So no benefit.
B-vitamin supplementation on mitigating post-stroke cognition and neuropsychiatric sequelae: A randomized controlled trial
Hong Kuang Tanhttps://orcid.org/0000-0002-4743-85151, Kaavya Narasimhaluhttps://orcid.org/0000-0001-6116-30202,*, Simon Kang Seng Tinghttps://orcid.org/0000-0003-4852-97902, Shahul Hameed2, Hui Meng Chang2, Deidre Anne De Silvahttps://orcid.org/0000-0002-7123-23812, Christopher Li Hsian Chen3, and Eng King Tan2
Background and Purpose:
Background and Purpose:
A third of stroke patients suffer from post-stroke cognitive decline, depressive symptoms, and anxiety symptoms. B-vitamin supplementation provides a possible safe and affordable treatment to mitigate post-stroke neuropsychiatric sequelae via reducing homocysteine levels. Our study aims to examine the effect of B-vitamin supplementation in the prevention of post-stroke cognitive decline, depressive symptoms, and anxiety symptoms. Our secondary aims were to investigate associations between baseline factors and the three outcomes.
Methods:
Patients were recruited as part of a Singaporean substudy of a randomized controlled trial that examined the effect of B-vitamin supplementation on recurrent cardiovascular events. Cognitive decline, depressive symptoms, and anxiety symptoms were assessed with neuropsychological assessments and Hospital Anxiety and Depression Scale 6 monthly. Cox regression analyses were performed to determine treatment efficacy. Logistic regression used to examine factors associated with cognitive decline, depressive symptoms, and anxiety symptoms.
Results:
A total of 707 were included in the analyses. Survival and hazards ratio analysis showed no treatment effect of B-vitamins on cognitive decline, depressive symptoms, and anxiety symptoms. Cognitive decline was only associated with age. Depressive symptoms were associated with large anterior cerebral infarcts and hyperlipidemia.
Conclusions:
Our study showed no benefit of supplementation with B-vitamins for post-stroke cognitive decline, depressive symptoms, or anxiety symptoms. Depressive symptoms were associated with larger anterior cerebral infarcts, which may be reflective of the disability associated with larger infarcts.
Keywords
Stroke, B-vitamin, cognition, depression, anxiety
1Duke-NUS Medical School, Singapore
2Department of Neurology, National Neuroscience Institute (Singapore General Hospital Campus), Singapore
3Department of Pharmacology, National University of Singapore, Singapore
Corresponding author(s):
Kaavya Narasimhalu, Department of Neurology, National Neuroscience Institute, SGH Campus, Outram Road, Singapore 169608. Email: nkaavya@gmail.com
*
Kaavya Narasimhalu is now affiliated to Duke-NUS Medical School, Singapore
Introduction
Background
Stroke is a leading cause of disability and 30% of stroke patients experience some form of cognitive decline within 3 years post-stroke1—which includes both dementia2 and cognitive impairment no dementia (CIND).3 In addition, stroke can also result in neuropsychiatric complications such as depression4 and anxiety.5 One contributory factor postulated for these post-stroke sequelae is hyperhomocystinemia6—which may reduce brain volume possibly through oxidative stress resulting in cognitive decline.7 Serum homocysteine levels can be lowered by B12, Folate, and B6 through methylation.8 Furthermore, an Australian substudy of the same international trial as our study found that B-vitamin supplementation associated with a lower hazard of major depression compared with placebo.9 Hence, using B-vitamins to lower post-stroke homocysteine serum concentrations was proposed as a treatment to mitigate these post-stroke cognitive decline, depression, and anxiety.10
Besides potential efficacy, there are other advantages of using B-vitamins as an alternative or augmentation to current conventional treatments. They can be safely added to current cardiovascular medications and are safer and have less adverse effects compared with prophylactically administering antidepressants to prevent post-stroke depression and anxiety. Furthermore, they may augment the efficacy of current standard antidepressant treatment as well.11 Therefore, studying the efficacy of B-vitamin supplementation on post-stroke cognitive and neuropsychiatric sequelae is important.
Although other studies examined the relationship between stroke and cognitive impairment, few studies looked specifically at the effects of B-vitamin on CIND, depressive, symptoms, and anxiety symptoms on patients who had either an ischemic stroke or a transient ischemic attack (TIA). Hence, our study (a substudy of a large scale multicentre randomized control trial designed to assess the impact of B-vitamin supplementation on recurrent vascular events in patients with a prior stroke—VITAmins To Prevent Stroke (VITATOPS))12 endeavors to examine the effects of B-vitamins on reducing the rates of post-stroke cognitive decline, depressive symptoms, and anxiety symptoms.
Aims and hypothesis
We aim to bridge this gap in the current literature by comparing the incidence rates of post-stroke cognitive impairment, depressive symptoms, and anxiety symptoms in those who received B-vitamin supplements post-stroke with those who did not. We hypothesize that patients who received B-vitamin supplementation would have reduced risk of acquiring post-stroke cognitive decline, depressive symptoms, and anxiety symptoms compared with those in the placebo group.
Methods
Patients
Patients were recruited as part of a larger multicentre, double blinded, placebo-controlled randomized trial VITATOPS12 that examined B-vitamins’ effects in reducing risk of major cardiovascular events in people who had a recent ischemic stroke or TIA link to CONSORT Statement (Appendix 1). In the Singapore study site, ischemic stroke and TIA patients who presented to Singapore General Hospital between July 2000 and May 2007 were approached for inclusion in this study within the first 7 months of their index event. Patients were excluded if they were on folate antagonists or B-vitamin supplements and were not willing to switch, had a short remaining life-expectancy, or were women of childbearing potential. However, patients on B-vitamin supplements and were willing to switch to the study medication were included.
In addition, those who fulfilled the diagnostic criteria for dementia at baseline were excluded.
Randomization
Patients were randomized by a telephone service or an interactive website using blocks stratified by hospital.
Protocol approvals, registrations, and patient consent
The study protocol was approved by Singapore General Hospital’s Institutional Review Board and Ethics Committee. Written consent was obtained from patients or legal guardians. The VITATOPS study was registered under clinical-trials.gov with the identification number NCT00097669.
Intervention
Patients enrolled in the study were required to take one composite tablet that contained 25 mg vitamin B6, 2 mg Folic acid, and 0.5 mg vitamin B12 or a placebo pill of similar color and texture daily. Adherence was checked by questioning, tablet count, and return of tablet container. Patients, clinicians, and caregivers were blinded. Patients were then followed up every 6 months up to a 5-year period after randomization.
Demographic and stroke-related measures
Baseline characteristics such as age, sex, ethnic background, smoking status, education level, medical history of diabetes, hypertension, dyslipidemia, stroke classification, ischemic heart disease, angina, myocardial infarction, previous stroke, and peripheral vascular diseases of patients were assessed. Baseline serum homocysteine, folate, and B12 levels were also recorded.
Neuropsychological assessment
Assessment of cognitive domains was performed with a battery of tests validated in Singapore13 at baseline and subsequently in 6 monthly intervals. Education-adjusted cutoffs of 1.5 standard deviations below established normal means14 were used on individual tests. Failure in at least half of the tests in a domain constituted failure in that domain. The assessment was administered in English, Malay, Mandarin, or Chinese dialects according to the subject’s habitual language. The cognitive domains measured included (a) Attention: Wechsler Memory Scale–Revised (WMS-R) Digit and Visual span,15 Auditory detection test;16 (b) Language function: Category Fluency (Animal and Food subtask)17 and Modified Boston naming test;18 (c) Verbal memory: Word List Recall (immediate, delayed recall, and delayed recognition),14 Story Recall (immediate and delayed recall); (d) Visual memory: Picture recall (immediate, delayed recall, and delayed recognition), WMS-R visual reproduction (immediate, delayed recall, and delayed recognition);15 (e) Visuoconstruction: Wechsler Memory Scale–Revised (WMS-R) subtest Visual Reproduction Copy task,15 Wechsler Adult Intelligence Scale–Revised (WAIS-R) Block Design subtest,15 and Clock Drawing;19 (f) Visuomotor speed: Symbol Digit Modality Test,20 Digit Cancelation,21 and Maze Task,22 and (g) Executive function: frontal assessment battery (FAB).23
Patients were characterized as having dementia as per Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) criteria,24 cognitive impairment, no dementia (CIND),25 or no cognitive impairments (NCI). Patients with CIND had at least one objective domain impaired but did not meet criteria for dementia diagnosis. Cognitive decline was defined as a decline from NCI to CIND or dementia, or a decline from CIND to dementia. Patients were categorized as reporting depressive symptoms or anxiety symptoms if they scored 8 or more on the respective subscales on the Hospital Anxiety and Depression Scale (HADS).26
Statistical analysis
All demographic and clinical factors were summarized based on treatment status. Differences between treatment groups and other categorical and continuous variables were tested using chi—square test and independent t-test, respectively. Univariate and multivariate cox proportional hazard (CPH) regression were used for time to cognitive decline, depressive symptoms, and anxiety symptoms to compare treatment and placebo group over the duration post enrolment. Results from CPH were expressed as hazards ratio (HR) analysis with 95% confidence interval (95% CI). Kaplan–Meier curves were also presented for cognitive decline, depressive symptoms, and anxiety symptoms. Univariate and multivariate logistic regressions were subsequently used to examine relationships between the baseline characteristics and the development of cognitive decline, depressive symptoms, or anxiety symptoms in patients. Results from logistic regression were represented in terms of odds ratios (ORs) with 95% confidence intervals (CIs). A two-tailed alpha of 0.05 was deemed statistically significant. All analyses were done using the SPSS version 25.
Results
A total of 867 patients were recruited and randomized in VITATOPS in Singapore. One hundred thirty-two were excluded as they declined cognitive assessments. An additional 28 were excluded because they had dementia at baseline. In total, 707 patients were included in the analyses—349 patients in the placebo group and 358 patients in the treatment group (Figure 1). The mean time to recruitment was 7.5 days of the index stroke (median = 11 days, interquartile range = 3–24 days). There were no significant differences in the baseline characteristics between the placebo and treatment group (Table 1).
Figure 1. Consort figure. Number of patient included for analysis after excluding those who were not in VITATOPS-Cog and had dementia at baseline.
CPH used to compare differences between treatment and placebo group on the duration to cognitive decline (Total n = 594, Event n = 89, Censored n = 505, HR: 1.06, CI: 0.70–1.61, p = 0.79), depressive symptoms (Total n = 604, Event n = 74, Censored n = 530, HR: 1.45, CI: 0.89–2.35, p = 0.14), and anxiety symptoms (Total n = 604, Event n = 84, Censored n = 520, HR: 1.33, CI: 0.86–2.04, p = 0.20) found no differences between the groups. The Kaplan–Meier curves showed treatment group having lower incidence of anxiety symptoms compared with placebo group although the difference was not statistically significant (Figure 2).
Figure 2. Kaplan–Meier curves for duration to cognitive decline, severe depressive and anxiety symptoms. (a) Survival function for cognitive decline. (b) Survival function for depressive symptoms. (c) Survival function for anxiety symptoms.
Table 2 summarizes the logistic regression analysis results for outcome of cognitive decline. In univariate analysis, cognitive decline was associated with posterior circulation infarction (OR: 2.70, CI: 1.12–6.49, p = 0.03) and age (OR: 1.02, CI: 1.00–1.04, p = 0.03). In multivariate analysis, age (OR: 1.02, CI: 1.02–1.04, p = 0.03) was the only independent predictor of post-stroke cognitive decline.
Table 3 summarizes the logistic regression analysis results for the outcome of depressive symptoms. In univariate analysis, depressive symptoms were associated with LACI (OR: 2.37, CI: 0.91–7.70, p = 0.04), TACI (OR: 9.51, CI: 1.88–48.19, p = 0.01), hypertension (OR: 2.27, CI: 1.19–4.43, p = 0.01), and hyperlipidemia (OR: 0.58, CI: 0.35–0.95, p = 0.03). In multivariate analysis, TACI (OR: 10.60, CI: 2.02–55.76, p = 0.01), hypertension (OR: 2.42, CI: 1.24–4.72, p = 0.01), and lack of hyperlipidemia (OR: 0.54, CI: 0.23–0.90, p = 0.02) were independent predictors of depressive symptoms.
Table 4 summarizes the univariate analysis results for the outcome of anxiety symptoms. Univariate analysis showed no association between anxiety symptoms and baseline variables.
Discussion
In our study, B-vitamin supplementation was not associated with a reduction of post-stroke cognitive impairment, depressive symptoms, or anxiety symptoms. We found that the development of post-stroke cognitive impairment associated with older age, while the development of post-stroke depressive symptoms associated with large anterior hemispheric strokes, hypertension, and lack of hyperlipidemia.
B-vitamin supplementation’s inefficacy in mitigating cognitive decline in our study echoes results from other studies.27 This is likely because hyperhomocysteinemia is just one of many factors mediating post-stroke cognitive decline. Other risk factors for post-stroke cognitive decline include advanced age and type of stroke—although the latter association was not found in our study.28,29 This suggests that current standard of care, even when coupled with B-vitamin supplementation, is insufficient in mitigating post-stroke cognitive decline. Another possibility for the absent treatment effects could be the mean baseline homocysteine levels in both groups being in the normal range (<15μmol/L)—which are already protective against cognitive decline.30
Large anterior hemispheric strokes were associated with developing depressive symptoms in our study. We postulate that extensive neurostructural damage results in greater disability—which may precipitate depressive symptoms.31 This association was reflected in our findings and that of others.32,33 We postulate that the association between hypertension and depressive symptoms is due to increased incidence of small vessel disease34 among hypertensives. Our finding that hyperlipidemia protects against developing depressive symptoms is possibly due to statins’ potential protective effects against depression.35,36 Previous studies demonstrated that statins modulate inflammation—a postulated cause of depression in stroke patients.37 Since almost half of our participants in each group have hyperlipidemia, and were likely on statins, the therapeutic effect of B-vitamins on mitigating post-stroke depressive symptoms could have been masked. An earlier Australian substudy of VITATOPS found that B-vitamin supplementation reduces the hazard of major depression compared with placebo. Our findings likely differed from theirs’ due to different measurement tools being used (Mini-international Neuropsychiatric Interview (MINI) instead of HADS). Also, they split their sample into major and minor depression based on clinical assessment while our outcome measure was a screening tool (HADS). Furthermore, they had a longer duration of follow-up.
This study has several limitations. First, the outcome of depressive and anxiety symptoms was based on the HADS rather than clinical diagnosis—making it difficult to ascertain whether patients were truly clinically depressed or anxious which may dilute our associations. Second, while previous studies showed rates of 30–50% for post-stroke depressive and anxiety symptoms, our study’s rates (13–18%) likely resulted in a lack of statistical power. Third, the study design meant that patients could be on B-vitamin supplementation prior to randomization—diluting the effect of B-vitamin supplementation on our post-stroke outcomes.
In conclusion, our study did not find significant benefits of B-vitamin supplementation in mitigating post-stroke neuropsychiatric sequelae. Hence, further studies would need to address the limitations in this study and explore other factors affecting post-stroke cognitive decline, depressive symptoms, and anxiety symptoms.
Keywords
Stroke, B-vitamin, cognition, depression, anxiety
1Duke-NUS Medical School, Singapore
2Department of Neurology, National Neuroscience Institute (Singapore General Hospital Campus), Singapore
3Department of Pharmacology, National University of Singapore, Singapore
Corresponding author(s):
Kaavya Narasimhalu, Department of Neurology, National Neuroscience Institute, SGH Campus, Outram Road, Singapore 169608. Email: nkaavya@gmail.com
*
Kaavya Narasimhalu is now affiliated to Duke-NUS Medical School, Singapore
Introduction
Background
Stroke is a leading cause of disability and 30% of stroke patients experience some form of cognitive decline within 3 years post-stroke1—which includes both dementia2 and cognitive impairment no dementia (CIND).3 In addition, stroke can also result in neuropsychiatric complications such as depression4 and anxiety.5 One contributory factor postulated for these post-stroke sequelae is hyperhomocystinemia6—which may reduce brain volume possibly through oxidative stress resulting in cognitive decline.7 Serum homocysteine levels can be lowered by B12, Folate, and B6 through methylation.8 Furthermore, an Australian substudy of the same international trial as our study found that B-vitamin supplementation associated with a lower hazard of major depression compared with placebo.9 Hence, using B-vitamins to lower post-stroke homocysteine serum concentrations was proposed as a treatment to mitigate these post-stroke cognitive decline, depression, and anxiety.10
Besides potential efficacy, there are other advantages of using B-vitamins as an alternative or augmentation to current conventional treatments. They can be safely added to current cardiovascular medications and are safer and have less adverse effects compared with prophylactically administering antidepressants to prevent post-stroke depression and anxiety. Furthermore, they may augment the efficacy of current standard antidepressant treatment as well.11 Therefore, studying the efficacy of B-vitamin supplementation on post-stroke cognitive and neuropsychiatric sequelae is important.
Although other studies examined the relationship between stroke and cognitive impairment, few studies looked specifically at the effects of B-vitamin on CIND, depressive, symptoms, and anxiety symptoms on patients who had either an ischemic stroke or a transient ischemic attack (TIA). Hence, our study (a substudy of a large scale multicentre randomized control trial designed to assess the impact of B-vitamin supplementation on recurrent vascular events in patients with a prior stroke—VITAmins To Prevent Stroke (VITATOPS))12 endeavors to examine the effects of B-vitamins on reducing the rates of post-stroke cognitive decline, depressive symptoms, and anxiety symptoms.
Aims and hypothesis
We aim to bridge this gap in the current literature by comparing the incidence rates of post-stroke cognitive impairment, depressive symptoms, and anxiety symptoms in those who received B-vitamin supplements post-stroke with those who did not. We hypothesize that patients who received B-vitamin supplementation would have reduced risk of acquiring post-stroke cognitive decline, depressive symptoms, and anxiety symptoms compared with those in the placebo group.
Methods
Patients
Patients were recruited as part of a larger multicentre, double blinded, placebo-controlled randomized trial VITATOPS12 that examined B-vitamins’ effects in reducing risk of major cardiovascular events in people who had a recent ischemic stroke or TIA link to CONSORT Statement (Appendix 1). In the Singapore study site, ischemic stroke and TIA patients who presented to Singapore General Hospital between July 2000 and May 2007 were approached for inclusion in this study within the first 7 months of their index event. Patients were excluded if they were on folate antagonists or B-vitamin supplements and were not willing to switch, had a short remaining life-expectancy, or were women of childbearing potential. However, patients on B-vitamin supplements and were willing to switch to the study medication were included.
In addition, those who fulfilled the diagnostic criteria for dementia at baseline were excluded.
Randomization
Patients were randomized by a telephone service or an interactive website using blocks stratified by hospital.
Protocol approvals, registrations, and patient consent
The study protocol was approved by Singapore General Hospital’s Institutional Review Board and Ethics Committee. Written consent was obtained from patients or legal guardians. The VITATOPS study was registered under clinical-trials.gov with the identification number NCT00097669.
Intervention
Patients enrolled in the study were required to take one composite tablet that contained 25 mg vitamin B6, 2 mg Folic acid, and 0.5 mg vitamin B12 or a placebo pill of similar color and texture daily. Adherence was checked by questioning, tablet count, and return of tablet container. Patients, clinicians, and caregivers were blinded. Patients were then followed up every 6 months up to a 5-year period after randomization.
Demographic and stroke-related measures
Baseline characteristics such as age, sex, ethnic background, smoking status, education level, medical history of diabetes, hypertension, dyslipidemia, stroke classification, ischemic heart disease, angina, myocardial infarction, previous stroke, and peripheral vascular diseases of patients were assessed. Baseline serum homocysteine, folate, and B12 levels were also recorded.
Neuropsychological assessment
Assessment of cognitive domains was performed with a battery of tests validated in Singapore13 at baseline and subsequently in 6 monthly intervals. Education-adjusted cutoffs of 1.5 standard deviations below established normal means14 were used on individual tests. Failure in at least half of the tests in a domain constituted failure in that domain. The assessment was administered in English, Malay, Mandarin, or Chinese dialects according to the subject’s habitual language. The cognitive domains measured included (a) Attention: Wechsler Memory Scale–Revised (WMS-R) Digit and Visual span,15 Auditory detection test;16 (b) Language function: Category Fluency (Animal and Food subtask)17 and Modified Boston naming test;18 (c) Verbal memory: Word List Recall (immediate, delayed recall, and delayed recognition),14 Story Recall (immediate and delayed recall); (d) Visual memory: Picture recall (immediate, delayed recall, and delayed recognition), WMS-R visual reproduction (immediate, delayed recall, and delayed recognition);15 (e) Visuoconstruction: Wechsler Memory Scale–Revised (WMS-R) subtest Visual Reproduction Copy task,15 Wechsler Adult Intelligence Scale–Revised (WAIS-R) Block Design subtest,15 and Clock Drawing;19 (f) Visuomotor speed: Symbol Digit Modality Test,20 Digit Cancelation,21 and Maze Task,22 and (g) Executive function: frontal assessment battery (FAB).23
Patients were characterized as having dementia as per Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) criteria,24 cognitive impairment, no dementia (CIND),25 or no cognitive impairments (NCI). Patients with CIND had at least one objective domain impaired but did not meet criteria for dementia diagnosis. Cognitive decline was defined as a decline from NCI to CIND or dementia, or a decline from CIND to dementia. Patients were categorized as reporting depressive symptoms or anxiety symptoms if they scored 8 or more on the respective subscales on the Hospital Anxiety and Depression Scale (HADS).26
Statistical analysis
All demographic and clinical factors were summarized based on treatment status. Differences between treatment groups and other categorical and continuous variables were tested using chi—square test and independent t-test, respectively. Univariate and multivariate cox proportional hazard (CPH) regression were used for time to cognitive decline, depressive symptoms, and anxiety symptoms to compare treatment and placebo group over the duration post enrolment. Results from CPH were expressed as hazards ratio (HR) analysis with 95% confidence interval (95% CI). Kaplan–Meier curves were also presented for cognitive decline, depressive symptoms, and anxiety symptoms. Univariate and multivariate logistic regressions were subsequently used to examine relationships between the baseline characteristics and the development of cognitive decline, depressive symptoms, or anxiety symptoms in patients. Results from logistic regression were represented in terms of odds ratios (ORs) with 95% confidence intervals (CIs). A two-tailed alpha of 0.05 was deemed statistically significant. All analyses were done using the SPSS version 25.
Results
A total of 867 patients were recruited and randomized in VITATOPS in Singapore. One hundred thirty-two were excluded as they declined cognitive assessments. An additional 28 were excluded because they had dementia at baseline. In total, 707 patients were included in the analyses—349 patients in the placebo group and 358 patients in the treatment group (Figure 1). The mean time to recruitment was 7.5 days of the index stroke (median = 11 days, interquartile range = 3–24 days). There were no significant differences in the baseline characteristics between the placebo and treatment group (Table 1).
Figure 1. Consort figure. Number of patient included for analysis after excluding those who were not in VITATOPS-Cog and had dementia at baseline.
CPH used to compare differences between treatment and placebo group on the duration to cognitive decline (Total n = 594, Event n = 89, Censored n = 505, HR: 1.06, CI: 0.70–1.61, p = 0.79), depressive symptoms (Total n = 604, Event n = 74, Censored n = 530, HR: 1.45, CI: 0.89–2.35, p = 0.14), and anxiety symptoms (Total n = 604, Event n = 84, Censored n = 520, HR: 1.33, CI: 0.86–2.04, p = 0.20) found no differences between the groups. The Kaplan–Meier curves showed treatment group having lower incidence of anxiety symptoms compared with placebo group although the difference was not statistically significant (Figure 2).
Figure 2. Kaplan–Meier curves for duration to cognitive decline, severe depressive and anxiety symptoms. (a) Survival function for cognitive decline. (b) Survival function for depressive symptoms. (c) Survival function for anxiety symptoms.
Table 2 summarizes the logistic regression analysis results for outcome of cognitive decline. In univariate analysis, cognitive decline was associated with posterior circulation infarction (OR: 2.70, CI: 1.12–6.49, p = 0.03) and age (OR: 1.02, CI: 1.00–1.04, p = 0.03). In multivariate analysis, age (OR: 1.02, CI: 1.02–1.04, p = 0.03) was the only independent predictor of post-stroke cognitive decline.
Table 3 summarizes the logistic regression analysis results for the outcome of depressive symptoms. In univariate analysis, depressive symptoms were associated with LACI (OR: 2.37, CI: 0.91–7.70, p = 0.04), TACI (OR: 9.51, CI: 1.88–48.19, p = 0.01), hypertension (OR: 2.27, CI: 1.19–4.43, p = 0.01), and hyperlipidemia (OR: 0.58, CI: 0.35–0.95, p = 0.03). In multivariate analysis, TACI (OR: 10.60, CI: 2.02–55.76, p = 0.01), hypertension (OR: 2.42, CI: 1.24–4.72, p = 0.01), and lack of hyperlipidemia (OR: 0.54, CI: 0.23–0.90, p = 0.02) were independent predictors of depressive symptoms.
Table 4 summarizes the univariate analysis results for the outcome of anxiety symptoms. Univariate analysis showed no association between anxiety symptoms and baseline variables.
Discussion
In our study, B-vitamin supplementation was not associated with a reduction of post-stroke cognitive impairment, depressive symptoms, or anxiety symptoms. We found that the development of post-stroke cognitive impairment associated with older age, while the development of post-stroke depressive symptoms associated with large anterior hemispheric strokes, hypertension, and lack of hyperlipidemia.
B-vitamin supplementation’s inefficacy in mitigating cognitive decline in our study echoes results from other studies.27 This is likely because hyperhomocysteinemia is just one of many factors mediating post-stroke cognitive decline. Other risk factors for post-stroke cognitive decline include advanced age and type of stroke—although the latter association was not found in our study.28,29 This suggests that current standard of care, even when coupled with B-vitamin supplementation, is insufficient in mitigating post-stroke cognitive decline. Another possibility for the absent treatment effects could be the mean baseline homocysteine levels in both groups being in the normal range (<15μmol/L)—which are already protective against cognitive decline.30
Large anterior hemispheric strokes were associated with developing depressive symptoms in our study. We postulate that extensive neurostructural damage results in greater disability—which may precipitate depressive symptoms.31 This association was reflected in our findings and that of others.32,33 We postulate that the association between hypertension and depressive symptoms is due to increased incidence of small vessel disease34 among hypertensives. Our finding that hyperlipidemia protects against developing depressive symptoms is possibly due to statins’ potential protective effects against depression.35,36 Previous studies demonstrated that statins modulate inflammation—a postulated cause of depression in stroke patients.37 Since almost half of our participants in each group have hyperlipidemia, and were likely on statins, the therapeutic effect of B-vitamins on mitigating post-stroke depressive symptoms could have been masked. An earlier Australian substudy of VITATOPS found that B-vitamin supplementation reduces the hazard of major depression compared with placebo. Our findings likely differed from theirs’ due to different measurement tools being used (Mini-international Neuropsychiatric Interview (MINI) instead of HADS). Also, they split their sample into major and minor depression based on clinical assessment while our outcome measure was a screening tool (HADS). Furthermore, they had a longer duration of follow-up.
This study has several limitations. First, the outcome of depressive and anxiety symptoms was based on the HADS rather than clinical diagnosis—making it difficult to ascertain whether patients were truly clinically depressed or anxious which may dilute our associations. Second, while previous studies showed rates of 30–50% for post-stroke depressive and anxiety symptoms, our study’s rates (13–18%) likely resulted in a lack of statistical power. Third, the study design meant that patients could be on B-vitamin supplementation prior to randomization—diluting the effect of B-vitamin supplementation on our post-stroke outcomes.
In conclusion, our study did not find significant benefits of B-vitamin supplementation in mitigating post-stroke neuropsychiatric sequelae. Hence, further studies would need to address the limitations in this study and explore other factors affecting post-stroke cognitive decline, depressive symptoms, and anxiety symptoms.
No comments:
Post a Comment