ORIGINAL ARTICLE
The Effect of Lithium in Post-Stroke Motor Recovery: A Double-Blind, Placebo-Controlled, Randomized Clinical Trial Seyed Ehsan Mohammadianinejad, MD,* Nastaran Majdinasab, MD,* Seyed Aidin Sajedi, MD,* Fahimeh Abdollahi, MD,† Mehdi Masoudi Moqaddam, MD,* and Fatemeh Sadr, MD* Objective: Evidences from cultured cells and animal models of ischemia suggest that lithium has neuroprotective and neurotrophic effects and may play a desirable role in reducing infarct volume and even improving the brain insults from stroke. The aim of this study was to evaluate the efficacy of lithium in early motor recovery of patients after ischemic stroke. Methods: Eighty patients with first ever stroke, allocated randomly in lithium, 300 mg twice daily, or placebo. Treatment was initiated 48 hours after stroke and continued for 30 days. Modified National Institute of Health Stroke Scale (mNIHSS) and hand subsection of Fugl-Meyer Assessment (hFMA) were used to evaluate impairment on the fifth and 30th day of treatment. Results: Sixty-six subjects (32 subjects in the lithium group and 34 subjects in the placebo group) completed the study. There were no significant differences in the improvement in mNIHSS (P=0.40) and hFMA (P=0.07) after 30 days. However, a subgroup analysis showed that patients with cortical strokes in the lithium group had significantly better improvement in both mNIHSS and hFMA in comparison to the placebo group (P=0.003). Approximately 44% (n=14) of patients in the lithium group, mainly from the cortical stroke subgroup, regained more than 25% of full function based on hFMA, whereas this rate in the placebo group was 14.7% (n=5; P=0.009). Conclusion: The observed discrete difference between the lithium group and the placebo group in the cortical stroke subgroup may suggest an enhanced motor recovery after stroke by using an early treatment with a low dose of lithium carbonate. However, a larger trial with more patients with cortical stroke is needed to investigate this effect better. Key Words: ischemic stroke, treatment, lithium, neuroprotection, motor rehabilitation, functional recovery, clinical trial (Clin Neuropharm 2014;37: 73–78)
I
n the past 2 decades, considerable amount of data about the neuroprotective and neurotrophic effects of lithium have been gathered.1 Now, we know at least some parts of neurochemical processes underlying the lithium-induced neuroprotection. These mainly include prevention of N-methyl-D-aspartate receptor– induced excitotoxicity,2 up-regulation of brain-derived neurotrophic factor (BDNF),3 induction of the phosphatidylinositol 3-kinase/Akt cell survival pathway,4 inhibition of glycogen synthase kinase-3β (GSK-3β) that is a proapoptotic molecule,5 and increasing expression of Bcl-2 as a major antiapoptotic protein.6 In addition to studies in cultured cells, there are significant evidences from animal models of ischemia suggesting that lithium
Departments of *Neurology and †Internal Medicine, Golestan Hospital, Ahvaz Jundishapur University of Medical Sciences, Ahvaz, Iran. Conflicts of Interest and Source of Funding: This study was supported by the deputy of research of Ahvaz Jundishapur University of Medical Sciences. The authors have no conflicts of interest to declare. Address correspondence and reprint requests to Seyed Aidin Sajedi, MD, Department of Neurology, Golestan Hospital, Ahvaz, Iran; E-mail: [email protected] Copyright © 2014 by Lippincott Williams & Wilkins DOI: 10.1097/WNF.0000000000000028
may exert a desirable role in reducing infarct volume, neurological deficits, and even improving the brain insults from stroke.1 Considering the supporting evidences from in vitro and in vivo studies, and regarding the fact that lithium has been used as a mood stabilizer since 1950 7 and its safe serum level, toxic dose, and potential adverse effects for human have been determined,8 we felt that it is the appropriate time to examine its potential efficacy on the recovery of post-stroke population. As there have been no previous studies of lithium effects in recovery of patients with ischemic stroke, we decided to evaluate the efficacy of lithium in early motor recovery after stroke with a special focus on distal upper limb motor recovery of patients with ischemic stroke in middle cerebral artery (MCA) territory to obtain pilot data for supporting larger trials in the future.
MATERIALS AND METHODS Study Design and Population This was a single-center, placebo-controlled, double-blind, and randomized clinical trial. Subjects were recruited from patients hospitalized in neurology ward with diagnosis of first ever ischemic stroke, immediately (ie, within 2 days) after stroke onset and among them who were not eligible for antithrombotic therapy owing to its time limitation. All patients were allocated randomly to group A (lithium carbonate, 300-mg tablet 2 times daily + routine stroke treatment regimen) and group B (placebo, 1 tablet [talc] 2 times daily + routine stroke treatment regimen) according to a balanced 1:1 randomization based on a computer-generated randomization table. Lithium and placebo were prescribed for 30 days and started 48 hours after initiation of stroke manifestations. Our routine stroke treatment included antiplatelet and statin as well as any drug necessary for patients’ underlying diseases, and a program of physical and occupational therapy every 2 days. Stroke rehabilitation strategy in our center generally consisted of 2 main stages: (1) acute stage (the first month), which mainly uses splint and orthosis, positioning, stretching, exercises for strengthening, transcutaneous electrical nerve stimulation, initials of occupational therapy, and so on based on the patient’s situation. (2) subacute and chronic stage, which includes more focuses on strengthening, functional and occupational therapy, balance training, and constrained induced movement therapy for suitable candidates. Except for the first week of hospitalization that the program is offered every day, the rest of the program is offered for 3 hours every 2 days in our routine practice. Rehabilitation program of this study was set according to, and the same as, the program that is offered to all stroke patients in this center. As our study included patients in their first 30 days after stroke, our rehabilitation program involved the acute stage of our center’s rehabilitation program. By definition, this rehabilitation regimen cannot be regarded an intense physical therapy. However, intense physical therapy has better proven effects, but due to limitation of resources of our health system and high rate of stroke in this country, it is the maximum rehabilitation treatment that can be offered to stroke
Clinical Neuropharmacology • Volume 37, Number 3, May/June 2014
www.clinicalneuropharm.com
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
73
Mohammadianinejad et al
Clinical Neuropharmacology • Volume 37, Number 3, May/June 2014
patients, according to our national insurance system policies. Because in real practice, it is the best that we can offer to our patients and the ability to extend the result to our daily practice was important for researchers, we decided to act the same as our center’s stroke policy. Physical and occupational therapies were done in our hospital rehabilitation center or at home by appointed staff of this center for patients whom their transfer to rehabilitation center was problematic for their relatives. Our physiotherapists were equipped with portable devices to offer most of the physical therapy offered in rehabilitation facility to patients who could not attend the rehabilitation center. Except for 2 members of research team, one neurologist and one internist, all assessors and patients were unaware about the treatment allocation. Randomization was performed under the supervision of the unblinded neurologist who also had the responsibility of managing patients’ treatment and workup during hospitalization. The unblinded internist had the responsibility to monitor probable drug adverse effects and exert appropriate management in necessary cases. The inclusion and exclusion criteria are listed below.
Inclusion Criteria 1. Adults aged between 50 and 80 years old 2. First ever ischemic stroke in the territory of middle cerebral artery and including classes 1, 3, 4, and 5 subtypes of trial of Org 10172 in acute stroke treatment (TOAST) classification system9 (ie, all subtypes of ischemic stroke except embolic stroke with cardiac origin) with a compatible brain magnetic resonance imaging or computed tomography scan 3. Appropriate level of consciousness, defined as score 1 or less from each of 1b and 1c section of the modified National Institutes of Health Stroke Scale (mNIHSS)10 at the time of entrance in the study 4. Motor power of grade 0/5 to 3/5 on the Medical Research Council scale in at least one upper limb at the time of entrance in the study 5. Maximum time passed of less than 48 hours from stroke onset 6. Nonmenopause women should not be pregnant or lactating mothers 7. Written informed consent for participation in study (obtained from patient or his/her legal representative)
Exclusion Criteria 1. Coexistence of intracerebral hemorrhage (subarachnoid or intraparenchymal) or hemorrhagic conversion during study period. 2. Severe sustained loss of consciousness (score >2 from 1b and 1c questions of mNIHSS during study period 3. Wernicke aphasia (pure or as a part of constellation of other symptoms) 4. Organ failures that may interfere with consumption of lithium, including: Acute or chronic kidney disease (serum creatinine >1.2) Acute or chronic heart failure (ejection fraction 16). For evaluating improvement of impairment after stroke, our primary efficacy end points were changes in patients’ scores of mNIHSS and subsection of hand assessment of Fugl-Meyer assessment11 (hFMA) including subsections 8a, 8b, 8c, 8d, 8e, 8f, and 8f . The secondary end point was regaining more than 25% of the full motor function of hand movements based on hFMA. According to hFMA, a patient with full hand motor function will gain a total score of 14; therefore, regaining more than 3.5 points in hFMA after treatment was regarded as secondary end point achievement. For example, if one patient’s hFMA on the first assessment was 4 and after 30 days he/she achieved the score of 8 on hFMA, recovery percentage was calculated as [(8–4)/14]100=28%, and it was categorized in the group with regaining more than 25% of the full function. Fugl-Meyer assessment and NIHSS were used as our assessing tools mainly because these are impairment scores rather than being disability scores like Barthel index or Rankin scale. The advancement of an impairment score is that it can assess the degree of recovery or improvement of function rather than measure the amount of compensation. Compensation plays a major role in decreasing the level of disability, like the time that a right-handed patient with right arm paresis learn to conduct daily living activities with the left hand. In such instances, disability in practice has been decreased, but it cannot be regarded as true improvement. The first aim of this study was to find whether a low to mid dose of lithium may have any probable effect on actual improvement of neurological function, if being used in acute state of stroke. As lithium reaches to its effective serum level after 5 days of consumption, we compared impairment scores at the fifth day and then at the 30th day of entrance in the study.
Other Tests We checked all of the following tests at baseline for all subjects. Electrocardiogram; chest x-ray; echocardiography; © 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Clinical Neuropharmacology • Volume 37, Number 3, May/June 2014
and routine blood tests involving complete blood count, renal function test, thyroid function test, lipid profile, glucose, and electrolyte. At the fifth and the 30th days of entrance in the study, routine blood tests were checked again for all participants Serum lithium level was checked from the blood samples taken at the fifth and 30th days after initiation of study for patients in group A. Results of routine blood tests and lithium level were evaluated by the unblinded internist who was responsible for checking and managing adverse events. Our aim was to achieve a serum lithium level of 0.4 to 0.8 mmol/L by the fifth day of treatment. In the cases where serum lithium level at the fifth day was between 0.8 and 1.2 mmol/L and serum creatinine was less than 1.2 mg/dL, lithium carbonate dosage was decreased to 300 mg daily. For each case where decreasing the lithium dosage was done, dosage of placebo was decreased in one randomly selected patient from group B accordingly. The study was approved by the medical ethics committee of Ahvaz Jundishapur University of Medical Sciences (registration number: ETH-403), and it was registered in Iranian Registry of Clinical Trials (IRCT), available at www.irct.ir under registration number: IRCT2012072210361N1.
Sample Size This study was designed basically as a pilot study. Previously, Duncan et al12 in a cohort study of improvement after stroke by using FMA had estimated that achieving 50% improvement after 6 months can be an acceptable improvement for determining the effect size. In addition, their study showed that near half of spontaneous improvement will take place at the first month after stroke. Therefore, we estimated that sample size should be 30 patients in each arm of study based on a priori power analysis13 to detect an effect size d of 27 points (23% vs 50%) with α=0.05 (1-tailed) and 80% power. Nevertheless, with the prediction of
Lithium's Effect in Post-Stroke Motor Recovery
loss due to morbidity or mortality due to ischemic stroke or other related conditions, it was decided to recruit at least 40 patients in each arm of the study.
Statistical Analysis Kolmogorov–Smirnov Z was used to check whether variables are normally distributed. Baseline variables were compared using a 2-group t test and the Mann-Whitney U test for continuous variables, and the χ2 test and the Fisher exact test for categorical variables. For evaluating efficacy variables, we made comparison between 2 groups at baseline and the end of treatment. Since the change from baseline was the primary outcome, a per-protocol analysis was done. All analyses were done using the Statistical Package for Social Sciences version 16, and P
The Effect of Lithium in Post-Stroke Motor Recovery: A Double-Blind, Placebo-Controlled, Randomized Clinical Trial Seyed Ehsan Mohammadianinejad, MD,* Nastaran Majdinasab, MD,* Seyed Aidin Sajedi, MD,* Fahimeh Abdollahi, MD,† Mehdi Masoudi Moqaddam, MD,* and Fatemeh Sadr, MD* Objective: Evidences from cultured cells and animal models of ischemia suggest that lithium has neuroprotective and neurotrophic effects and may play a desirable role in reducing infarct volume and even improving the brain insults from stroke. The aim of this study was to evaluate the efficacy of lithium in early motor recovery of patients after ischemic stroke. Methods: Eighty patients with first ever stroke, allocated randomly in lithium, 300 mg twice daily, or placebo. Treatment was initiated 48 hours after stroke and continued for 30 days. Modified National Institute of Health Stroke Scale (mNIHSS) and hand subsection of Fugl-Meyer Assessment (hFMA) were used to evaluate impairment on the fifth and 30th day of treatment. Results: Sixty-six subjects (32 subjects in the lithium group and 34 subjects in the placebo group) completed the study. There were no significant differences in the improvement in mNIHSS (P=0.40) and hFMA (P=0.07) after 30 days. However, a subgroup analysis showed that patients with cortical strokes in the lithium group had significantly better improvement in both mNIHSS and hFMA in comparison to the placebo group (P=0.003). Approximately 44% (n=14) of patients in the lithium group, mainly from the cortical stroke subgroup, regained more than 25% of full function based on hFMA, whereas this rate in the placebo group was 14.7% (n=5; P=0.009). Conclusion: The observed discrete difference between the lithium group and the placebo group in the cortical stroke subgroup may suggest an enhanced motor recovery after stroke by using an early treatment with a low dose of lithium carbonate. However, a larger trial with more patients with cortical stroke is needed to investigate this effect better. Key Words: ischemic stroke, treatment, lithium, neuroprotection, motor rehabilitation, functional recovery, clinical trial (Clin Neuropharm 2014;37: 73–78)
I
n the past 2 decades, considerable amount of data about the neuroprotective and neurotrophic effects of lithium have been gathered.1 Now, we know at least some parts of neurochemical processes underlying the lithium-induced neuroprotection. These mainly include prevention of N-methyl-D-aspartate receptor– induced excitotoxicity,2 up-regulation of brain-derived neurotrophic factor (BDNF),3 induction of the phosphatidylinositol 3-kinase/Akt cell survival pathway,4 inhibition of glycogen synthase kinase-3β (GSK-3β) that is a proapoptotic molecule,5 and increasing expression of Bcl-2 as a major antiapoptotic protein.6 In addition to studies in cultured cells, there are significant evidences from animal models of ischemia suggesting that lithium
Departments of *Neurology and †Internal Medicine, Golestan Hospital, Ahvaz Jundishapur University of Medical Sciences, Ahvaz, Iran. Conflicts of Interest and Source of Funding: This study was supported by the deputy of research of Ahvaz Jundishapur University of Medical Sciences. The authors have no conflicts of interest to declare. Address correspondence and reprint requests to Seyed Aidin Sajedi, MD, Department of Neurology, Golestan Hospital, Ahvaz, Iran; E-mail: [email protected] Copyright © 2014 by Lippincott Williams & Wilkins DOI: 10.1097/WNF.0000000000000028
may exert a desirable role in reducing infarct volume, neurological deficits, and even improving the brain insults from stroke.1 Considering the supporting evidences from in vitro and in vivo studies, and regarding the fact that lithium has been used as a mood stabilizer since 1950 7 and its safe serum level, toxic dose, and potential adverse effects for human have been determined,8 we felt that it is the appropriate time to examine its potential efficacy on the recovery of post-stroke population. As there have been no previous studies of lithium effects in recovery of patients with ischemic stroke, we decided to evaluate the efficacy of lithium in early motor recovery after stroke with a special focus on distal upper limb motor recovery of patients with ischemic stroke in middle cerebral artery (MCA) territory to obtain pilot data for supporting larger trials in the future.
MATERIALS AND METHODS Study Design and Population This was a single-center, placebo-controlled, double-blind, and randomized clinical trial. Subjects were recruited from patients hospitalized in neurology ward with diagnosis of first ever ischemic stroke, immediately (ie, within 2 days) after stroke onset and among them who were not eligible for antithrombotic therapy owing to its time limitation. All patients were allocated randomly to group A (lithium carbonate, 300-mg tablet 2 times daily + routine stroke treatment regimen) and group B (placebo, 1 tablet [talc] 2 times daily + routine stroke treatment regimen) according to a balanced 1:1 randomization based on a computer-generated randomization table. Lithium and placebo were prescribed for 30 days and started 48 hours after initiation of stroke manifestations. Our routine stroke treatment included antiplatelet and statin as well as any drug necessary for patients’ underlying diseases, and a program of physical and occupational therapy every 2 days. Stroke rehabilitation strategy in our center generally consisted of 2 main stages: (1) acute stage (the first month), which mainly uses splint and orthosis, positioning, stretching, exercises for strengthening, transcutaneous electrical nerve stimulation, initials of occupational therapy, and so on based on the patient’s situation. (2) subacute and chronic stage, which includes more focuses on strengthening, functional and occupational therapy, balance training, and constrained induced movement therapy for suitable candidates. Except for the first week of hospitalization that the program is offered every day, the rest of the program is offered for 3 hours every 2 days in our routine practice. Rehabilitation program of this study was set according to, and the same as, the program that is offered to all stroke patients in this center. As our study included patients in their first 30 days after stroke, our rehabilitation program involved the acute stage of our center’s rehabilitation program. By definition, this rehabilitation regimen cannot be regarded an intense physical therapy. However, intense physical therapy has better proven effects, but due to limitation of resources of our health system and high rate of stroke in this country, it is the maximum rehabilitation treatment that can be offered to stroke
Clinical Neuropharmacology • Volume 37, Number 3, May/June 2014
www.clinicalneuropharm.com
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
73
Mohammadianinejad et al
Clinical Neuropharmacology • Volume 37, Number 3, May/June 2014
patients, according to our national insurance system policies. Because in real practice, it is the best that we can offer to our patients and the ability to extend the result to our daily practice was important for researchers, we decided to act the same as our center’s stroke policy. Physical and occupational therapies were done in our hospital rehabilitation center or at home by appointed staff of this center for patients whom their transfer to rehabilitation center was problematic for their relatives. Our physiotherapists were equipped with portable devices to offer most of the physical therapy offered in rehabilitation facility to patients who could not attend the rehabilitation center. Except for 2 members of research team, one neurologist and one internist, all assessors and patients were unaware about the treatment allocation. Randomization was performed under the supervision of the unblinded neurologist who also had the responsibility of managing patients’ treatment and workup during hospitalization. The unblinded internist had the responsibility to monitor probable drug adverse effects and exert appropriate management in necessary cases. The inclusion and exclusion criteria are listed below.
Inclusion Criteria 1. Adults aged between 50 and 80 years old 2. First ever ischemic stroke in the territory of middle cerebral artery and including classes 1, 3, 4, and 5 subtypes of trial of Org 10172 in acute stroke treatment (TOAST) classification system9 (ie, all subtypes of ischemic stroke except embolic stroke with cardiac origin) with a compatible brain magnetic resonance imaging or computed tomography scan 3. Appropriate level of consciousness, defined as score 1 or less from each of 1b and 1c section of the modified National Institutes of Health Stroke Scale (mNIHSS)10 at the time of entrance in the study 4. Motor power of grade 0/5 to 3/5 on the Medical Research Council scale in at least one upper limb at the time of entrance in the study 5. Maximum time passed of less than 48 hours from stroke onset 6. Nonmenopause women should not be pregnant or lactating mothers 7. Written informed consent for participation in study (obtained from patient or his/her legal representative)
Exclusion Criteria 1. Coexistence of intracerebral hemorrhage (subarachnoid or intraparenchymal) or hemorrhagic conversion during study period. 2. Severe sustained loss of consciousness (score >2 from 1b and 1c questions of mNIHSS during study period 3. Wernicke aphasia (pure or as a part of constellation of other symptoms) 4. Organ failures that may interfere with consumption of lithium, including: Acute or chronic kidney disease (serum creatinine >1.2) Acute or chronic heart failure (ejection fraction 16). For evaluating improvement of impairment after stroke, our primary efficacy end points were changes in patients’ scores of mNIHSS and subsection of hand assessment of Fugl-Meyer assessment11 (hFMA) including subsections 8a, 8b, 8c, 8d, 8e, 8f, and 8f . The secondary end point was regaining more than 25% of the full motor function of hand movements based on hFMA. According to hFMA, a patient with full hand motor function will gain a total score of 14; therefore, regaining more than 3.5 points in hFMA after treatment was regarded as secondary end point achievement. For example, if one patient’s hFMA on the first assessment was 4 and after 30 days he/she achieved the score of 8 on hFMA, recovery percentage was calculated as [(8–4)/14]100=28%, and it was categorized in the group with regaining more than 25% of the full function. Fugl-Meyer assessment and NIHSS were used as our assessing tools mainly because these are impairment scores rather than being disability scores like Barthel index or Rankin scale. The advancement of an impairment score is that it can assess the degree of recovery or improvement of function rather than measure the amount of compensation. Compensation plays a major role in decreasing the level of disability, like the time that a right-handed patient with right arm paresis learn to conduct daily living activities with the left hand. In such instances, disability in practice has been decreased, but it cannot be regarded as true improvement. The first aim of this study was to find whether a low to mid dose of lithium may have any probable effect on actual improvement of neurological function, if being used in acute state of stroke. As lithium reaches to its effective serum level after 5 days of consumption, we compared impairment scores at the fifth day and then at the 30th day of entrance in the study.
Other Tests We checked all of the following tests at baseline for all subjects. Electrocardiogram; chest x-ray; echocardiography; © 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Clinical Neuropharmacology • Volume 37, Number 3, May/June 2014
and routine blood tests involving complete blood count, renal function test, thyroid function test, lipid profile, glucose, and electrolyte. At the fifth and the 30th days of entrance in the study, routine blood tests were checked again for all participants Serum lithium level was checked from the blood samples taken at the fifth and 30th days after initiation of study for patients in group A. Results of routine blood tests and lithium level were evaluated by the unblinded internist who was responsible for checking and managing adverse events. Our aim was to achieve a serum lithium level of 0.4 to 0.8 mmol/L by the fifth day of treatment. In the cases where serum lithium level at the fifth day was between 0.8 and 1.2 mmol/L and serum creatinine was less than 1.2 mg/dL, lithium carbonate dosage was decreased to 300 mg daily. For each case where decreasing the lithium dosage was done, dosage of placebo was decreased in one randomly selected patient from group B accordingly. The study was approved by the medical ethics committee of Ahvaz Jundishapur University of Medical Sciences (registration number: ETH-403), and it was registered in Iranian Registry of Clinical Trials (IRCT), available at www.irct.ir under registration number: IRCT2012072210361N1.
Sample Size This study was designed basically as a pilot study. Previously, Duncan et al12 in a cohort study of improvement after stroke by using FMA had estimated that achieving 50% improvement after 6 months can be an acceptable improvement for determining the effect size. In addition, their study showed that near half of spontaneous improvement will take place at the first month after stroke. Therefore, we estimated that sample size should be 30 patients in each arm of study based on a priori power analysis13 to detect an effect size d of 27 points (23% vs 50%) with α=0.05 (1-tailed) and 80% power. Nevertheless, with the prediction of
Lithium's Effect in Post-Stroke Motor Recovery
loss due to morbidity or mortality due to ischemic stroke or other related conditions, it was decided to recruit at least 40 patients in each arm of the study.
Statistical Analysis Kolmogorov–Smirnov Z was used to check whether variables are normally distributed. Baseline variables were compared using a 2-group t test and the Mann-Whitney U test for continuous variables, and the χ2 test and the Fisher exact test for categorical variables. For evaluating efficacy variables, we made comparison between 2 groups at baseline and the end of treatment. Since the change from baseline was the primary outcome, a per-protocol analysis was done. All analyses were done using the Statistical Package for Social Sciences version 16, and P
