Safety of Dexamphetamine in Acute Ischemic Stroke
A Randomized, Double-Blind, Controlled Dose-Escalation Trial
Background and Purpose— Amphetamine is reported to enhance recovery after experimental stroke, but results from the first few trials in humans are inconclusive. Limited information is available on treatment safety. This study intended to investigate the safety and tolerability of dexamphetamine in patients with acute cerebral ischemia.
Methods— Forty-five patients with cerebral ischemia were enrolled within 72 hours after onset of symptoms. Patients were randomized to 1 of 3 dose levels (2.5, 5, or 10 mg orally twice daily) or placebo for 5 consecutive days. Adverse events, blood pressure, heart rate, body temperature, consciousness level, and functional outcome measures were followed up daily during treatment. Follow-ups were made at day 7 and 1 and 3 months after stroke.
Results— Mean systolic and diastolic blood pressures and heart rate increased 14 mm Hg, 8 mm Hg, and 9 bpm, respectively, with dexamphetamine treatment compared with placebo (P≤0.01). There was no difference between dexamphetamine and placebo regarding adverse events, body temperature, or consciousness level. During treatment, there was a benefit of dexamphetamine in neurological and functional outcome (P<0.05), but differences were not maintained at follow-up.
Conclusions— Overall, dexamphetamine was safe and well tolerated by patients with acute cerebral ischemia, but blood pressure and heart rate increased during treatment in comparison to placebo. These observations may be important in the future planning of amphetamine trials because changes in these variables have been observed to interfere with clinical outcome. The suggestions of improvement in outcome must be confirmed in further studies.
The effect of amphetamine on promoting recovery after focal cortical lesions has been extensively studied in animal models. Most evidence shows an acceleration of recovery with amphetamine treatment, leading to a final higher functional level than without this therapy1 if the treatment is combined with concomitant training.1,2 The acceleration of recovery has been found to correspond to enhanced neural sprouting and synaptogenesis after experimental infarction.3 In contrast, a detrimental effect of amphetamine on recovery has been reported after lesions in the cerebellum4 or substantia nigra.5
An increase in central noradrenergic activity seems to underlie the effect of amphetamine on recovery in experimental models6 (see References 1 and 2 for a review). Boyeson and coworkers7,8 have shown that infusion of norepinephrine but not dopamine into the brain facilitates motor recovery in animals after brain injury. The noradrenergic theory is further supported by the observation that drugs acting as antagonists to norepinephrine have reinstated motor deficits in animal models.9
So far, few clinical studies have been published. Results from the first relatively small studies (8 to 42 patients) are inconclusive, with some studies reporting effects in line with the animal models10–12 and others reporting no such effects.13–15 From human studies, no adverse events or side effects have been reported besides insomnia.16 To the best of our knowledge, no published study has been designed specifically to evaluate adverse effects and tolerability of amphetamine in patients with stroke. We therefore intended to investigate the safety and tolerability of dexamphetamine (d-amph) at 3 different doses compared with placebo in patients with acute cerebral ischemia. A secondary aim was to study the effects of d-amph on neurological and functional recovery compared with placebo.
Subject and Methods
The study was performed at the Department of Neurology, Karolinska Hospital (Stockholm, Sweden). Patient recruitment and follow-up were done by the authors. A safety committee was responsible for unblinded examination of the adverse events reports. The study was performed in accordance with the Declaration of Helsinki, and the protocol was approved by the local ethics committee at Karolinska Hospital and by the Medical Products Agency in Sweden. All patients or their relatives gave informed consent.
Patients with first or recurrent stroke admitted to the stroke unit at the Department of Neurology, Karolinska Hospital, were screened for inclusion in the study from October 1998 to January 2001. Patients fulfilling the following criteria were included: (1) age, 18 to 85 years; (2) clinical diagnosis of acute hemispheric cerebral infarction within the carotid-supplying area confirmed by excluding hemorrhagic stroke on a CT scan; (3) onset of symptoms ≤72 hours before the start of treatment; (4) score of ≤15 on arm, hand, and leg motor items of the Scandinavian Stroke Scale (SSS)17 or a presumed need for hospital care at the stroke unit for at least 5 days because of other focal neurological symptoms such as aphasia or neglect; (5) symptoms present for ≥1 hour and still present at the start of treatment; and (6) informed consent. The diagnosis of stroke was based on the medical history and clinical examination of the patient according to World Health Organization criteria.18 Main exclusion criteria were other serious diseases with short life expectancy or that could interfere with the study protocol and known ongoing alcohol or drug abuse. There were no limits set to blood pressure (BP).
Design and Randomization
A prospective randomized, placebo-controlled, double-blind approach was used. The study was designed as a dose-escalation trial with 3 groups of 15 patients (10 d-amph and 5 placebo) in each group. If unacceptable side effects arose in >4 patients in a group of 15, this level would be stopped for an analysis. If all patients with unacceptable side effects had received d-amph, this dose level would have been closed and a new one opened at a lower intermediate dose level. The basis for the dimensioning of sample size was a judgment on clinical relevance of the side effects.
The Karolinska Pharmacy prepared and numbered 45 matching boxes with d-amph capsules of different doses (2.5, 5, and 10 mg) and placebo randomly. Patients received boxes in consecutive order. The pharmacy kept the codes until the end of the trial.
Patients received either d-amph or placebo in capsules given orally twice daily for 5 consecutive days (the treatment period). The first capsule was given within 72 hours after stroke on the day after baseline assessments (day 1) at approximately 8 am; the second was given 4 hours later. Treatment was carried out in a double-blind manner, with d-amph or placebo in identical packages.
Demographic details, medical and cerebrovascular history, general characteristics of the present stroke, ECG, and laboratory tests were collected at baseline. Adverse events were recorded daily during the treatment period and 2 days after drug discontinuation. After day 7, only serious adverse events (stroke, progressive stroke, transient ischemic attack, myocardial infarction, psychic disturbances, and epilepsy) and death were registered. Progressive stroke was defined as a loss of ≥2 points on the SSS17 and fever as a temperature of ≥37.5°C. Systolic and diastolic BPs and heart rate was recorded hourly from 8 am until 4 pm during the first, third, and fifth days of treatment and at 8 am, noon, and 4 pm the second and fourth days of treatment. Registrations were made in the supine or seated position after 5 minutes of rest with a Datascope Accutorr 3 (Datascope Corporation). The same arm was used in the same patient at all registrations. Body temperature was recorded in the ear with a ThermoScan pro 1 (Thermoscan Inc) at 8 am, noon, and 4 pm during the treatment period. Level of consciousness was recorded with the Reaction Level Scale (RLS 85)19 hourly between 8 am and 4 pm during the treatment period. Patients with slight disorientation but who were fully awake were accepted in reaction grade 1, because our assessment focused on the level of consciousness rather than on cognitive functions. We used a category of 0 for normal sleep. The quantity of physiotherapy, occupational therapy, and speech therapy was recorded on a special form daily during the treatment period until the first follow-up (day 7) 2 days after discontinuation of treatment. At baseline, patients were assessed for the following measures: (1) the SSS-5817 and SSS-68, which is an internal but unvalidated modification of the scale in which neglect has been added with a maximum of 10 points; (2) Lindmark motor assessment chart (LMAC)20,21; (3) Activity Index (AI)22; (4) Barthel ADL Index (BAI)23; and (5) 10-m self-paced walk24 and a walk at maximum walking speed (patients were asked to walk at a maximum safe speed). Patients were reassessed with LMAC and SSS at days 1, 2, 3, and 5 of treatment and with AI, BAI, and the 10-m walking test at day 3. Follow-ups were made at day 7 and at 1 and 3 months.
All patients who were alive, including treatment withdrawals, had follow-up visits at day 7 and after 1 and 3 months or until death.
Descriptive statistics were used to estimate demographic parameters and baseline scores. The number of adverse events was compared between groups with the χ2 test. Repeated-measures analysis was used to analyze time-dependent data. Statistical significance for intergroup differences in outcome was assessed by the Mann-Whitney U test, χ2, Kruskal-Wallis, and Student’s t test, as appropriate. The analyses of LMAC, SSS, AI, and BAI were based on linear transformation, which gives a relative change in score. The transformed score was calculated according to a previously used formula25 and ranged from −100 (maximal worsening) to 100 (maximal improvement). The analyses of body temperature, BP, heart rate, and 10-m walking were based on differences from each assessment point compared with placebo. The study used multiple testing in which each hypothesis (intergroup differences) was analyzed separately and the existence of patterns and the consistency of the results were considered in the analysis. All analyses were carried out by use of Statistica 6.0 (Statsoft, Inc).
Safety analysis was based on all patients randomized. Efficacy analysis was based on surviving d-amph and placebo patients who completed the 5-day treatment protocol.
Recruitment, Patient Baseline Characteristics, and Amount of Rehabilitation
Forty-eight patients fulfilled the inclusion criteria and were asked to participate in the trial; of these, 45 gave informed consent for participation and were randomized to the trial. All patients were followed up for 3 months or until death according to the principles of intention to treat and formed the basis for the safety analysis. Of the 45 patients included in the study, 41 were valid for efficacy analysis at day 1, 40 at day 2, 39 at day 3, 38 at days 5 and 7, 33 at 1 month, and 32 at 3 months. Four patients were excluded because they did not complete the treatment program for the following reasons: epileptic seizure (placebo), consent withdrawn (placebo), icterus (d-amph), and hallucination (d-amph). The other patients were excluded because of death. Treatment initiation did not cause any identifiable effects of d-amph, which could cause unblinding of the study.
Demographic and medical history are summarized in Table 1; baseline characteristics are given in Table 2. There was a statistically significant difference between the whole group of d-amph–treated patients (all d-amph, n=30) and placebo-treated patients (n=15) in number of smokers, with the number of smokers being higher in the all d-amph group (P=0.042). The number of patients not alert was statistically higher in the all d-amph group compared with the placebo group (P=0.042).
There were no statistically significant differences between treatment groups in the amount of rehabilitation during or after the treatment period.
Adverse Events and Mortality
All adverse events during the treatment and follow-up periods are listed in Table 3. The mortality rate in the all d-amph group was 20% compared with 13.3% in the placebo group (P=NS). The primary causes of death in the low-dose d-amph group were pneumonia (n=3) and coronary heart disease (n=1); in the median-dose d-amph group, pneumonia (n=1), coronary heart disease (n=1), and pleurocarcinosis (n=1); and in the high-dose d-amph group, coronary heart disease (n=1) and initial stroke (n=1). In the placebo group, 1 patient died because of pneumonia and 1 because of coronary heart disease. Coronary deaths occurred 12, 20, and 27 days after drug discontinuation. Deaths were clinically classified as not being related to the study drug in 9 of the cases and as probably not related in 2 cases (1 d-amph and 1 placebo) according to the safety committee. Severe adverse events occurred in 66.7% of the patients in the all d-amph group and in 66.7% of the placebo patients. The mean number of total adverse events per patient was 2.0 in the all d-amph group and 1.8 for placebo (P=NS).
Differences between the all d-amph and the placebo groups were not statistically significant.
The systolic BPs for the all d-amph groups and placebo at each time point are shown in Figure 1A; the mean systolic BPs for the whole treatment period compared with baseline are given in Figure 2A. Differences between treatment arms were statistically significant (analysis of variance [ANOVA], P=0.023), with the d-amph–treated groups having the highest systolic BP. The mean systolic BP for the whole treatment period increased 14 mm Hg in the all d-amph group (P=0.004), 16 mm Hg in the median-dose d-amph group (P=0.008), and 16 mm Hg in the high-dose d-amph group (P=0.005) compared with placebo (P<0.05, r=0.44).
The diastolic BPs for the all d-amph group and placebo for each time point are shown in Figure 1B, and the mean diastolic BPs for the whole treatment period compared with baseline are given in Figure 2B. Differences between the 3 dose groups and placebo were significant (ANOVA, P=0.024). The mean diastolic BP for the whole treatment period increased 8 mm Hg in the all d-amph group (P=0.010), 9 mm Hg in the median-dose d-amph group (P=0.009), and 10 mm Hg in the high-dose d-amph group (P=0.005) compared with placebo (P<0.05, r=0.43).
The heart rates for the all d-amph and placebo groups at each time point are shown in Figure 1C; the mean heart rates for the whole treatment period compared with baseline are given in Figure 2C. Differences between the 3 dose groups and placebo were significant (P=0.0001), with the d-amph–treated patients having the highest heart rate (P<0.05, r=0.61). The mean heart rate during the whole period increased 9 bpm in the all d-amph group (P=0.007), 8 bpm in the median-dose group (P=0.023), and 17 bpm in the high-dose group (P=0.00008) compared with placebo and 15 bpm in the high-dose d-amph group compared with the low-dose group (P=0.001).
Level of Consciousness
During treatment, there were no statistically significant differences between treatment groups in level of consciousness.
The all d-amph–treated group had significantly better improvement (P<0.05) during the treatment period assessed with the LMAC motor function score (days 1, 2, 3, and 5; see Figure 3), SSS-58 (days 1 and 5), SSS-68 (day 5), AI motor score (day 3), and BAI (day 3). At the day 7 follow-up, all d-amph group had significantly better improvement with the LMAC motor function score (P=0.025), SSS-68 (P=0.044), and AI motor score (P=0.009). There were no statistically significant differences at 1 or 3 months. No other outcome measure, including the 10-m walking test, was statistically significant at any time point.
There were no statistically significant increases in total number of severe adverse events, deaths, body temperature, or level of consciousness. Systolic and diastolic BPs and heart rate increased during treatment with d-amph compared with placebo. The increase seemed to be dose dependent, most obviously for heart rate. During treatment, there were statistically significant differences between the groups in functional and neurological outcome in favor of d-amph, but these differences were not maintained at follow-up.
Differences in mortality between the all d-amph and placebo groups were not statistically significant, but there were more deaths in the all d-amph group. The all d-amph group had a higher extent of cardiac failure, atrial fibrillation, and neglect at baseline compared with the placebo group, as well as having a higher number of smokers. Some of these factors may predict a worse outcome. One patient in the all d-amph group who died during the follow-up period was diagnosed with pleurocarcinosis after inclusion in the study and would not have been included if this condition had been known. In the judgment of the safety committee, none of the deaths were related to the study drug. There was no increase in the number of deaths (or serious adverse events) with increasing dose.
The increases in BP and heart rate are in agreement with the sympathomimetic properties of amphetamines, but they have not previously been described in studies providing information on treatment safety in stroke patients.10,11,15,27 Our study differs from earlier studies in a number of ways. Walker-Batson et al11,12 started treatment 16 to 45 days after stroke onset, administered 10 mg d-amph at an interval of 3 to 4 days, and excluded patients with uncontrolled hypertension (>160/110 mm Hg), whereas we started d-amph treatment within 72 hours, administered 5 to 20 mg d-amph twice daily, and included patients with hypertension. It seems likely that the effect on BP and heart rate might be more pronounced with more frequent administration. In this study, BP and heart rate were monitored hourly 3 of 5 days during the treatment period; Walker-Batson et al reported a baseline measurement followed by a measure 2 hours after drug administration. From our own experience, the effect on BP and heart rate differs substantially between patients and does not always appear within the first hours after the first dose administration. Therefore, late effects might be missed if the patient is not followed for a longer period (Figure 1A through C).
The importance of combining d-amph administration with training has been emphasized in the context of improving recovery.2 It has also been argued that a “continuous [daily] administration of d-amph could have a diminishing effect on therapeutic efficacy”11 because daily dosing theoretically could lead to neurotransmitter depletion.1 In this study, we did not systematically correlate d-amph treatment and rehabilitation in time, but all patients except 1 had training (physiotherapy, occupational therapy, speech therapy) during the treatment period. It seems likely that the patients had their training sessions within the time of potentially effective plasma concentration because the drug was administered twice daily. The effect of a more strict combination of d-amph and rehabilitation and a more spaced d-amph administration remains to be investigated.
Overall, d-amph seemed to be safe and well tolerated by patients with acute cerebral ischemia during the first week after symptom onset. Because at least changes in BP28 and body temperature29 are related in different ways to stroke outcome and because BP and heart rate seem to be related to amphetamine dose, monitoring these parameters should be relevant in future amphetamine studies. Whether the effect of BP may contribute to a better outcome with d-amph should be studied further. Efficacy results should be interpreted with caution because of the limited sample size and short treatment period in the subacute phase when the neurological condition frequently is unstable. However, a benefit of d-amph was suggested for patients who followed the study protocol. This needs to be confirmed in future trials.
This trial was supported by grants for the Federation of County Councils (main sponsor), Family Janne Elgqvist foundation, Swedish Association of Neurologically Disabled, Åke Wiberg Foundation, Karolinska Institute, and Swedish Stroke Association. We gratefully acknowledge Anita Hansson-Thyrén and Lena Lundqvist for assistance with data collection and Dr Hans-Göran Hårdemark (chairman), Professor Jesper Swedenborg, and Associate Professor Jan Östergren for participating in the safety monitoring committee. Per Näsman is acknowledged for statistical advice.
- Received June 13, 2002.
- Revision received August 15, 2002.
- Accepted August 27, 2002.
Feeney DM, Gonzalez A, Law WA. Amphetamine, haloperidol, and experience interact to affect rate of recovery after motor cortex injury. Science. 1982; 217: 855–857.
Stroemer RP, Kent TA, Hulsebosch CE. Enhanced neocortical neural sprouting, synaptogenesis, and behavioral recovery with d-amphetamine therapy after neocortical infarction in rats. Stroke. 1998; 29: 2381–2395.
Feeney D. Mechanisms of noradrenergic modulation of physical therapy: effects on functional recovery after cortical injury. In: Goldstein LB, ed. Restorative Neurology: Advances in Pharmacotherapy for Recovery After Stroke. New York, NY: Futura Publishing Co Inc; 1998: 35–78.
Walker-Batson D, Smith P, Curtis S, Unwin H, Greenlee R. Amphetamine paired with physical therapy accelerates motor recovery after stroke: further evidence. Stroke. 1995; 26: 2254–2259.
Walker-Batson D, Curtis S, Natarajan R, et al. A double-blind, placebo-controlled study of the use of amphetamine in the treatment of aphasia. Stroke. 2001; 32: 2093–2098.
Mazagri R, Shuaib A, McPherson M, Deighton M. Amphetamine failed to improve motor function in acute stroke. J Can Sci Neurol. 1995; 22: S25.Abstract.
Reding M, Solomon B, Borucki S. Effect of dextroamphetamine on motor recovery after stroke. Neurology. 1995; 45 (suppl 4): S01.Abstract.
Scandinavian stroke study group. Multicenter trial of hemodilution in ischemic stroke: background and study protocol. Stroke. 1985; 16: 855–890.
Recommendations on stroke prevention, diagnosis, and therapy. Stroke. 1989; 20: 1407–1431.
Starmark J-E, Stålhammar D, Holmgren E. The Reaction Level Scale (RLS 85): manual and guidelines. Acta Neurochir (Wien). 1988; 91: 12–20.
Mahoney F, Barthel B. Functional evaluation: the Barthel Index. Md Med J. 1965; 14: 61–65.
Ahmed N, Näsman P, Wahlgren N. Effect of intravenous nimodipine on blood pressure and outcome after acute stroke. Stroke. 2000; 31: 1250–1255.