
Temporal	8 (22.2)	2 (11.8)	0.471
Parietal	10 (27.8)	4 (23.5)	1.000
Occipital	4 (11.1)	2 (11.8)	1.000
Cerebellar	5 (13.9)	1 (5.9)	0.651
Thalamus	3 (8.3)	0 (0)	0.543
Basal Ganglion	7 (19.4)	6 (35.3)	0.306
Brain stem	2 (5.6)	2 (11.8)	0.585

TOAST – Trial of ORG 10172 in acute stroke treatment; AIS – Acute ischemic stroke; MD – Major depressive

Table 3 demonstrates hematologic (including complete blood count) and biochemical (metabolic panel, C-reactive protein, uric acid) parameters in the major depressive or nondepressive AIS patients. There was no significant difference between the preexisting major depressive and nondepressive AIS patients regarding sociodemographic characteristics, medical histories, mean duration between beginning of AIS symptoms and collection of serum samples, the average length of stay in the hospital, NIHSS score at admission, TOAST classification, stroke location, and hematologic and biochemical parameters. In addition, there was no significant difference between the major depressive and nondepressive AIS patients in terms of serum levels of TNF-α, IL-1 β, IL-18, BDNF, and NSE [Table 3].

Table 3

Biological parameters of the patients

	Patient with AIS (n=36)	Patient with AIS and MD (n=17)	P
Biological parameters (mean±SD)
Blood glucose level	148.25±67.26	123.24±37.52	0.365
Urea	40.58±16.38	40.88±16.12	0.951
Creatine	0.27±0.45	0.41±0.50	0.334
Hemoglobin	12.69±2.14	12.8±2.17	0.752
WBC	8283.33±2080.31	8476.47±2036.70	0.235
C-reactive protein	4.47±3.30	4.88±3.88	0.692
Uric acid	5.36±1.75	5.29±1.75	0.898
TNF-α-pg/ml	34.50±40.46	35.65±36.99	0.856
IL-1β-pg/ml	3.72±4.59	3.35±1.86	0.760
IL-18-pg/ml	165.42±87.04	210.71±165.97	0.197
NSE-ng/ml	6.19±6.29	4.47±2.71	0.320
BDNF-pg/ml	729.39±500.90	751.35±642.86	0.892

WBC – White blood cells; TNF-alfa – Tumor necrosis factor-alpha; IL – Interleukin; NSE – Neuron-specific enolase; BDNF – Brain-derived neurotrophic factor; AIS – Acute ischemic stroke; MD – Major depressive; SD – Standard deviation

DISCUSSION

As far as we know, our trial is the first study in which the relationship between major depression in patients with AIS and proinflammatory cytokines (TNF-α, IL-1 β, and IL-18), BDNF, and NSE has been investigated. In previous studies, most commonly the relationship between these biological parameters and aforementioned diseases in patient groups diagnosed with AIS,[383940] major depression[17181920] and poststroke depression.[414243] In our sample, we determined that there is no significant relationship between major depression and basal proinflammatory cytokines (TNF-α, IL-1 β, and IL-18), BDNF and NSE. It is known that in studies concerning acute ischemia, an increase in proinflammatory cytokines and NSE serum levels were linearly related with infarction size, edema formation, and severity of neurological deficits whereas decreased BDNF serum levels were linearly related to increased stroke risk.[2122233034] On the other hand, studies conducted with individuals with depression revealed an increase in levels of proinflammatory cytokines and a decrease in BDNF levels.[171819202728] Although some authors reported that CSF levels of NSE were increased in psychiatric patients with major depression,[33] when studies regarding TNF-α, primarily proinflammatory serum cytokines and BDNF, are compared, studies concerning the relationship between major depression and NSE were found to be insufficient.[2044]

In our study, we had expected that in patients with major depression who experienced AIS, compared to nondepressive patients who experienced AIS the serum levels of proinflammatory cytokines would be much higher, and BDNF levels would be much lower. However, the results from our sample indicate that, even if the presence of major depression in patients affects these parameters, whether just before AIS or not, this effect disappears within hours after AIS. The insufficiency of studies concerning the relationship between NSE and depression makes it hard to clearly our results about NSE. However, a similar interpretation could be possible for NSE, too.

It has been reported that increased levels of proinflammatory cytokines and decreased serum levels of BDNF exacerbate neurological deficit and increase mortality using worsening of prognosis by effecting functional recovery.[2445] Studies have also shown that the relationship between depressive symptoms and increased stroke risk increases ischemic stroke-related morbidity and mortality.[510] Therefore, although we had posited that proinflammatory cytokines and BDNF mediate the negative effect of major depression in patients with AIS to probable prognosis, our results do not support this hypothesis.

Among the possible causes for results not in accordance with our expectations, the first place goes to proinflammatory cytokines (TNF-α, IL-1 β, IL-18) and BDNF. Before we conducted the study, our expectations were shaped because of results obtained from previous studies conducted with psychiatric patients with major depression. Most of these studies were conducted in adolescent and middle-age groups. However, there was a scarcity of relevant studies and limited data available regarding older psychiatric patients with depression.[4647] Therefore, the reason that these parameters to be irrelevant for depressive and nondepressive patient groups right after AIS could be specific to this patient group. Second, given the strict exclusion criteria of our study and NIHSS of patients at admission, it can be observed that the study group is composed of patients with strokes of low and moderate severity. Therefore, we can say that our findings show relevance for patients with strokes with low and moderate severity, rather than patients with strokes with of high severity. Third, we collected the serum samples of the patients within 24 h after the initiation of AIS symptoms. Therefore, even though our aforementioned parameters were affected before AIS in patients with major depression, hours after AIS, they had come to be at similar levels as in nondepressive patients.

We found no study examining the association between preexisting major depression and specific brain regions in AIS in the literature. In studies up to now, even some data that supported the relationship between poststroke depression and specific stroke region,[48] was concluded that the likelihood of a relationship between poststroke depression and a specific stroke region being a major risk factor was weak.[49] On the other hand, some authors stated that, in patient groups where patients were divided as depressive and nondepressive by investigation for diagnosis of poststroke depression, mean hospitalization times were similar.[50] In our sample, preexisting depression was not associated with hospital stay and specific stroke location in AIS. We think this situation may be related to the patient group comprising our sample.

When the results of our study are interpreted, some limitations should be taken into consideration. First, the study has a relatively small sample size. The strictness of the exclusion criteria of the study led to expulsion of a majority of the recruited patients. In addition, in the presence of these exclusion criteria, we think it will be very difficult to study aforementioned biological parameters by recruiting large patient numbers from a single center. Second, the study has a cross-sectional but not a prospective observational design. As noted above, in our patient group, we collected serum samples within hours after AIS. Ideally, studying the relevant biological parameters before AIS by following up patients with major depression who do not have any disease that can cause any neuroinflammatory response could provide more reliable data. However, the prospective of following up depressive patients who are not receiving antidepressant treatment and being able to collect serum samples just before AIS are quite difficult both ethically and methodologically. In addition, ensuring that post-AIS mean serum sample collecting time is similar for the depressive and nondepressive patients in our study would partially compensate for this limitation. Regardless of all these limitations, a diagnosis of major depression just before AIS using a psychiatric interview device structured by a psychiatrist and the expulsion of patients diagnosed with psychiatric disorders other than major depression are powerful parts of our study. In addition, we evaluated current depression, not a preexisting one. Diagnosis of current major depression shows the relationship between depression and relevant parameters just before AIS, compared to preexisting major depression. This is another powerful part of the study.

CONCLUSION

Regardless of limitations, our study results show that, in patients with major depression, an experience of AIS has no effect on proinflammatory cytokines (TNF-α, IL-1 β, IL-18), BDNF, and NSE in acute ischemic stroke patients. For obtaining clearer data and clearer interpretations, controlled studies that are cross-sectional and multi-centered with a large sample are needed.

Financial support and sponsorship

This research was supported by the research foundation of the Meram Medical Faculty of Necmettin Erbakan University.

Conflict of interest

There are no conflicts of interest.