An Exploration of Central Nervous System Medication Use and Outcomes in Stroke Rehabilitation

Article Outline

• Abstract
• Methods
  • Patient Variables
  • Process Variables (Including Medications)
  • Outcome Variables
  • Patient Subsample With Neurobehavioral Impairment
  • Statistical Methods
• Results
  • Descriptive Statistics
  • Associations of Neurologic and Behavioral Impairments With Outcomes by CMGs
• Discussion
• Conclusions
• Acknowledgments
• References
• Copyright

Abstract 

Conroy B, Zorowitz R, Horn SD, Ryser DK, Teraoka J, Smout RJ. An exploration of central nervous system medication use and outcomes in stroke rehabilitation.

Objective

To study associations between neurobehavioral impairments, use of neurotropic medications, and outcomes for inpatient stroke rehabilitation, controlling for a variety of confounding variables.

Design

Observational cohort study of poststroke rehabilitation.

Setting

Six inpatient rehabilitation hospitals in the United States.

Participants

Patients with moderate or severe strokes (N=919).

Interventions

Not applicable.

Main Outcome Measures

Discharge disposition, FIM score change, and rehabilitation length of stay (LOS).

Results

Neurobehavioral impairments and use of many medications, including first-generation selective serotonin reuptake inhibitors, older traditional antipsychotic medications, and anti-Parkinsonian neurostimulants, have a statistical association with poorer outcomes, whereas use of the atypical antipsychotic medications has a positive association with improvement in motor FIM scores. Counterintuitively, use of opioid analgesics is associated with a larger motor FIM score change but not an increase in LOS or reduced percentage of discharge to community. There was significant variation in use of neurotropic medications among the 6 study sites during inpatient stroke rehabilitation.

Conclusions

There are many opportunities to enhance a stroke survivor’s ability to benefit from acute inpatient stroke rehabilitation through improved understanding of associations of neurotropic medications with outcomes for different patient groups.

Key Words:  Antipsychotic agents , Clinical practice variations , Rehabilitation , Stroke , Treatment outcome

ANNUAL MEDICARE EXPENDITURES for hospital-based rehabilitation in the United States reached $5.9 billion in 2004.¹, ² Stroke, a leading cause of adult onset disability, is the second leading cause for admission to inpatient rehabilitation and is associated with high costs and intensive utilization of rehabilitation resources.³, ⁴, ⁵ Neurologic and behavioral impairments, such as delirium, dementia, agitation, anxiety, apathy, psychomotor slowing, impulsivity, and depression, are common in stroke survivors and can have a negative association with participation in therapy, length of stay (LOS), discharge disposition, resultant functional outcome, and ultimate quality of life.⁶, ⁷, ⁸, ⁹, ¹⁰, ¹¹, ¹², ¹³, ¹⁴, ¹⁵

Stroke-related depression literature⁶, ⁷, ⁸, ⁹, ¹⁰, ¹¹, ¹², ¹³, ¹⁴, ¹⁵ states that depression is probably the most common neurologic and behavioral impairment disorder after stroke, that it occurs in 30% to 50% of stroke patients,⁸ and that depressive symptoms and pharmacologic treatments extend well beyond the first few weeks after stroke.⁷, ⁸, ⁹, ¹⁰, ¹¹, ¹², ¹³, ¹⁴, ¹⁵ Importantly, there are studies that have found that major depression may not become diagnosable until several months after stroke onset.¹⁶ In contrast, during the immediate poststroke period—when patients are most likely to undergo intensive rehabilitation therapy—other mood and behavior disturbances are more prevalent than a major depressive disorder, but few studies exist on this subject.⁶ Examples of neurologic and behavioral impairments that can occur soon after the onset of a stroke and can interfere with rehabilitation care include apathy, agitation, anxiety, insomnia, psychosis, disinhibition, adjustment disorder with depressed mood, delusions, delirium, abulia, pathologic affect, psychomotor slowing, neurogenic and somatic pain, mania, catastrophic reactions, and poststroke fatigue. Several pharmacologic classes of medications (eg, benzodiazepines, antipsychotics, sedatives and hypnotics, anticonvulsants, stimulants, antidepressants) often are used empirically, alone or in combinations, to treat these symptoms. Some of these medications, such as benzodiazepines, the anticonvulsants phenytoin and phenobarbital, and older dopamine receptor antagonists have been associated with poorer upper-extremity motor function and less independence in activities of daily living 84 days poststroke.¹⁷

According to current literature, the potential benefit of choosing 1 neurotropic medication over another in poststroke mood and behavior disturbances other than depression is particularly unclear, especially in the early poststroke interval (0–4wk after stroke). Do newer neurotropic medications (usually more costly) offer substantial benefits compared with the older, less expensive, and more commonly used medications? Limited access to newer agents because of formulary cost control, as well as a limited number of studies in stroke patients, has impeded the adoption of these medications in clinical practice, thus hindering clinical knowledge of potential benefits.¹⁸ Judicious study of selected neurotropic medications, such as olanzapine or quetiapine in poststroke patients with agitation or delirium as opposed to buspirone, benzodiazepines, or haloperidol, has potential to affect outcomes.

Many reasons exist for the paucity of information on effects of neurotropic medications in stroke rehabilitation. There is a concern whether randomized control methods for this type of study are ethically and logistically appropriate in this population. Cognitive and emotional aberrations often affect recruitment into clinical trials because of lack of understanding or altered mental status; randomized controlled trials often exclude these types of impairments. Henon et al¹⁹ found evidence of preexisting dementia in 16% of a series of admissions to their stroke unit. Interactions between the mechanism and anatomic location of the brain lesion in relation to the timing of drug administration, which are not understood completely, may influence a drug’s apparent impact on functional recovery.²⁰

Analysis of the Post-Stroke Rehabilitation Outcomes Project (PSROP) database uncovered significant variation in the use of medications among 6 U.S. inpatient rehabilitation facilities (IRFs) that cannot be explained by patient differences.²¹ This was especially evident in those agents specifically used for their effects on the central nervous system. Physician preferences seemed to be primary determinants of medication choice. Drug formulary restrictions, experience using a particular medication, and other factors may influence physicians’ prescriptions.

This study attempts to identify neurotropic medication treatments associated with better outcomes with regard to mood, behavioral, and/or cognitive impairments in stroke rehabilitation. We hypothesize that use of medications that modulate the noradrenergic, dopaminergic, cholinergic, and serotinergic neuroendocrine systems is associated with better outcomes after stroke rehabilitation. A secondary hypothesis is that newer neuroleptic medications are associated with better outcomes compared with older neuroleptic agents. Newer antipsychotic agents purportedly have mechanisms of action that are more effective than older antipsychotics and have a lesser side-effect profile; thus they are better tolerated in patients with stroke and the elderly at risk of iatrogenic disturbances.

Back to Article Outline

Methods 

The methodology governing the full PSROP is discussed in the article by Gassaway et al,²¹ which provides a detailed description of the larger study’s participating facilities, patient selection criteria, data collection instruments including their validity and reliability, and a detailed description of the project’s final study group. The methodology is summarized in Maulden et al.²² The institutional review boards at Boston University and at each participating IRF approved the study.

Patient Variables 

PSROP patient variables²¹ included age, sex, race, payer, type and location of stroke, admission FIM instrument score, case-mix group (CMG), time from stroke symptom onset to rehabilitation admission, and severity of illness. The Comprehensive Severity Index (CSI), the study’s principal severity-of-illness adjuster, is a disease-specific severity assessment system that provides a consistent method for defining levels of severity using over 2200 individual physical findings and laboratory results.²³, ²⁴, ²⁵, ²⁶, ²⁷ The CSI was measured separately for admission to rehabilitation (first 24h), discharge from rehabilitation (discharge day), and maximum (the full rehabilitation stay, including admission and discharge periods).

Process Variables (Including Medications) 

Details about each neurotropic medication received by study patients were obtained, including drug name, dose, frequency (including as required [PRN] or regular dosing), route of administration, start date and time, and stop date and time. For medications to be included in the PSROP database, the medication needed to be initialized as given on the medication administration record in the patient’s chart. PRN medications that were ordered but not given were not included.

We grouped neurotropic medications into categories by consensus of prescribing members of the PSROP clinical team based on similarity of drug content and effects on patients. Drug categories (structured roughly around medication groupings found in ePocrates²⁸) used in these analyses are listed in table 1 along with the medications they contain.

Table 1. Descriptions of Medication Categories

Abbreviation: SSRIs, selective serotonin reuptake inhibitors.

Outcome Variables 

PSROP outcome variables include rehabilitation LOS, discharge FIM and CSI scores, functional gain as measured by increased FIM score from admission to discharge, increase in severity of illness as measured by increase in CSI from admission to maximum, and discharge disposition.²¹

Patient Subsample With Neurobehavioral Impairment 

In the 1161-subject U.S. PSROP sample, we identified patients with indications of neurobehavioral impairment, defined as mood and behavioral disturbances, cognitive impairment, both, or symptoms of neither but presence of certain neurotropic medications indicative of previously treated symptoms. Patients were included in the neurobehavioral impairment group if they met 1 of 3 selection criteria, each of which is analyzed as an independent variable:

CMGs were combined into moderate (CMGs 104–107) and severe (CMGs 108–114) stroke patient groups, which were large enough to detect small effects. There were too few patients with mild stroke to be analyzed at this time (CMGs 101–103; n=108).

To include the full inpatient rehabilitation course in these analyses, patients discharged to other acute facilities were excluded (n=134). This left 474 patients in the moderate stroke group and 445 patients in the severe stroke group to allow us to evaluate effectiveness of various medication approaches, including polypharmaceutical combination therapies found to be of benefit in a recent study of long-term-care patients.¹⁸

Statistical Methods 

We performed a systematic analysis to examine associations of various neurobehavioral impairments and neurotropic medication categories with stroke rehabilitation outcomes using descriptive statistics, 2-way associations, analysis of variance, correlation analyses, and ordinary least squares or logistic regression analyses. We controlled for important covariates, such as admission functional status (FIM instrument), severity of illness (CSI), and comorbidities, by using detailed patient data contained in the PSROP database.²¹

Back to Article Outline

Results 

Descriptive Statistics 

Table 1 lists specific medications that were included in each neurotropic medication group. Each group contains multiple medications used in PSROP facilities; however, there is often a predominate medication. For example, gabapentin accounts for 90% of all new anticonvulsants.

Patients with moderate stroke (CMGs 104–107) had different demographic and other characteristics than patients in CMGs 108 to 114 (severe stroke). There were 345 (72.8%) patients in the moderate stroke group who had a documented neurobehavioral impairment or received neurotropic medications, compared with 381 (85.6%) patients in the severe stroke group (P<.001) (table 2).

Table 2. Descriptive Statistics for Patients in Moderate (CMGs 104–107) and Severe (CMGs 108–114) Stroke Groups

Variables	CMGs 104–107 (n=474)	CMGs 108–114 (n=445)	P
Female (%)	50.0	47.6	.509⁎
Mean age ± SD	65.4±14.8	67.8±14.1	.013†
Age groups (%)			.224⁎
19–40y	5.5	3.8
41–60y	26.4	23.2
61–80y	52.1	53.0
>80y	16.0	20.0
Race (%)			.060⁎
White	56.5	56.4
Black	28.1	24.5
Other	14.4	18.4
Missing	1.1	0.7
Side of stroke (%)			.727⁎
Right	45.4	43.4
Left	42.8	43.4
Bilateral	9.1	11.0
Unknown	2.7	2.3
Type of stroke (%)			.063⁎
Hemorrhagic	21.3	26.7
Ischemic	78.7	73.3
Neurobehavioral impairments (%)			<.001⁎
Mood/behavior disturbances	36.5	36.2
Cognitive impairment	4.4	7.6
Both	8.9	18.4
Neurotropic medications	23.0	23.4
None	27.2	14.4
Discharge disposition (%)			<.001⁎
Home/community	95.2	73.3
SNF	4.8	28.7
Mean admission CSI continuous score ± SD	16.0±10.2	26.7±14.7	<.001†
Mean maximum CSI continuous score ± SD	23.3±14.2	40.1±21.9	<.001†
Mean discharge CSI continuous score ± SD	6.0±6.8	14.1±12.9	<.001†
Mean increase (maximum − admission) in CSI score ± SD	7.3±8.0	13.4±12.2	<.001†
Mean admission motor FIM score ± SD	47.9±5.6	27.0±7.1	<.001†
Mean discharge motor FIM score ± SD	70.2±9.4	51.5±16.3	<.001†
Mean Increase motor FIM score ± SD	22.4±8.8	24.5±13.9	.006†
Mean admission cognitive FIM score ± SD	24.2±7.2	16.9±7.7	<.001†
Mean discharge cognitive FIM score ± SD	27.9±6.0	22.2±7.6	<.001†
Mean increase cognitive FIM score ± SD	3.7±3.7	5.3±4.6	<.001†
Mean LOS ± SD	15.2±7.2	24.9±10.5	<.001†

Abbreviations: SNF, skilled nursing facility; SD, standard deviation.

There were several other significant differences between the moderate and severe stroke CMG patient groups. Patients with severe stroke were sicker as measured by admission and maximum CSI scores (higher), more functionally disabled as measured by FIM scores (lower), had higher percentage of hemorrhagic (vs ischemic) strokes, and had longer LOSs. A smaller percentage of patients with severe stroke were discharged to home (73.3% vs 95.2%).

Neurotropic medication use served as a surrogate for indication of neurobehavioral impairment for 23% of patients in both groups.

Associations of Neurologic and Behavioral Impairments With Outcomes by CMGs 

Associations between neurobehavioral impairment and outcomes by severity of stroke are shown in table 3. For patients with moderate strokes, having both of the defined components of neurobehavioral impairment (mood and behavior disturbances, cognitive impairment) was associated with the longest LOS (17.3d). Patients with no documentation (diagnosis, chart descriptions, or neurotropic medication use) of neurobehavioral impairment had the shortest LOS (12.9d, P<.001) and the highest rate of discharge to home (99.2%).

Table 3. Bivariate Associations of Neurobehavioral Impairment With Outcomes by CMG Group

Variables	Mood/Behavior Disturbances (n=334)	Cognitive Impairment (n=55)	Both Mood/Behavior Disturbances and Cognitive Impairment (n=124)	Neurotropic Medications (No Mood/Behavioral/Cognitive Impairment Signs Recorded) (n=213)	None (n=193)	P
LOS (d)
CMGs 104–107 (n)	173	21	42	109	129
Mean LOS ± SD	16.2±6.5	15.2±6.8	17.3±9.1	15.4±7.9	12.9±6.3	<.001⁎
CMGS 108–114 (n)	161	34	82	104	64
Mean LOS ± SD	26.4±9.9	22.6±9.4	26.0±12.3	24.2±10.7	22.1±9.4	.030⁎
Mean increase motor FIM score ± SD
CMGs 104–107	22.5±8.5	22.1±10.2	20.5±7.1	21.5±10.1	23.5±8.1	.271⁎
CMGs 108–114	25.4±13.2	21.8±22.4	20.6±14.3	26.4±11.5	25.4±12.6	.034⁎
Discharge disposition (%)
CMGs 104–107						<.001†
Home	97.1	81.0	85.7	93.6	99.2
SNF	2.9	19.0	14.3	6.4	0.8
CMGs 108–114						.003†
Home	72.7	61.8	61.0	83.7	79.7
SNF	27.3	38.2	39.0	16.3	20.3

NOTE. CMGs 104–107: moderate stroke; CMGs 108–114: severe stroke.

For patients with severe strokes, having the mood and behavior disturbances component or both components (mood and behavioral disturbances, cognitive impairment) was associated with a longer mean LOS (≥26d), whereas patients in the severe group with no indication of neurobehavioral impairment had the shortest LOS for their group (22.1d, P=.030). Also, patients with severe stroke with both components of neurobehavioral impairment had significantly less improvement in motor FIM score and were more likely to be discharged to a skilled nursing facility (SNF). Patients with severe stroke with neurotropic medications only had the largest increase in motor FIM score and the lowest percentage of discharge to an SNF. In addition, the presence of cognitive impairment was associated most strongly with a higher percentage of patients discharged to an SNF in both the moderate and severe stroke groups.

Table 4 presents significant associations between neurotropic medication groups and 2 outcomes: rehabilitation LOS and increases in motor FIM score. Patients with moderate and severe stroke who received medications within specific neurotropic medication groups are compared with all patients in each group, controlling for patient characteristics (listed below table 4). Patients with moderate stroke (n=20) who were given atypical antipsychotic medications are compared with a control group of patients with moderate stroke who did not receive any atypical antipsychotic medicine. Patients given atypical antipsychotics statistically had the same LOS (15.7d) but a significantly improved motor FIM score with a change of 27.8 points as compared with the overall mean LOS and change in motor FIM score for the moderate stroke control group (15.2d and 22.4 points, respectively).

Table 4. Associations of Types of Medications and Outcomes by CMG Groups

Variables	CMG 104–107 (n=474)			CMG 108–114 (n=445)
Therapeutic Medication Class	n	Mean LOS (mean, 15.2)⁎	Mean Increase Motor FIM Score (mean, 22.4)⁎	n	Mean LOS (mean, 24.9)†	Mean Increase Motor FIM Score (mean, 24.5)†
Atypical antipsychotics	20	15.7(+0.5)	27.8‡(+5.4)	53	26.9(+2.0)	25.6(+1.1)
Traditional antipsychotics	7	18.4§(+3.2)	22.1(−0.3)	10	37.4‡(+12.5)	24.5(same)
Tricyclic antidepressants	15	18.6(+3.4)	20.2§(−2.2)	21	25.1(+0.2)	25.8(+1.3)
Old SSRIs	90	16.5(+1.3)	21.3§(−1.1)	104	25.8(+0.9)	21.4∥(−3.1)
New SSRIs	31	19.5§(+4.3)	24.2§(+1.8)	59	29.3(+4.4)	25.9(+1.4)
Analgesic; muscle relaxant	23	18.0(+2.8)	19.3‡(−3.1)	43	30.2‡(+5.3)	23.0(−1.5)
Anti-Parkinson’s medications	41	18.0§(+2.8)	18.1∥(−4.3)	68	28.4‡(+3.5)	22.8‡(−1.7)
Anxiolytics	4	23.8§(+8.6)	19.8(−2.6)	13	36.2‡(+11.3)	27.6(+3.1)
Hypnotics	87	17.2§(+2.0)	21.5(−0.9)	96	28.1‡(+3.2)	27.5‡(+3.0)
Modafinil	2	32.0∥(+16.8)	29.0(+6.6)	32	27.7(+2.8)	21.1§(−3.4)
Neurostimulants	16	19.9(+4.7)	18.0∥(−4.4)	57	28.4(+3.5)	22.1(−2.4)
Opioid analgesics	86	15.8(+0.6)	24.7‡(+2.3)	115	27.1(+2.2)	25.2(+0.7)
New antinausea/vomiting medications	15	15.9(+0.7)	21.3(−1.1)	34	29.2§(+4.3)	24.7(+0.2)
Old antinausea/vomiting medications	41	17.3(+2.1)	20.0‡(−2.4)	76	27.3(+2.4)	24.7(+0.2)
Sedating antihistamines	43	16.8(+1.6)	23.5(+1.1)	31	29.8§(+4.9)	26.6(+2.1)

NOTE. Values are means for patients with the specified medication and, in parentheses, the difference between cell mean value and overall mean for all patients in the CMG. Patient characteristics controlled in regression analyses include sex, age, race, side of stroke, type of stroke, mental status, pre–prospective payment system, discharge disposition, maximum CSI continuous score, admission motor FIM score, and admission cognitive FIM score.

For patients with moderate stroke, neurotropic medication groups associated with significantly longer rehabilitation LOSs were the traditional antipsychotics, modafinil, hypnotics, anxiolytics, anti-Parkinson’s medications, and newer selective serotonin reuptake inhibitors (SSRIs). Newer SSRIs, atypical antipsychotics, and opioid analgesics were associated with significantly greater increase in motor FIM score. Use of older antinausea medications, tricyclic antidepressants, anti-Parkinson’s medications, muscle relaxants, neurostimulants, and older SSRIs was associated with significantly less increase in motor FIM score.

For patients with severe strokes, use of muscle relaxants, anti-Parkinson’s medications, anxiolytics, hypnotics, new antinausea medications, sedating antihistamines, or traditional antipsychotics was associated with significantly longer rehabilitation LOSs. For these same severe patients, use of older SSRIs, anti-Parkinson’s medications, and modafinil was associated with significantly less improvement in motor FIM score, but use of hypnotics was associated with significantly more improvement in motor FIM.

Table 5, Table 6 show, for each PSROP facility, the percentage of patients who received medications in the neurotropic medication groups found to be associated with better or poorer outcomes (see table 4). Use of neurotropic medications varied significantly among the facilities for patients with moderate (see table 5) and severe strokes (see table 6), with the latter group having the greatest variation. Site variation is noticeable in the increase or decrease of neurotropic medication use as the severity of the stroke increases. For example, at site 4, use of new SSRIs is infrequent for all patients and use of old SSRIs increases from 27% for patients with moderate stroke (see table 5) to 52% for patients with severe stroke (see table 6). In contrast, at site 5 the overall use of old SSRIs is less frequent and the use of newer SSRIs increases as stroke severity increases.

Table 5. Percentage of Patients With Moderate Stroke (CMG 104–107) Using Specified Medication Categories by Site

Table 6. Percentage of Patients With Severe Stroke (CMG 108–114) Using Specified Medication Categories by Site

Back to Article Outline

Discussion 

Patients with severe strokes (CMGs 108–114) were older; were sicker at admission to, discharge from, and during their rehabilitation stays (CSI scores); were less likely to be discharged to home; and had longer LOSs than patients with moderate strokes. However, both patients with severe and moderate strokes had about the same increase in motor FIM and cognitive FIM scores from admission to discharge from rehabilitation. Within the moderate and severe stroke CMG groupings patients with no neurobehavioral impairments (no mood or behavior disturbances, no cognitive impairment, and no use of neurotropic medications) had the shortest LOSs and larger increases in motor FIM. When severity of illness (CSI) and its related components were not allowed to enter models by not including them in the variable selection list, the R² and c statistics changed little. Because none or very few of the other predictors changed, the models were stable.

We found several neurotropic medications associated with better outcomes and others that were associated with poorer outcomes. These varied by patient characteristics and severity of stroke. Generally, the newer medications (eg, newer SSRIs, atypical antipsychotics) were associated with better outcomes. Newer SSRIs were associated with greater improvement in FIM scores but also were associated with longer LOSs, making it difficult to draw definite conclusions about overall benefit. Older antinausea medications were associated with less FIM improvement for patients with moderate stroke and had no effect on LOS, suggesting a rationale for using the newer antinausea agents in this patient population, because the older antinausea medications may reduce FIM efficiency. Finally, atypical antipsychotics generally were associated with more increase in motor FIM score (primarily in the moderate stroke group), corresponding to our initial hypothesis that the more favorable side-effect profile of the atypical antipsychotic medication group in patients with stroke should translate into better outcomes.

Most facilities used newer medications sparingly. However, site 5 used newer SSRIs, newer antinausea medications, neurostimulants, and atypical antipsychotic medications more frequently, for patients with both moderate and severe stroke. After controlling for many patient characteristics (see table 2), we found that the association of neurobehavioral impairments with better or poorer outcomes in bivariate analyses remained significant in multiple regression analyses for LOS and increase in motor FIM score. That is, after using more thorough efforts to control for multiple patient characteristics in multiple sequences and combinations, outcomes consistently were better for patients with atypical antipsychotic medications than without.

There are a number of questions that can be raised about these initial observations. Many of these medications may have been used off-label in ways that their medication category description would not suggest. For instance, low-dose chlorpromazine is often used as a cure for intractable hiccups, and haloperidol is rarely used in poststroke rehabilitation except in the case of an elderly person who may be demented and experiencing sundowning. Use of anti-Parkinson’s neurostimulants has entirely different implications in the absence of Parkinson’s disease (of all the study patients who were given anti-Parkinson’s medications, only 3.9% had a documented diagnosis of Parkinson’s disease). Future analyses will attempt to understand discrepant uses of medications of interest.

However, there is evidence in the literature that these medications might be beneficial and justifies investigation of their effectiveness. During the early 1980s, studies were conducted on animals investigating the use of adrenergic agents on brain recovery after injury.³⁰, ³¹, ³² Later, Gualtieri³³ and Goldstein¹⁷ published articles advocating that other adrenergic agents, as well as their precursors, could facilitate recovery. Studies on the use of dopamine agonists (so-called “anti-Parkinson agents”) for brain injury in humans began in the 1990s, showing that these agents also could be used to help initiation and attention in these patients.³⁴, ³⁵, ³⁶, ³⁷, ³⁸, ³⁹, ⁴⁰ Dopamine agonists have since been used commonly in the treatment of brain injury. The use of dopamine agonists in patients with stroke so far has been limited to anecdotal or pilot studies; however, these articles are suggestive of their ability to facilitate cognitive capacity and recovery.

A meta-analysis of 7 generally high-level studies involving a total of 172 patients suggested that amphetamine treatment reduced death and dependence and relatively improved motor and language function.⁴¹ However, there were too few patients to draw any definite conclusions about effects of amphetamine treatment on recovery from stroke. A randomized, double-blind, placebo-controlled trial of 40 subjects using intravenous amantadine or placebo for 5 days showed statistically significant improvements in cadence, length of heel-to-toe movements in the single support phase, and variability in double support phase and double support time.⁴² A prospective, randomized, placebo-controlled, double-blind study of physical therapy combined with 3 weeks of daily levodopa or placebo and then 3 weeks of physical therapy alone showed increased motor function at both endpoints. Finally, 21 stroke survivors randomized to methylphenidate or placebo for 3 weeks scored lower on one depression scale and higher on a functional scale.⁴³

Atypical antipsychotics, particularly olanzapine, have been reported to enhance cognitive function, providing a possible basis for the positive association of these medications with better outcomes during stroke rehabilitation.⁴⁴, ⁴⁵, ⁴⁶ These positive reports need to be balanced with recent controversy about the off-label use of atypical antipsychotics in the management of elderly patients with dementia. A U.S. Food and Drug Administration Public Health Advisory⁴⁷ in April 2005 warned that a review of 17 controlled trials involving the use of atypical antipsychotics in elderly demented patients showed a 1.6- to 1.7-fold increase in mortality, mostly because of heart-related events and pneumonia. Like the present study, this report only indicates an association of increased mortality with these medications in a population with some similarity to our stroke population, not a cause-and-effect relation. Caution and further investigation are needed to confirm these findings.

Finally, we have not yet examined the specific ramifications of medication dosing, duration, or timing or medications being given simultaneously or in sequence. Nonetheless, these findings add to the body of quantified knowledge of how a stroke survivor is treated during poststroke inpatient rehabilitation and strengthen previously established observations that limiting access to newer medications may lead to higher overall costs through longer LOSs without concomitant improvements in motor FIM score change or rate of discharge to community.¹⁸, ⁴⁸

Back to Article Outline

Conclusions 

We found significant differences in the ways stroke rehabilitation physicians approach common neurocognitive impairments after stroke and in the choice of medications to lessen their negative impacts. This exploration of neurotropic medication utilization practice patterns and outcomes can be used to guide the design of future studies to enhance the efficient use of inpatient stroke rehabilitation resources and improve patient outcomes. Although they do not confirm a cause-and-effect relation, our results indicate that certain medications or classes of medications are associated with positive and negative effects on stroke rehabilitation outcomes and should be studied further.

Back to Article Outline

Acknowledgments 

We acknowledge contributions of collaborators at each clinical site represented in the Post-Stroke Rehabilitation Outcomes Project: Brendan Conroy, MD (Stroke Recovery Program, National Rehabilitation Hospital, Washington, DC); Richard Zorowitz, MD (Department of Rehabilitation Medicine, University of Pennsylvania Medical Center, Philadelphia, PA); David Ryser, MD (Neuro Specialty Rehabilitation Unit, LDS Hospital, Salt Lake City, UT); Jeffrey Teraoka, MD (Division of Physical Medicine and Rehabilitation, Stanford University, Palo Alto, CA); Frank Wong, MD, and LeeAnn Sims, RN (Rehabilitation Institute of Oregon, Legacy Health Systems, Portland, OR); Murray Brandstater, MD (Loma Linda University Medical Center, Loma Linda, CA); and Harry McNaughton, MD (Wellington and Kenepuru Hospitals, Wellington, NZ). We also acknowledge the role of Alan Jette, PhD (Rehabilitation Research and Training Center on Medical Rehabilitation Outcomes, Boston University, Boston, MA).

Back to Article Outline

References 

1. DeJong G , Horn S , Conroy B , Nichols D , Healton EB . Opening the black box of post-stroke rehabilitation (stroke rehabilitation patients, processes, and outcomes) . Arch Phys Med Rehabil . 2005;86(12 Suppl 2):S1–S7
   • View In Article
   • MEDLINE
2. Medicare Payment Advisory Commission . In: A data book (healthcare spending and the Medicare program) . Washington (DC): MedPAC; 2005;p. 150
   • View In Article
3. Duncan PW , Horner RD , Reker DM , et al.   Adherence to post acute rehabilitation guidelines is associated with functional recovery . Stroke . 2002;33:167–178
   • View In Article
   • CrossRef
4. Buchanan JL , Rumpel JD , Hoenig H . Charges for outpatient rehabilitation (growth and differences in provider types) . Arch Phys Med Rehabil . 1996;77:320–328
   • View In Article
   • Abstract
   • Full-Text PDF (1173 KB)
   • CrossRef
5. Gresham GE , Duncan PW , Stason WB , et al.   Post-stroke rehabilitation (priorities for future research) . Clinical practice guideline no. 16 . Rockville: U.S. Department of Health and Human Services, Agency for Health Care Policy and Research; 1995; AHCPR Publication No. 95-0062
   • View In Article
6. Carota A , Staub F , Bogousslavsky J . Emotions, behaviours and mood changes in stroke . Curr Opin Neurol . 2002;15(1):57–69
   • View In Article
   • MEDLINE
7. Bogousslavsky J . William Feinberg lecture 2002 (emotions, mood, and behavior after stroke) . Stroke . 2003;34:1046–1050
   • View In Article
   • CrossRef
8. Starkstein SE , Robinson RG . Affective disorders and cerebral vascular disease . Br J Psychiatry . 1989;154:170–182
   • View In Article
   • MEDLINE
   • CrossRef
9. Narushima K , Robinson RG . Stroke-related depression . Curr Atheroscler Rep . 2002;4:296–303
   • View In Article
   • MEDLINE
   • CrossRef
10. Fruehwald S , Gatterbauer E , Rehak P , Baumhackl U . Early fluoxetine treatment of post-stroke depression—a three-month double-blind placebo-controlled study with an open-label long-term follow up . J Neurol . 2003;250:347–351
    • View In Article
    • MEDLINE
    • CrossRef
11. Gupta A , Pansari K , Shetty H . Post-stroke depression . Int J Clin Pract . 2002;56:531–537
    • View In Article
    • MEDLINE
12. Turner-Stokes L , Hassan N . Depression after stroke (a review of the evidence base to inform the development of an integrated care pathway. Part 2: treatment alternatives) . Clin Rehabil . 2002;16:248–260
    • View In Article
    • MEDLINE
    • CrossRef
13. Paolucci S , Antonucci G , Grasso MG , et al.  A case-control study   Post-stroke depression, antidepressant treatment and rehabilitation results . Cerebrovasc Dis . 2001;12:264–271
    • View In Article
    • MEDLINE
    • CrossRef
14. Gall A . Post stroke depression . Hosp Med . 2001;62:268–273
    • View In Article
    • MEDLINE
15. Cole MG , Elie LM , McCusker J , Bellavance F , Mansour A . Feasibility and effectiveness of treatments for post-stroke depression in elderly inpatients (systematic review) . J Geriatr Psychiatry Neurol . 2001;14:37–41
    • View In Article
    • MEDLINE
    • CrossRef
16. Khan F . Post stroke depression . Aust Fam Physician . 2004;33:831–834
    • View In Article
    • MEDLINE
17. Goldstein LB . Pharmacology of recovery after stroke . Stroke . 1990;21(11 Suppl):III139–III142
    • View In Article
    • MEDLINE
18. Horn SD . Limiting access to psychiatric services can increase total health care costs . Drug Benefit Trends . 2003;15(Suppl 1):12–18
    • View In Article
19. Henon H , Pasquier F , Durieu  , Godefroy O . Preexisting dementia in stroke patients . Stroke . 1997;28:2429–2436
    • View In Article
    • MEDLINE
    • CrossRef
20. Goldstein LB  The Sygen In Acute Stroke Study Investigators . Common drugs may influence motor recovery after stroke . Neurology . 1995;45:865–871
    • View In Article
    • MEDLINE
    • CrossRef
21. Gassaway J , Horn SD , DeJong G , Smout RJ , Clark C , James R . Applying the clinical practice improvement approach to stroke rehabilitation (methods used and baseline results) . Arch Phys Med Rehabil . 2005;86(12 Suppl 2):S16–S33
    • View In Article
    • MEDLINE
22. Maulden S , Gassaway J , Horn SD , Smout RJ , DeJong G . Timing of initiation of rehabilitation after stroke . Arch Phys Med Rehabil . 2005;86(12 Suppl 2):S34–S40
    • View In Article
    • MEDLINE
23. Horn SD , Sharkey PD , Gassaway J . Managed Care Outcomes Project (study design, baseline patient characteristics, and outcome measures) . Am J Managed Care . 1996;2:237–247
    • View In Article
24. Horn SD , Sharkey PD , Buckle JM , Backofen JE , Averill RF , Horn RA . The relationship between severity of illness and hospital length of stay and mortality . Med Care . 1991;29:305–317
    • View In Article
    • MEDLINE
    • CrossRef
25. Averill RF , McGuire TE , Manning BE , et al.   A study of the relationship between severity of illness and hospital cost in New Jersey hospitals . Health Serv Res . 1992;27:587–606 ; discussion 607-12
    • View In Article
    • MEDLINE
26. Thomas TJ , Ashcraft ML . Measuring severity of illness (comparison of severity and severity systems in terms of ability to explain variation in costs) . Inquiry . 1991;28:39–55
    • View In Article
    • MEDLINE
27. Alemi F , Rice J , Hankins R . Predicting in-hospital survival of myocardial infarction, a comparative study of various severity measures . Med Care . 1990;28:762–774
    • View In Article
    • MEDLINE
    • CrossRef
28. Epocrates Inc. Available at: http://www2.epocrates.com. Accessed September 4, 2005
    • View In Article
29. In:  Hart AC ,  Schmidt KM ,  Aaron WS editor. ICD-9-CM code book . Reston: St. Anthony’s Publishing; 1999;
    • View In Article
30. Feeney DM , Gonzalez A , Law WA . Amphetamine, haloperidol and experience interact to affect the rate of recovery after motor cortex injury . Science . 1982;217:885–887
    • View In Article
31. Hovda DA , Feeney DM . Amphetamine with experience promotes recovery of locomotor function after unilateral frontal cortex injury in the cat . Brain Res . 1984;298:358–361
    • View In Article
    • MEDLINE
    • CrossRef
32. Evans RW , Gualtieri CT , Patterson D . Treatment of chronic closed head injury with psychostimulant drugs (a controlled case study and an appropriate evaluation procedure) . J Nerv Ment Dis . 1987;175:106–110
    • View In Article
    • MEDLINE
    • CrossRef
33. Gualtieri CT . Pharmacotherapy and the neurobehavioural sequelae of traumatic brain injury . Brain Inj . 1988;2:101–129
    • View In Article
    • MEDLINE
    • CrossRef
34. Gualtieri CT , Chandler M , Coons TB , Brown LT . Amantadine (a new clinical profile for traumatic brain injury) . Clin Neuropharmacol . 1989;12:258–270
    • View In Article
    • MEDLINE
    • CrossRef
35. Kraus MF , Maki P . Combined use of amantadine and L-dopa/carbidopa in the treatment of chronic brain injury . Brain Inj . 1997;11:455–460
    • View In Article
    • MEDLINE
    • CrossRef
36. Nickels JL , Schneider WN , Dombovy ML , Wong TM . Clinical use of amantadine in brain injury rehabilitation . Brain Inj . 1994;8:709–718
    • View In Article
    • MEDLINE
    • CrossRef
37. Tillerson JL , Cohen AD , Philhower J , Miller GW , Zigmond MJ , Schallert T . Forced limb use effects on the behavioral and neurochemical effects of 6-hydroxydopamine . J Neurosci . 2001;21:4427–4435
    • View In Article
38. Mukand JA , Guilmette TJ , Allen DG , et al.   Dopaminergic therapy with carbidopa L-dopa for left neglect after stroke . Arch Phys Med Rehabil . 2001;82:1279–1282
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (55 KB)
    • CrossRef
39. Kujac TL , Thomander D , Teraoka JK . Efficacy of amantadine in facilitating cognitive recovery in stroke patients [abstract] . Arch Phys Med Rehabil . 2001;82:1342
    • View In Article
40. Scheidtmann K , Fries W , Muller F , Koenig E . Effect of levodopa in combination with physiotherapy on functional motor recovery after stroke (a prospective, randomized, double-blind study) . Lancet . 2001;358:787–790
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (75 KB)
    • CrossRef
41. Martinsson L , Wahlgren NG , Hardemark HG . Amphetamines for improving recovery after stroke . Cochrane Database Syst Rev . 2003;3: CD002090
    • View In Article
42. Baezner H , Oster M , Henning O , Cohen S , Hennerici MG . Amantadine increases gait steadiness in frontal gait disorder due to subcortical vascular encephalopathy (a double-blind randomized placebo-controlled trial based on quantitative gait analyses) . Cerebrovasc Dis . 2001;11:235–244
    • View In Article
    • MEDLINE
    • CrossRef
43. Grade C , Redford B , Chrostowski J , Toussaint L , Blackwell B . Methylphenidate in early post stroke recovery (a double-blind, placebo-controlled study) . Arch Phys Med Rehabil . 1998;79:1047–1050
    • View In Article
    • Abstract
    • Full-Text PDF (610 KB)
    • CrossRef
44. Meltzer HY . Beyond control of acute exacerbation (enhancing affective and cognitive outcomes) . CNS Spectr . 2003;8(11 Suppl 2):16–18 , 22
    • View In Article
    • MEDLINE
45. Gur RE , Kohler C , Ragland JD , et al.   Neurocognitive performance and clinical changes in olanzapine-treated patients with schizophrenia . Neuropsychopharmacology . 2003;28:2029–2036
    • View In Article
    • MEDLINE
    • CrossRef
46. Tyson PJ , Roberts KH , Mortimer AM . Are the cognitive effects of atypical anti-psychotics influenced by their affinity to 5HT-2A receptors? . Int J Neurosci . 2004;114:593–611
    • View In Article
    • MEDLINE
    • CrossRef
47. U.S. Food and Drug Administration Center for Drug Evaluation and Research . FDA public health advisory (deaths with antipsychotics in elderly patients with behavioral disturbances) . 2005; April 11, Available at: http://www.fda.gov/cder/drug/advisory/antipsychotics.htm. Accessed September 4, 2005
    • View In Article
48. Horn SD , Sharkey PD , Tracy DM , Horn CE , James B , Goodwin F . Intended and unintended consequences of HMO cost-containment strategies (results from the Managed Care Outcomes Project) . Am J Managed Care . 1996;2:253–264
    • View In Article