- Email: [email protected]
Poststroke depression: An examination of the literature
658
REVIEW ARTICLE
Poststroke Depression:
An Examination
of the Literature
Wayne A. Gordon, PhD, Mary R. Hibbard, PhD ABSTRACT. Gordon WA, Hibbard MR. Poststroke depression: an examination of the literature. Arch Phys Med Rehabil 1997;78:658-63. Objective: To examine literature on poststroke depression (PSD). Data Sources: More than 200 articles related to stroke and depression were selected from a computer-based search spanning 1985 to 1995. Study Selection: All relevant articles on PSD. Articles in foreign languages, case studies, anecdotal reports, book chapters, and reviews were excluded. Data Extraction: Summary findings were independently reviewed by the authors. Data Synthesis: PSD remains a frequent sequela of stroke; its prevalence remains uncertain because of continued methodologic problems in defining subject groupings and in utilizing psychiatrically normed assessmenttools with neurologically impaired individuals, and because of the poor specificitylsensitivity of neuroendocrine markers in determining a diagnosis. The etiology of PSD appearsto be complex and not fully understood. Although there has been much research on PSD, this review highlights the sparsity of available literature on its treatment. Conclusion: The review points out the further need for more carefully designed studies of PSD that examine both assessment and treatment. 0 1997 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation
ORE THAN 1.5YEARS AGO, Labi and colleagues’ deM scribed depression as the most common untreated disability secondary to stroke. Poststroke depression (PSD) continues to be an area of significant research interest. A search of the Medline and Psychlit databases for the years 1985 to 1995, using the parameters “postroke and depression” and “poststroke depression”, uncovered 225 papers that were published in English. Fifty of the manuscripts were unique to Medline, 39 articles were unique to Psychlit, and 69 articles were common to both databases.Despite this abundance of research, considerable gaps remain in our understanding of PSD, especially with regard to such issues as its diagnosis, its neuroanatomic correlates, and its treatment. Interest in PSD stems from the fact that stroke is a major cause of disability, ie, more than 600,000 individuals sustain a new stroke each year and there are more than two million stroke From the Department of Rehabilitation Medicine, Mount Sinai Medical Center, New York, NY. Submitted for publication May 5, 1996. Accepted in revised form November 29, 1996. Supported in part by grant H133B30038, National Institute on Disability and Rehabilitation Research, United States Department of Education. No commercial party having a direct or indirect interest in the subject matter of this article has or will confer a benefit upon the authors or upon any organization with which the authors are associated. Reprint requests to Dr. Wayne A. Gordon, Department of Rehabilitation Medicine, Mount Sinai Medical Center, One Gustave Levy Place, Box 1240, New York, NY 10029. 0 1997 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation 0003.9993/97/7806-4133$3.00/O
Arch
Phys Mad
Rehabil
Vol78,
June
1997
survivors living in the United States. Stroke continues to be the most common diagnostic group among inpatient rehabilitation patients occupying more than one third of the beds.* Furthermore, PSD has been found to negatively affect length of stay3-5and functional outcomes.6-9PSD has also been associated with increased risk for subsequent mortality.” Thus, although innovations in the acute management of stroke have resulted in a decreasein its mortality rate and, to some extent, the magnitude of its impairments, little has been done to improve the quality of life of those who survive as that quality relates to depression. Consequently, many who survive the acute event will live lives that are compromised by depression.“-‘* This review will examine the diagnostic issues related to PSD, the variability in the reported prevalence of PSD, the neuroanatomic correlates of PSD, and the psychopharmacological and psychological treatment of PSD. THE PREVALENCE OF PSD: A MINOR OR MAJOR SEQUELA OF STROKE? In 1986, we reviewed extensively the early research on the prevalence of PSD and factors associatedwith its reported variability.13 Although PSD continues to be the subject of a significant amount of investigation, little research has been done that increases our understanding of, or accounts for, the variability in its reported prevalence. Reported PSD prevalence rates range from as low as 25% to as high as 79%.‘4-25Despite this variability, it is clear that a many individuals are debilitated by depression after stroke. Methodologic factors accounting for this variability include differing subject selection, methodologic problems with instrumentation, and limited utility of biochemical markers traditionally used in diagnosing depression. Each of these issues will be briefly discussed. Subject Selection Whether PSD is primarily associated with right or left brain damage is an issue that remains unresolved, since researchers often fail to separate the reported incidence of PSD in individuals after right and left cerebral strokes. Criteria for selection of patients in previous studies have differed widely and may also account, in part, for the variability in prevalence rates. For example, studies vary in the subsets of the stroke patients sampled (eg, inpatient versusoutpatient), in the time since the onset of the stroke (eg, acute versus chronic), in the neuroanatomy of the stroke (eg, cerebral versus noncerebral involvement, unilateral versus bilateral damage), in the inclusion and/or exclusion of individuals with global aphasia and/or concomitant dementia, and in the inclusion and/or exclusion of patients with prior psychiatric/substance abuse histories or other central nervous system diseases.These sampling differences limit the generalizability of present findings to other patient subgroups. Instrumentation Perhaps the most serious limitation of the PSD literature relates to the issue of diagnostic accuracy when standardized psychiatric assessmenttechniques, (ie, structured clinical interviews, self-report scales,and biochemical markers) are applied to neurologically impaired individuals. Traditional psychiatric
POSTSTROKE
DEPRESSION,
criteria for determining a mood disorder relies heavily on patients’ reports of specific symptoms in combination with depressed mood as indicators of depression. Implicit in this approach is the assumption that patients are aware of their situation, are capable of providing an accurate report of their mood, and do not have coexisting medical/neurological conditions that could either “mimic” targeted symptomatology or be interpreted more parsimoniously. When evaluating stroke patients, these basic assumptions must be reevaluated. Awareness cannot be assumed because stroke patients have been found to minimize and/or exaggerate their poststroke-related physical, cognitive, and affective changes in function.26-29 Indeed, individuals may be aware of some of their impairments yet minimize or exaggerate others. Thus, awareness is not a dichotomous variable but rather one that should be viewed along a continuum of impairment.2” This variability in awareness can bear directly on the accuracy of self-reports incorporated into traditional psychiatric assessment. In addition, when stroke patients have aphasia or severe dementia, they are often unable to provide accurate self-reports of their internal mood state. Hence, mood must be inferred from behavioral observations of these individuals or from data that are systematically gathered from others familiar with the patient’s affective states. It is our observation, along with that of other investigators3” that multiple sources of information are crucial to the diagnosis of PSD, yet they are not routinely or systematically pursued in making a psychiatric diagnosis. Given the problems inherent in the selfreports of stroke patients, a systematic multidimensional approach to the collection of information about the patients’ mood is clearly needed.3’ The validity of any psychiatric assessment is also clouded by the nature of the assessment tools used in making a diagnosis of PSD. Both traditional psychiatric criteria, ie, DSM-IV, and psychological measures of mood rely heavily on the specific somatic complaints in combination with depressed mood as indicators of depression. For example, five of the nine symptoms required for a DSM-IV diagnosis of a major mood disorder (ie, fatigue or loss of energy, weight loss or gain and/or appetite changes, insomnia, psychomotor changes, concentration difficulties) are somatic in nature. Similarly, 11 items on the Beck Depression Inventory and 8 items on the Hamilton Rating of Depression are somatic in nature. These symptoms are just as likely to reflect behavioral/neurological impairments, the effects of age, or the impact of the person’s immediate environment (eg, the effects of hospitalization) as they are to reflect the person’s altered mood. For example, impairments in memory, attention, and abstraction may frequently underlie the patients’ complaints of concentration difficulties after a stroke. Thus, a positive response to a question related to concentration could be mistakenly attributed to depression as opposed to an underlying cognitive impairment. Similarly, increased fatigue after a neurological insult could be misinterpreted as ahedonia, a symptom of depression, rather than be viewed as hypoarousal, a secondary impairment to the stroke. Thus, as a result of the over-representation of somatic symptoms in traditional psychiatric criteria, there is potential for the misdiagnosis of mood disorders. Several authors have argued for the creation of more sensitive criteria for the diagnosis of PSD. This belief is based in part on recent evidence that somatic symptoms are unreliable indicators of PSD. In fact, some investigators have found that somatic symptoms are sensitive 32 indicators of a mood disturbance in poststroke individuals, whereas other? have found the opposite to be true. Indeed it has been reported that “. . somatic symptoms were neither specific to PSD nor added incremental validity over nonsomatic symptoms for diagnosing PSD .‘r33 Furthermore this methodologic issue has been highlighted by recent
659
Gordon
research indicating that stroke patients are unreliable reporters of their mood.34 A third issue that limits accuracy in diagnosising PSD is the lack of a systematic way to assess the impact of neurobehavioral syndromes and cognitive changes on the validity of depressive symptoms reported by patients. How severe do impairments such as hypoarousal, alterations in affective regulation (ie, euphoria, pathological laughter, or crying), and difficulties with prosodic expression have to be for the person’s self-report to be unreliable? Two examples illustrate how neurobehavioral syndromes complicate the diagnosis of depression. First, hypoarousal is a common sequella of PSD,35 as we have found in our work36; tt is not, however, correlated with depression in stroke patients.37 Second, alterations in affect expression, ie, excessive tearfulness and constricted affect, are indicative of depressed mood in non-neurologically impaired individuals. In stroke patients, these same behaviors are more likely to be attributable to a neurologically based lability or to a prosodic deficit. Unlike neurologically impaired individuals, a stroke patient’s external affective behaviors may not match his or her internal mood.30,38 Recently, Starkstein and colleagues3’ provided additional support for the independence of a prosodic deficit and PSD. Of the stroke patients they examined, 49% had aprosodic deficits unrelated to depression. Depressed patients, as well as nondepressed patients, showed similar patterns of prosodic deficits. Thus, neurobehavioral sequellae of stroke, taken together or separately, complicate the diagnosis of PSD and directly affect the validity of PSD diagnostic statements.
Biochemical Markers of PSD In non-brain damaged individuals, the Dexamethasone Suppression Test (DST) is a neuroendocrine marker of melancholia that has acceptable levels of sensitivity and specificity.a Following the administration of dexamethasone, depressed individuals do not show suppressed plasma cortisol levels; nondepressed individuals do. Because the DST does not rely on the person’s self-report of symptoms or of his/her mood, it has been suggested as an alternative approach to diagnosing PSD. Unfortunately, research reports wide variability in both sensitivity and specificity of this diagnostic tool when it is used to assess poststroke individuals. In poststroke inpatients, reported sensitivities are as low as 0%41 and as high as 75%.42 The DST’s specificity is also quite varied and ranges from 70%43 to 95%42 for rehabilitation inpatients. Hamey and co11eagues44 found that 75% of their sample had abnormal DSTs 1 week after stroke while 50% had an abnormal DST 3 weeks after stroke. Although individuals with abnormal DSTs had higher scores on the Hamilton Rating of Depression, none met DSM-III criteria for depression. The authors suggest that abnormal DSTs at 3 weeks after stroke might be predictive of future depression. Thus, it is difficult to interpret these data with regard to the utility of the DST in diagnosing PSD during the acute phase of poststroke recovery. Studies of outpatients are even more confusing with respect to the relative merits of the DST as an indicator of PSD. In our study of a group of poststroke outpatients,45 we found the DST’s sensitivity was 15% and its specificity was 67%. We concluded the DST was not a measure useful in the diagnosis of PSD. A similar conclusion was reached by Dam and colleagues4’j In contrast, Astrom and colleagues“’ reported that the DST is useful, (sensitivity = 70%, specificity = 96%) in diagnosing depression in chronic, but not acute, strokes. Thus, the DST does not appear to be a useful adjunct to the diagnosis of PSD. It remains to be seen whether other neuroendocrine markers, eg, thyrotrophin releasing hormone or corticotropin releasing hormone, will be effective in diagnosing poststroke depression.
Arch
Phys
Med Rehabil
Vol78,
June
1997
660
POSTSTROKE
DEPRESSION,
CEREBRAL LATERALITY AND PSD Our understanding of the relation between the neuroanatomic loci of brain damage and the incidence of PSD is an unresolved issue. During the late 1970s and early 1980s several researchers reported a clear-cut relation between laterality of brain injury and PSD, with the preponderance of research supporting PSD as an impairment associated with left brain damage.48-51 In subsequent attempts to replicate these earlier studies, some researchers reported reverse findings, ie, PSD more commonly observed following right brain damage,‘“~2’~52-54 whereas others failed to demonstrate any association between the laterality of brain damage and PSD.55-59 PSD may reflect a more complex phenomenon than merely the neuroanatomic loci of brain damage. Schwartz and colleague?’ found PSD to be independent of lesion location, more severe in persons with right hemisphere lesions and correlated with functional disability, lesion size, and previous history of depression. In contrast, Downhill and Robinson6’ found that coexisting depression and impairments in cognition were more likely to be observed in individuals with left hemisphere strokes. Within the last 5 years, cerebral blood flow studies have been used to further examine PSD. These advanced imaging techniques have supported the notion that PSD is, indeed. a more complex phenomenon than anticipated. Using this imaging technique Grass0 and colleague@* reported a reduction in blood flow to the area of the mesial temporal cortex of the affected hemisphere in depressed stroke patients. Reduced blood flow was strongly correlated with the severity of depression observed in these individuals. Yamaguchi and colleague@’ found that individuals with lesions in the left frontal and/or right parieto-occipital regions were more likely to be depressed. In another study by Stern and colleagues,M lesion location within a hemisphere was found to be a more sensitive indicator of PSD than side of brain damage. More specifically, patients with lesions in the left parietal/occipital, left inferior frontal, right superior frontal, and right temporal lobe were more likely to be depressed than individuals whose strokes affected other areas of the brain. Over the past decade, Robinson and colleagues65 have refined their understanding of the neuroanatomic correlates of PSD. In more recent investigations, they report a relation between PSD and lesion location primarily in patients with typical occipital asymmetry.66 In patients with left hemispheric lesions, PSD is found primarily in those without a history of depression6’ Additionally, these researcher@* have reported that PSD and a generalized anxiety disorder are associated solely with cortical lesions, whereas depression alone is associated with subcortical lesions. In patients with right hemispheric strokes, depression is associated with lesions involving the parietal cortex and a family history of psychiatric disturbance. Finally, this research team has concluded that catastrophic reaction is a subtype of PSD. THE
TREATMENT
OF PSD
Given the prevalence of PSD and the amount of research focused on delineating its neuropathology, there is a surprising dearth of research focused on treatment of PSD. As in other depressive disorders, there are two traditional approaches to the treatment of PSD, one being antidepressant medication and the other being psychotherapy. When severe depressive reactions fail to respond to more traditional approaches, electroconvulsive therapy is considered. A review of the PSD literature suggests that there have been few studies on the efficacy of the various approaches to intervention and no published research on the relation among treatment efficacy, patient characteristics, and
Arch
Phys Med
Rehabil
Vol78,
June
1997
Gordon
the medical economics of intervention. Literature pertinent these domains of treatment are highlighted below.
to
Psychopharmacological Interventions Treatment of PSD with antidepressant medication is often effective.69 However, drug therapy must be carefully evaluated to avoid polypharmacological interactions and potential adverse side effects of a particular antidepressant on a patient’s comorbidities.70 Nortriptyline, a tricyclic antidepressant, was among the early choices of antidepressants reported to be successful in the treatment of PSD.7’,72 Indeed, Lipsey’s7’ study documenting the benefits of this medication is one of the few randomized clinical trials in the literature. Balunov and colleagues7’ compared the effects of amitriptyline to that of a tranquilizer (Seduxen) in individuals after stroke and a comparison group of individuals who received no treatment. They reported that amitriptyline was more effective in reducing depression, as measured by the Hamilton Rating Scale of Depression. Nortriptyline has also been reported to be an effective treatment of both emotional lability and depression after stroke.74 Andersen and colleagues75 compared the effects of two tricyclic antidepressant regimens (imipramine plus mianserin vs. desipramine plus mianserin) in 20 poststroke individuals whose Hamilton Rating Scale of Depression scores were 15 or above. A combination of imipramine plus mianserin was found to be superior to that of desipramine plus miaserin.72 Other antidepressant medications have been evaluated. In a randomized trial, Reding and colleagues76 found that Trazadone was useful in the treatment of PSD. In a 6-week double-blind, placebo control study, the effects of the drug Citalopram, a selective serotonin reuptake inhibitor, were examined in a sample of 66 stroke patients. It was found to significantly reduce depression, as measured by the Hamilton Rating Scale of Depression. Support for alterations in levels of serotonin in depressed stroke patients can be found in work by Bryer and colleagues.78 They found that depressed stroke patients exhibit lower concentrations of 5hydroxyindolacetic acid (5HIAA) in their cerebrospinal fluid when contrasted with either nondepressed stroke patients or nondepressed individuals without a stroke. Since 5-HIAA is a serotonin metabolite, this study again points to altered levels of serotonin in poststroke depressed individuals and suggests that more recently developed serotonin reuptake inhibitors, eg, Prozac, Paxil, etc, are worthy of clinical trials as treatment for PSD. An alternative pharmacologic approach to treatment of PSD has been the use of psychostimulants. In a comparison study79 of the relative effectiveness of two differing psychostimulants, ie, Dextroamphetamine and Methylphenidate, both drugs were equally effective in the treatment of PSD. Of the patients tested, 82% reported improved mood, with 47% showing “marked or moderate” improvement. Lazarus and colleagues” compared the psychostimulant methylphenidate to the antidepressant nortriptyline. An equivalent rate of remission of depressive symptoms was observed in both treatment groups (53% methlyphenidate, 43% nortriptyline); however, the response to treatment was more rapid in the individuals administered methylphenidate (average response time = 2.4 days for methylphenidate vs. 27 days for nortriptyline). Earlier studies by this same research teams1 and by others** described a reduction in depression following treatment with methylphenidate in 70% to 80% of patients (n = IO/study). Unfortunately the validity of these studies is compromised by a lack of adequate comparison groups, absence of randomization, and small sample sizes. While there is growing evidence to support the use of various antidepressant medications and psychostimulants in the treatment of PSD, further research is needed to identify the specific characteristics
POSTSTROKE
DEPRESSION,
of patients who respond best to either of these types of medications and whether there are benefits in using antidepressants and psychostimulants in combination. Use of Electroconvulsive Therapy Electroconvulsive therapy (ECT) has been viewed as an approach to treatment of depression, particularly in severely depressed individuals who fail to respond to traditional approaches. The efficacy of ECT in the treatment of PSD has been explored in two retrospective studies.s3,84 The first study, conducted by Murray and colleagues,83found that 14 of 193 (7%) individuals had been treated with ECT for PSD between 1969 and 1981. The authors report improved mood in 12 of the 14 patients (86%), with only one patient experiencing an adverse effect as a result of treatment. In contrast, Currier and associatess4reviewed charts covering a lo-year period from 1982 to 1991 and found only 20 depressed poststroke individuals who had been treated with ECT. The authors report improvement in 95% of the cases,and conclude that ECT is a “generally well tolerated and effective treatment.” Despite this positive interpretation by the authors, significant side effects were observed in 60% (12 of 20 patients) of the cases, ie, 7 patients suffered relapses despite their maintenance on antidepressant medication, 5 developed ECT-related medical complications and 3 pts experienced postinterictal confusion or amnesia. Given the prevalence of complications, we do not believe that the data reported support Currier and his colleagues’ conclusion. Thus, although data from retrospective studies appear to support the utility of ECT in the treatment of PSD, side effects of this intervention appear to be frequent. Therefore, a clinical trial examining the efficacy of ECT in poststroke individuals is indicated. Psychotherapeutic Interventions Given the high prevalence of PSD and the attention focused on pharmacological management, it is curious that minimal effort has been directed to the study of psychotherapy, either alone or combined with antidepressant medication as a treatment modality for PSD. Psychopharmacological treatment of PSD using the older tricylics, while often effective, is frequently contraindicated in patients with stroke because of age considerations and coexisting medical problems. More than 10 years ago, we compared the effectiveness of cognitive psychotherapy, Nortriptyline, or a combination of both in the treatment of PSD and found that many patients either refused to take these earlier types of antidepressant or stopped taking medications because of their side effects (unpublished data). Although it is not known whether the refusal rate for antidepressants is any different from that for any other type of medication (eg, anticoagulants, antihypertensives, etc), it is a still a factor to consider when contemplating available ‘therapeutic options. Unfortunately, only one study”j on the effectiveness of these newer forms of antidepressant medications (eg, Prozac, Zoloft, and Paxil) has included poststroke individuals. These drugs, however, appear to offer the potential for effective treatment of PSD with minimal side effects. Given the aforementioned considerations, psychotherapy frequently becomes the only viable intervention for PSD treatment. An obvious benefit is that the intervention can help depressed individuals learn new coping and social skills, thus increasing their senseof control over their current mood and their environment and helping prevent the recurrence of depressive episodes.85-87 The paucity of research on psychological interventions for PSD may reflect the inherent difficulty clinicians face when
661
Gordon
attempting to apply traditional psychotherapies to individuals who are cognitively compromised after stroke. We have reported a successfuladaptation of Beck’s” cognitive-behavioral psychotherapy to the treatment of PSD. Focused principles of cognitive therapy have also been developed as guidelines for successful application of this approach to the treatment of PSD.85*86~88,89 Clearly, further systematic investigation of the relative utility of psychotherapy alone or in combination with psychopharmacological agents is needed to advance our understanding of the treatment of PSD. SUMMARY This review of the literature on PSD has examined an extensive body of research. PSD remains a frequent sequela of stroke, but its actual prevalence remains unknown. This is due in large part to continued methodologic problems in the literature that stem from a lack of clarity regarding such issues as subject selection, tools used for assessment, and an apparent lack of specificity and sensitivity of traditional neuroendocrine markers of depression in individuals after stroke. While studies published in the early 1980s indicated a clear-cut relation between PSD and lesion location, the neuroanatomic correlates of PSD are more complex than initially anticipated. Thus, the early work of Robinson and colleagues’5*17*50 relating PSD to laterality of brain damage has evolved as more recent research has illustrated the importance of intrahemispheric loci of the lesion rather than the laterality of brain damage in the etiology of PSD. In addition, the extent of coexisting cognitive deficits and histories of psychiatric illness appear to play important roles in the presence and severity of the depression observed after stroke. Finally, this review has highlighted that treatment of PSD remains a relatively underexamined domain. Although a variety of antidepressant medications and psychostimulants appear to be effective in reducing the severity of PSD, these studies are short term, pay inadequate attention to the specific patient characteristics of those who respond to treatment, and fail to examine the long-term efficacy of these medications. Clearly, the cost benefits of ECT have not been adequately demonstrated in poststroke individuals. Furthermore, the demonstration that psychotherapy is a useful intervention for PSD, either alone or in combination with other therapeutic approaches, remains a challenge for future researchers. In summary, this review points out how much research has been done within the past decade, but we believe that little new information has been learned about PSD, which in turn has resulted in little forward movement in the field. Consequently much remains to be learned about the diagnosis and treatment of PSD. Acknowledgments: ments of Steven Flanagan, of this manuscript.
The authors appreciatethe constructivecomMD, and Jennifer Bodgany on earlier versions
References Labi MLC, Phillips TF, Gresham GE. Psychosocial disability in physically restored long-term stroke survivors. Arch Phys Med Rehabil 1980;61:561-5. Granger CV, Ottenbacher KJ, Fiedler C. The uniform data system for medical rehabilitation: report of first admissions for 1993. Am J Phys Med Rehabil 1995;74:62-9. Malec JF, Richardson JW, Sinaki M, O’Brien MW. Types of affective response to stroke. Arch Phys Med Rehabil 1990;71:27984. Shubert DS, Bums R, Para W, Sioson E. Increase of medical hospital length of stay by depression in stroke and amputation patients: a pilot study. Psychother Psychosom 1992;57:61-6.
Arch
Phys Med Rehebil
Vol78,
June
1997
662
POSTSTROKE
DEPRESSION,
5. Zalewshik C, Keller B, Bowers C, Miske P. Depressive symptomatology and post-stroke rehabilitation outcome. Clin Geront 1994; 14:62-76. 6. Par&h RM, Robinson RG, Lipsey JR. The impact of poststroke depression on recovery in activities of daily living over a two-year follow-up. Arch Nemo1 1990;47:785-9. 7. Morris PL, Raphael B, Robinson RG. Clinical depression is associated with impaired recovery from stroke. Med J Aust 1992; 157: 239-42. 8. Koivisto K, Viinamaki H, Reikkinen P. Poststroke depression and rehabilitation outcome. Nord J Psychiatry 1993;47:245-9. 9. Sharpe M, Hawton K, Seagroatt V, Bamford J, House A, Molyneux A, et al. Depressive disorders in long-term survivors of stroke. Associations with demographic and social factors, functional status, and brain lesion volume. Br J Psychiatry 1994; 160:380-6. 10. Morris PL, Robinson RG, Andrzejewski P, Samuels J, Price TR. Association of depression with lo-year poststroke mortality. Am J Psychiatry 1993; 150: 124-9. 11. Gordon WA, Diller L. Stroke: coping with a cognitive deficit. In: Burish E, Bradley LA, editors. Coping with chronic disease: research and applications 1983. New York: Academic Press, 1983: 113-35. 12. Astrom M, Asplund K, Astrom T. Psychosocial function and life satisfaction after stroke. Stroke 1992; 23:527-3 1. 13. Ruckdeschel-Hibbard M, Gordon WA, Diller L. Affective disturbances associated with brain damage. In: Filskov S, Boll T, editors. Handbook of clinical neuropsychology, vol 2. New York: John Wiley and Sons, 1986:457-72. 14. Gasparrini WG, Satz P, Heilman KM, Coolidge FL. Hemispheric asymmetries of affective processing as determined by Minnesota Multiphasic Personality Inventory. J Neurol Neurosurg Psychiatry 1978;41:470-3. 15. Robinson RG, Price TR. Post-stroke depressive disorders: a followup study of 103 patients. Stroke 1982;.13:635-41. 16. Finklestein S. Benowitz L. Baldessarini R. Arana GW, Levine D, Woo E, et al. Mood, vegetative disturbance and dexamethasone suppression test after stroke. Ann Neurol 1982; 12:463-8. 17. Robinson RG, Kubos KL, Starr LB, Rao K, Price TR. Mood changes in stroke patients: relationship to lesion location. Comp Psych 1983; 24:555-66. 18. Ruckdeschel-Hibbard M. Affective impairments in right brain damaged individuals [PhD dissertation]. New York: New York University, 1984. 19. Finset A, Goffeng L, Landro NI, Haakonsen M. Depressed mood and intra-hemispheric locations of lesion in right hemispheric stroke patients. Stand-J Rehabil Med 1989; 2 1: l-6. 20. Eaelko S. Simon D. Rilev E. Gordon WA. Ruckdeschel-Hibbard My Diller’L. First year after stroke: tracking cognitive and affective deficits. Arch Phys Med Rehabil 1989;70:297-302. 21. Gordon WA, Hibbard MR, Egelko S, Riley E, Simon D, Diller L, et al. Issues in the diagnosis of post-stroke depression. Rehabil Psych 1991;36:71-88. 22. Andersen G, Vestergaard K, Riis JO, Lauritzen L. Incidence of post-stroke depression during the first year in a large unselected stroke population determined using a valid standardized rating scale. Acta Psychiatr Stand 1994;90: 190-5. 23. Shima S, Kitagawa Y, Kitmaura T, Fujinawa A. Poststroke depression. Gen Hosp Psychiatry 1994; 16:286-9. 24. Morris PLP, Shields RB, Hopwood MJ, Robinson RG. Are there two depressive syndromes after stroke? J Nerv Ment Dis 1994; 182: 230-4. 25. Burvill AG, Johnson KD, Jamrozik D, Anderson CS, StewartWynne EG, Chakera TMH. Prevalence of depression after stroke: the Perth community stroke study. Br J Psychiatry 1995; 166:3207. 26. Gianotti G. Emotional behavior and hemispheric side of lesion. Cortex 1972;8:42-55. 27. McGlynn SM, Schacter DL. Unawareness of deficits in neuropsychological syndromes. J Clin Exp Neuropsychol 1989; 11:143-205. 28. Hibbard MR, Gordon WA, Stein PN, Grober S, Sliwinski MJ. Unawareness in stroke patients: A problem of the mind not the body [abstract]. Arch Phys Med Rehabil 1990;71:776. 29. Hibbard MR. Gordon WA, Stein PN, Grober S, Sliwinski M.
Arch
Phys Med
Rehabil
Vol78,
June
1997
30.
31.
32.
33.
34.
35.
36.
37.
38. 39.
40.
41.
42.
43.
44.
45.
46.
47.
48. 49.
50. 51. 52.
53.
Gordon
Awareness of disability in patients following stroke. Rehabil Psych 1992; 37: 103-20. Ross ED, Rush J. Diagnosis and neuroanatomical correlates of depression in brain-damaged patients. Arch Gen Psychiatry 1991;38: 1344-54. Hibbard MR, Gordon WG, Stein PN, Grober SW, Sliwinski M. A multi modal approach to the diagnosis of post-stroke depression. In: Gordon WA, editor. Advances in stroke rehabilitation. Andover (MA): Andover Medical Publishers, 1993:185-214. Fedoroff JP, Starkstein SE, Parikh RM, Price TR, Robinson RG. Are depressive symptoms nonspecific in patients with acute stroke? Am J Psychiatry 1991; 148:1172-6. Stein PN. Sliwinski M. Gordon WA. Hibbard MR. The disciminative properties of somatic and non-somatic symptoms for post-stroke depression. Clin Neuropsychol 1996; 10: 141-8. Toedter LJ, Schall R, Reese CA, Hyland DT, Berk SN, Dunn DS. Psychological measures: reliability in the assessment of stroke patients. Arch Phys Med Rehabil 1995;76:719-25. Heilman KM, Schwartz HD, Watson RT. Hypoarousal in patients with neglect syndrome and emotional indifference. Neurology 1978; 281229-32. Gordon WA, Hibbard MR, Egelko S, Diller L. The multifaceted nature of the cognitive deficits following stroke: an unexpected finding [abstract]. Arch Phys Med Rehabil 1985;66:538. Gordon WA. Methodlogical considerations in cognitive remediation. In: Meier M, Benton A, Diller L, editors. Neuropsychological rehabilitation. London: Churchill Livingston, 1987: 11 l-3 1. Ross ED. The aurosodias. Arch Neurol 1981:38:561-9. Starkstein SE, Federoff JP, Price TR, Leiguarda RC, Robinson RG. Neuropsychological and neuroradiologic correlates of emotional prosody comprehension. Neurology 1994;44:515-22. Carroll BJ, Feinberg M, Greden JF, Tarika J, Albala AA, Haskett RF, et al. A specific laboratory test for the diagnosis of melancholia: standardization, validation and clinical utility. Arch Gen Psychiatry 1981;38:15-22. Bauer M, Gans JS, Harley JP, Cobb W. Dexamethasone suppression test and depression in a rehabilitation setting. Arch Phys Med Rehabil 1983;64:421-2. Finkelstein S, Benowitz LI, Baldessarini RJ, Arana GW, Levine D, Woo E, et al. Mood, vegetative disturbance, and dexamethasone suppression test after stroke. Ann Neurol 1982; 12:463-8. Lipsey JR, Robinson RG, Pearlson GD, Rao K, Price TR. The dexamethasone suppression test and mood following stroke. Am J Psychiatry 1985; 142:318-23. Hamey JH, Fulton C, Ross ED, Rush AJ. Dexamethasone suppression test and onset of poststroke depression in patients with ischemic infarction. J Clin Psychiatry 1993;54:34-8. Grober SE, Gordon WA, Sliwinski M, Hibbard MR, Aletta EG, Paddison PL. Utility of the Dexamethasone Test in the diagnosis of poststroke depression. Arch Phys Med Rehabil 1991;72: 1076-9. Dam H, Pedersen HE, Damkjaer M, Ahlgren P. Dexamethasone suppression test in depressive stroke patients. Acta Neurol Stand 1991;84:14-7. Astrom M, Olsson T, Asplund K. Different linkage of depression to hypercortisolism early versus late after stroke. A 3-year longitudinal study. Stroke 1993;24:52-7. Black FW. Unilateral brain lesions and MMPI performance: a preliminary study. Percpt Mot Skills 1975;40:87-93. Gasparrini W, Satz P, Heilman K, Coolidge F. Hemispheric asymmetries of affective processing determined by Minnesota Multiphasic Personality Inventory. J Neurol Neurosug Psychiatry 1978;41: 473-80. Robinson RG, Szetela B. Mood change following left-hemisphere brain injury. Ann Neurol 198 1; 9:447-53. Robin AH. Are stroke patients more depressed that other disabled subjects? J Chron Dis 1976;22:479-82. Folstein M, Maiberger R, McHugh P. Mood disturbance as specific complication of stroke. J Neurol Neurosurg Psychiatry 1977;40: 1018-20. Egelko S, Gordon WA, Hibbard MR. The relationship among CT scans, neurological exam and neuropsychological test performance in right brain damage stroke patients. J Clin Exp Neuropsychol 1989; 10:539-64.
POSTSTROKE
DEPRESSION,
54. Dikman S, Reitan R. MMPI correlates of localization in cerebral lesions. Percept Mot Skills 1974;22:831-40. 55. Kerns R. Depression following stroke: self-evaluation neuropsychological valuation and laterality of lesion as predictor variables [dissertation]. Carbondale (IL): Southern Illinois University, 1980. 56. Sinyor D, Jacques P, Kaloupek D, Becker R, Goldenberg M, Coopersmith H. Post-stroke depression and lesion location. Brain 1986; 109:537-46. 57. Agrell B, Dehlin 0. Depression in stroke patients with right and left hemisphere lesions: a study in geriatric rehabilitation in-patients. Aging-Milan0 1994;6:49-56. 58. Sharpe M, Hawton K, House A, Molyneux A, Sandercock P, Bamford .I, et al. Mood disorders in long-term survivors of stroke: association with brain lesion location and volume. Psycho1 Med 1990; 20:815-28. 59. House A, Dennis M, Warlow C, Hawton K, Molyneux A. Mood disorders after stroke and their relation to lesion location. Brain 1990; 113:1113-29. 60. Schwartz JA, Speed NM, Brunberg JA, Brewer TL, Brown M, Greden JF. Depression in stroke rehabilitation. Biol Psychiatry 1993;33:694-9. 6 1. Downhill JE, Robinson RG. Longitudinal assessment of depression and cognitive impairment following stroke. J Nerv Ment Dis 1994; 182:425-31. 62. Grass0 MG, Pantano P, Ricci M, Intiso DF, Pace A, Padovani A, et al. Mesial temporal hypoperfusion is associated with depression in subcortical stroke. Stroke 1994;25:980-5. 63. Yamaguchi S, Kobayashi S, Koide H, Tsunematsu S. Logitudinal study of regional cerebral blood flow changes in depression after stroke. Stroke 1992;23:1716-22. 64. Stem RA, Bachman DL. Depressive symptoms following stroke. Am J Psychiatry 1991; 148:351-6. 65. Robinson RG, Starkstein SE. Current research in affective disorders following stroke. J Neuropsychiatry Clin Neurosci 1990; 2: 1 - 14. 66. Starkstein SE, Bryer JB, Berthier ML, Cohen B, Price TR, Robinson RG. Depression after stroke: the importance of cerebral hemisphere asymmetries. J Neuropsychiatry Clin Neurosci 1991;3:276-85. 67. Morris PL, Robinson RG, Raphael B. The relationship between risk factors for affective disorder and poststroke depression in hospitalized stroke patients. Aust N Z J Psychiatry 1992;26:208-17. 68. Starkstein SE, Fedoroff JP, Price TR, Leiguarda R, Robinson RG. Catastrophic reaction after cerebrovascular lesions: frequency, correlates and validation of a scale. J Neuropsychiatry Clin Neurosci 1993;5:189-94. 69. Finklestein SP, Weintraub RJ, Karmouz N, Askinazi C, Davar G, Baldessarini RJ. Anti-depressant drug treatment for post-stroke depression: a retrospective study. Arch Phys Med Rehabil 1987;68: 772-6. 70. Cunningham LA. Depression in the medically ill: choosing an antidepressant. J Clin Psychiatry 1994;55 Suppl A:90-7. 7 1. Lipsey JR, Robinson RG, Pearlson GD, Rao L, Price TR. Nortriptyline treatment of post-stroke depression: a double-blind study. Lancet 1984; 1:297-300.
72.
73. 74.
75.
76.
77.
78.
79. 80.
81.
82.
83. 84.
85.
86.
87. 88.
89.
663
Gordon
Federoff JP, Robinson RG. Tricyclic antidepressants in the treatment of poststroke depression. J Clin Psychiatry 1989;50 Suppl: 18-23. Balunov OA, Sadov OG, Alemasova AY. Therapy of depression in post-stroke patients. Alaska Med 1990;32:20-9. Robinson RG, Robert G, Parikh RM, Lipsey JT, Starkstein SE, et al. Pathological laughing and crying following stroke: validation of a measurement scale and a double-blind treatment study. Am J Psychiatry 1993; 150:286-93. Lauritzen L, Bendsen BB, Vilmar T, Bendsen EB, Lunde M, Beth P. Post-stroke depression: combined treatment with imipramine or desipramine and mianserin: a controlled clinical study. Psychopharmacology 1994; 114: 119-22. Reding MJ, Orto LA, Winter SW, Fortuna IM, Di Ponte P, McDowell FH. Antidepressant therapy after stroke. a double blind trial. Arch Neurol 1986;43:763-5. Andersen G, Vestergaard K, Lauritzen L. Effective treatment of poststroke depression with the selective serotonin reuptake inhibitor citalopram. Stroke 1994;25:1099-1104. Bryer JP, Starksein SE, Votypka V, Parikh RM, Price TR, Robinson RG. Reduction of CSF monoamine metabolites in poststroke depression: a preliminary report. J Neuropsychiatry Clin Neurosci 1992;4:440-2. Masand P, Murray G, Pickett P. Psycho stimulants in post-stroke depression. J Neuropsychiatry Clin Neurosci 1991;3:23-7. Lazarus LW, Moberg PJ, Langsley PR, Lingam VR. Methylphenidate and nortriptyline in the treatment of post stroke depression: a retrospective comparison. Arch Phys Med Rehabil 1994;75:403-6. Lazarus LW, Winemiller DR. Lingam VR, Newman I, Martman C, Abassian M, et al. Efficacy and side effects of methlyphenidate for post stroke depression. J Clin Psychiatry 1992;53:447-9. Johnson ML, Roberts MD, Ross AR, Written CM. Methylphenidate in stroke patients with depression. Am J Phys Med Rehabil 1992; 73:239-41. Murray GB, Shea VM, Conn DK. Electroconvulsive therapy for poststroke depression. J Clin Psychiat 1986;47:258-60. Currier MB, Murray GB, Welch CC. Electroconvulsive therapy for post-stroke depressed geriatric patients. J Neuropsychiatry Clin Neurosci 1992;4:140-4. Hibbard MR, Grober SE, Gordon WA, Aletta EG, Freeman A. Cognitive therapy and the treatment of poststroke depression. Top Geriatr Rehabil 1990; 5:43-55. Grober S, Hibbard MR, Gordon WA, Stein PN, Freeman A. The psychotherapeutic treatment of post-stroke depression with cognitive behavioral therapy. In: Gordon WA, editor. Advances in stroke rehabilitation. Andover (MA): Andover Medical Publishers, 1993: 217-41. Beck AT, Rush AJ, Shaw BF, Emery G. Cognitive therapy of depression. New York: Guilford Press, 1979. Hibbard MR, Grober SA, Gordon WA, Aletta EA. Modification of cognitive psychotherapy for the treatment of post-stroke depression. The Behav Therapist 1990; Jan: 15-7. Hibbard MR, Grober SE, Stein PN, Gordon WA. Post stroke depression. In: Freeman A, Dattilio FM, editors. Casebook in cognitive behavior therapy. New York: Plenum Press, 1992:303-10.
Arch
Phys Med
Rehabil
Vol 78, June
1997
REVIEW ARTICLE
Poststroke Depression:
An Examination
of the Literature
Wayne A. Gordon, PhD, Mary R. Hibbard, PhD ABSTRACT. Gordon WA, Hibbard MR. Poststroke depression: an examination of the literature. Arch Phys Med Rehabil 1997;78:658-63. Objective: To examine literature on poststroke depression (PSD). Data Sources: More than 200 articles related to stroke and depression were selected from a computer-based search spanning 1985 to 1995. Study Selection: All relevant articles on PSD. Articles in foreign languages, case studies, anecdotal reports, book chapters, and reviews were excluded. Data Extraction: Summary findings were independently reviewed by the authors. Data Synthesis: PSD remains a frequent sequela of stroke; its prevalence remains uncertain because of continued methodologic problems in defining subject groupings and in utilizing psychiatrically normed assessmenttools with neurologically impaired individuals, and because of the poor specificitylsensitivity of neuroendocrine markers in determining a diagnosis. The etiology of PSD appearsto be complex and not fully understood. Although there has been much research on PSD, this review highlights the sparsity of available literature on its treatment. Conclusion: The review points out the further need for more carefully designed studies of PSD that examine both assessment and treatment. 0 1997 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation
ORE THAN 1.5YEARS AGO, Labi and colleagues’ deM scribed depression as the most common untreated disability secondary to stroke. Poststroke depression (PSD) continues to be an area of significant research interest. A search of the Medline and Psychlit databases for the years 1985 to 1995, using the parameters “postroke and depression” and “poststroke depression”, uncovered 225 papers that were published in English. Fifty of the manuscripts were unique to Medline, 39 articles were unique to Psychlit, and 69 articles were common to both databases.Despite this abundance of research, considerable gaps remain in our understanding of PSD, especially with regard to such issues as its diagnosis, its neuroanatomic correlates, and its treatment. Interest in PSD stems from the fact that stroke is a major cause of disability, ie, more than 600,000 individuals sustain a new stroke each year and there are more than two million stroke From the Department of Rehabilitation Medicine, Mount Sinai Medical Center, New York, NY. Submitted for publication May 5, 1996. Accepted in revised form November 29, 1996. Supported in part by grant H133B30038, National Institute on Disability and Rehabilitation Research, United States Department of Education. No commercial party having a direct or indirect interest in the subject matter of this article has or will confer a benefit upon the authors or upon any organization with which the authors are associated. Reprint requests to Dr. Wayne A. Gordon, Department of Rehabilitation Medicine, Mount Sinai Medical Center, One Gustave Levy Place, Box 1240, New York, NY 10029. 0 1997 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation 0003.9993/97/7806-4133$3.00/O
Arch
Phys Mad
Rehabil
Vol78,
June
1997
survivors living in the United States. Stroke continues to be the most common diagnostic group among inpatient rehabilitation patients occupying more than one third of the beds.* Furthermore, PSD has been found to negatively affect length of stay3-5and functional outcomes.6-9PSD has also been associated with increased risk for subsequent mortality.” Thus, although innovations in the acute management of stroke have resulted in a decreasein its mortality rate and, to some extent, the magnitude of its impairments, little has been done to improve the quality of life of those who survive as that quality relates to depression. Consequently, many who survive the acute event will live lives that are compromised by depression.“-‘* This review will examine the diagnostic issues related to PSD, the variability in the reported prevalence of PSD, the neuroanatomic correlates of PSD, and the psychopharmacological and psychological treatment of PSD. THE PREVALENCE OF PSD: A MINOR OR MAJOR SEQUELA OF STROKE? In 1986, we reviewed extensively the early research on the prevalence of PSD and factors associatedwith its reported variability.13 Although PSD continues to be the subject of a significant amount of investigation, little research has been done that increases our understanding of, or accounts for, the variability in its reported prevalence. Reported PSD prevalence rates range from as low as 25% to as high as 79%.‘4-25Despite this variability, it is clear that a many individuals are debilitated by depression after stroke. Methodologic factors accounting for this variability include differing subject selection, methodologic problems with instrumentation, and limited utility of biochemical markers traditionally used in diagnosing depression. Each of these issues will be briefly discussed. Subject Selection Whether PSD is primarily associated with right or left brain damage is an issue that remains unresolved, since researchers often fail to separate the reported incidence of PSD in individuals after right and left cerebral strokes. Criteria for selection of patients in previous studies have differed widely and may also account, in part, for the variability in prevalence rates. For example, studies vary in the subsets of the stroke patients sampled (eg, inpatient versusoutpatient), in the time since the onset of the stroke (eg, acute versus chronic), in the neuroanatomy of the stroke (eg, cerebral versus noncerebral involvement, unilateral versus bilateral damage), in the inclusion and/or exclusion of individuals with global aphasia and/or concomitant dementia, and in the inclusion and/or exclusion of patients with prior psychiatric/substance abuse histories or other central nervous system diseases.These sampling differences limit the generalizability of present findings to other patient subgroups. Instrumentation Perhaps the most serious limitation of the PSD literature relates to the issue of diagnostic accuracy when standardized psychiatric assessmenttechniques, (ie, structured clinical interviews, self-report scales,and biochemical markers) are applied to neurologically impaired individuals. Traditional psychiatric
POSTSTROKE
DEPRESSION,
criteria for determining a mood disorder relies heavily on patients’ reports of specific symptoms in combination with depressed mood as indicators of depression. Implicit in this approach is the assumption that patients are aware of their situation, are capable of providing an accurate report of their mood, and do not have coexisting medical/neurological conditions that could either “mimic” targeted symptomatology or be interpreted more parsimoniously. When evaluating stroke patients, these basic assumptions must be reevaluated. Awareness cannot be assumed because stroke patients have been found to minimize and/or exaggerate their poststroke-related physical, cognitive, and affective changes in function.26-29 Indeed, individuals may be aware of some of their impairments yet minimize or exaggerate others. Thus, awareness is not a dichotomous variable but rather one that should be viewed along a continuum of impairment.2” This variability in awareness can bear directly on the accuracy of self-reports incorporated into traditional psychiatric assessment. In addition, when stroke patients have aphasia or severe dementia, they are often unable to provide accurate self-reports of their internal mood state. Hence, mood must be inferred from behavioral observations of these individuals or from data that are systematically gathered from others familiar with the patient’s affective states. It is our observation, along with that of other investigators3” that multiple sources of information are crucial to the diagnosis of PSD, yet they are not routinely or systematically pursued in making a psychiatric diagnosis. Given the problems inherent in the selfreports of stroke patients, a systematic multidimensional approach to the collection of information about the patients’ mood is clearly needed.3’ The validity of any psychiatric assessment is also clouded by the nature of the assessment tools used in making a diagnosis of PSD. Both traditional psychiatric criteria, ie, DSM-IV, and psychological measures of mood rely heavily on the specific somatic complaints in combination with depressed mood as indicators of depression. For example, five of the nine symptoms required for a DSM-IV diagnosis of a major mood disorder (ie, fatigue or loss of energy, weight loss or gain and/or appetite changes, insomnia, psychomotor changes, concentration difficulties) are somatic in nature. Similarly, 11 items on the Beck Depression Inventory and 8 items on the Hamilton Rating of Depression are somatic in nature. These symptoms are just as likely to reflect behavioral/neurological impairments, the effects of age, or the impact of the person’s immediate environment (eg, the effects of hospitalization) as they are to reflect the person’s altered mood. For example, impairments in memory, attention, and abstraction may frequently underlie the patients’ complaints of concentration difficulties after a stroke. Thus, a positive response to a question related to concentration could be mistakenly attributed to depression as opposed to an underlying cognitive impairment. Similarly, increased fatigue after a neurological insult could be misinterpreted as ahedonia, a symptom of depression, rather than be viewed as hypoarousal, a secondary impairment to the stroke. Thus, as a result of the over-representation of somatic symptoms in traditional psychiatric criteria, there is potential for the misdiagnosis of mood disorders. Several authors have argued for the creation of more sensitive criteria for the diagnosis of PSD. This belief is based in part on recent evidence that somatic symptoms are unreliable indicators of PSD. In fact, some investigators have found that somatic symptoms are sensitive 32 indicators of a mood disturbance in poststroke individuals, whereas other? have found the opposite to be true. Indeed it has been reported that “. . somatic symptoms were neither specific to PSD nor added incremental validity over nonsomatic symptoms for diagnosing PSD .‘r33 Furthermore this methodologic issue has been highlighted by recent
659
Gordon
research indicating that stroke patients are unreliable reporters of their mood.34 A third issue that limits accuracy in diagnosising PSD is the lack of a systematic way to assess the impact of neurobehavioral syndromes and cognitive changes on the validity of depressive symptoms reported by patients. How severe do impairments such as hypoarousal, alterations in affective regulation (ie, euphoria, pathological laughter, or crying), and difficulties with prosodic expression have to be for the person’s self-report to be unreliable? Two examples illustrate how neurobehavioral syndromes complicate the diagnosis of depression. First, hypoarousal is a common sequella of PSD,35 as we have found in our work36; tt is not, however, correlated with depression in stroke patients.37 Second, alterations in affect expression, ie, excessive tearfulness and constricted affect, are indicative of depressed mood in non-neurologically impaired individuals. In stroke patients, these same behaviors are more likely to be attributable to a neurologically based lability or to a prosodic deficit. Unlike neurologically impaired individuals, a stroke patient’s external affective behaviors may not match his or her internal mood.30,38 Recently, Starkstein and colleagues3’ provided additional support for the independence of a prosodic deficit and PSD. Of the stroke patients they examined, 49% had aprosodic deficits unrelated to depression. Depressed patients, as well as nondepressed patients, showed similar patterns of prosodic deficits. Thus, neurobehavioral sequellae of stroke, taken together or separately, complicate the diagnosis of PSD and directly affect the validity of PSD diagnostic statements.
Biochemical Markers of PSD In non-brain damaged individuals, the Dexamethasone Suppression Test (DST) is a neuroendocrine marker of melancholia that has acceptable levels of sensitivity and specificity.a Following the administration of dexamethasone, depressed individuals do not show suppressed plasma cortisol levels; nondepressed individuals do. Because the DST does not rely on the person’s self-report of symptoms or of his/her mood, it has been suggested as an alternative approach to diagnosing PSD. Unfortunately, research reports wide variability in both sensitivity and specificity of this diagnostic tool when it is used to assess poststroke individuals. In poststroke inpatients, reported sensitivities are as low as 0%41 and as high as 75%.42 The DST’s specificity is also quite varied and ranges from 70%43 to 95%42 for rehabilitation inpatients. Hamey and co11eagues44 found that 75% of their sample had abnormal DSTs 1 week after stroke while 50% had an abnormal DST 3 weeks after stroke. Although individuals with abnormal DSTs had higher scores on the Hamilton Rating of Depression, none met DSM-III criteria for depression. The authors suggest that abnormal DSTs at 3 weeks after stroke might be predictive of future depression. Thus, it is difficult to interpret these data with regard to the utility of the DST in diagnosing PSD during the acute phase of poststroke recovery. Studies of outpatients are even more confusing with respect to the relative merits of the DST as an indicator of PSD. In our study of a group of poststroke outpatients,45 we found the DST’s sensitivity was 15% and its specificity was 67%. We concluded the DST was not a measure useful in the diagnosis of PSD. A similar conclusion was reached by Dam and colleagues4’j In contrast, Astrom and colleagues“’ reported that the DST is useful, (sensitivity = 70%, specificity = 96%) in diagnosing depression in chronic, but not acute, strokes. Thus, the DST does not appear to be a useful adjunct to the diagnosis of PSD. It remains to be seen whether other neuroendocrine markers, eg, thyrotrophin releasing hormone or corticotropin releasing hormone, will be effective in diagnosing poststroke depression.
Arch
Phys
Med Rehabil
Vol78,
June
1997
660
POSTSTROKE
DEPRESSION,
CEREBRAL LATERALITY AND PSD Our understanding of the relation between the neuroanatomic loci of brain damage and the incidence of PSD is an unresolved issue. During the late 1970s and early 1980s several researchers reported a clear-cut relation between laterality of brain injury and PSD, with the preponderance of research supporting PSD as an impairment associated with left brain damage.48-51 In subsequent attempts to replicate these earlier studies, some researchers reported reverse findings, ie, PSD more commonly observed following right brain damage,‘“~2’~52-54 whereas others failed to demonstrate any association between the laterality of brain damage and PSD.55-59 PSD may reflect a more complex phenomenon than merely the neuroanatomic loci of brain damage. Schwartz and colleague?’ found PSD to be independent of lesion location, more severe in persons with right hemisphere lesions and correlated with functional disability, lesion size, and previous history of depression. In contrast, Downhill and Robinson6’ found that coexisting depression and impairments in cognition were more likely to be observed in individuals with left hemisphere strokes. Within the last 5 years, cerebral blood flow studies have been used to further examine PSD. These advanced imaging techniques have supported the notion that PSD is, indeed. a more complex phenomenon than anticipated. Using this imaging technique Grass0 and colleague@* reported a reduction in blood flow to the area of the mesial temporal cortex of the affected hemisphere in depressed stroke patients. Reduced blood flow was strongly correlated with the severity of depression observed in these individuals. Yamaguchi and colleague@’ found that individuals with lesions in the left frontal and/or right parieto-occipital regions were more likely to be depressed. In another study by Stern and colleagues,M lesion location within a hemisphere was found to be a more sensitive indicator of PSD than side of brain damage. More specifically, patients with lesions in the left parietal/occipital, left inferior frontal, right superior frontal, and right temporal lobe were more likely to be depressed than individuals whose strokes affected other areas of the brain. Over the past decade, Robinson and colleagues65 have refined their understanding of the neuroanatomic correlates of PSD. In more recent investigations, they report a relation between PSD and lesion location primarily in patients with typical occipital asymmetry.66 In patients with left hemispheric lesions, PSD is found primarily in those without a history of depression6’ Additionally, these researcher@* have reported that PSD and a generalized anxiety disorder are associated solely with cortical lesions, whereas depression alone is associated with subcortical lesions. In patients with right hemispheric strokes, depression is associated with lesions involving the parietal cortex and a family history of psychiatric disturbance. Finally, this research team has concluded that catastrophic reaction is a subtype of PSD. THE
TREATMENT
OF PSD
Given the prevalence of PSD and the amount of research focused on delineating its neuropathology, there is a surprising dearth of research focused on treatment of PSD. As in other depressive disorders, there are two traditional approaches to the treatment of PSD, one being antidepressant medication and the other being psychotherapy. When severe depressive reactions fail to respond to more traditional approaches, electroconvulsive therapy is considered. A review of the PSD literature suggests that there have been few studies on the efficacy of the various approaches to intervention and no published research on the relation among treatment efficacy, patient characteristics, and
Arch
Phys Med
Rehabil
Vol78,
June
1997
Gordon
the medical economics of intervention. Literature pertinent these domains of treatment are highlighted below.
to
Psychopharmacological Interventions Treatment of PSD with antidepressant medication is often effective.69 However, drug therapy must be carefully evaluated to avoid polypharmacological interactions and potential adverse side effects of a particular antidepressant on a patient’s comorbidities.70 Nortriptyline, a tricyclic antidepressant, was among the early choices of antidepressants reported to be successful in the treatment of PSD.7’,72 Indeed, Lipsey’s7’ study documenting the benefits of this medication is one of the few randomized clinical trials in the literature. Balunov and colleagues7’ compared the effects of amitriptyline to that of a tranquilizer (Seduxen) in individuals after stroke and a comparison group of individuals who received no treatment. They reported that amitriptyline was more effective in reducing depression, as measured by the Hamilton Rating Scale of Depression. Nortriptyline has also been reported to be an effective treatment of both emotional lability and depression after stroke.74 Andersen and colleagues75 compared the effects of two tricyclic antidepressant regimens (imipramine plus mianserin vs. desipramine plus mianserin) in 20 poststroke individuals whose Hamilton Rating Scale of Depression scores were 15 or above. A combination of imipramine plus mianserin was found to be superior to that of desipramine plus miaserin.72 Other antidepressant medications have been evaluated. In a randomized trial, Reding and colleagues76 found that Trazadone was useful in the treatment of PSD. In a 6-week double-blind, placebo control study, the effects of the drug Citalopram, a selective serotonin reuptake inhibitor, were examined in a sample of 66 stroke patients. It was found to significantly reduce depression, as measured by the Hamilton Rating Scale of Depression. Support for alterations in levels of serotonin in depressed stroke patients can be found in work by Bryer and colleagues.78 They found that depressed stroke patients exhibit lower concentrations of 5hydroxyindolacetic acid (5HIAA) in their cerebrospinal fluid when contrasted with either nondepressed stroke patients or nondepressed individuals without a stroke. Since 5-HIAA is a serotonin metabolite, this study again points to altered levels of serotonin in poststroke depressed individuals and suggests that more recently developed serotonin reuptake inhibitors, eg, Prozac, Paxil, etc, are worthy of clinical trials as treatment for PSD. An alternative pharmacologic approach to treatment of PSD has been the use of psychostimulants. In a comparison study79 of the relative effectiveness of two differing psychostimulants, ie, Dextroamphetamine and Methylphenidate, both drugs were equally effective in the treatment of PSD. Of the patients tested, 82% reported improved mood, with 47% showing “marked or moderate” improvement. Lazarus and colleagues” compared the psychostimulant methylphenidate to the antidepressant nortriptyline. An equivalent rate of remission of depressive symptoms was observed in both treatment groups (53% methlyphenidate, 43% nortriptyline); however, the response to treatment was more rapid in the individuals administered methylphenidate (average response time = 2.4 days for methylphenidate vs. 27 days for nortriptyline). Earlier studies by this same research teams1 and by others** described a reduction in depression following treatment with methylphenidate in 70% to 80% of patients (n = IO/study). Unfortunately the validity of these studies is compromised by a lack of adequate comparison groups, absence of randomization, and small sample sizes. While there is growing evidence to support the use of various antidepressant medications and psychostimulants in the treatment of PSD, further research is needed to identify the specific characteristics
POSTSTROKE
DEPRESSION,
of patients who respond best to either of these types of medications and whether there are benefits in using antidepressants and psychostimulants in combination. Use of Electroconvulsive Therapy Electroconvulsive therapy (ECT) has been viewed as an approach to treatment of depression, particularly in severely depressed individuals who fail to respond to traditional approaches. The efficacy of ECT in the treatment of PSD has been explored in two retrospective studies.s3,84 The first study, conducted by Murray and colleagues,83found that 14 of 193 (7%) individuals had been treated with ECT for PSD between 1969 and 1981. The authors report improved mood in 12 of the 14 patients (86%), with only one patient experiencing an adverse effect as a result of treatment. In contrast, Currier and associatess4reviewed charts covering a lo-year period from 1982 to 1991 and found only 20 depressed poststroke individuals who had been treated with ECT. The authors report improvement in 95% of the cases,and conclude that ECT is a “generally well tolerated and effective treatment.” Despite this positive interpretation by the authors, significant side effects were observed in 60% (12 of 20 patients) of the cases, ie, 7 patients suffered relapses despite their maintenance on antidepressant medication, 5 developed ECT-related medical complications and 3 pts experienced postinterictal confusion or amnesia. Given the prevalence of complications, we do not believe that the data reported support Currier and his colleagues’ conclusion. Thus, although data from retrospective studies appear to support the utility of ECT in the treatment of PSD, side effects of this intervention appear to be frequent. Therefore, a clinical trial examining the efficacy of ECT in poststroke individuals is indicated. Psychotherapeutic Interventions Given the high prevalence of PSD and the attention focused on pharmacological management, it is curious that minimal effort has been directed to the study of psychotherapy, either alone or combined with antidepressant medication as a treatment modality for PSD. Psychopharmacological treatment of PSD using the older tricylics, while often effective, is frequently contraindicated in patients with stroke because of age considerations and coexisting medical problems. More than 10 years ago, we compared the effectiveness of cognitive psychotherapy, Nortriptyline, or a combination of both in the treatment of PSD and found that many patients either refused to take these earlier types of antidepressant or stopped taking medications because of their side effects (unpublished data). Although it is not known whether the refusal rate for antidepressants is any different from that for any other type of medication (eg, anticoagulants, antihypertensives, etc), it is a still a factor to consider when contemplating available ‘therapeutic options. Unfortunately, only one study”j on the effectiveness of these newer forms of antidepressant medications (eg, Prozac, Zoloft, and Paxil) has included poststroke individuals. These drugs, however, appear to offer the potential for effective treatment of PSD with minimal side effects. Given the aforementioned considerations, psychotherapy frequently becomes the only viable intervention for PSD treatment. An obvious benefit is that the intervention can help depressed individuals learn new coping and social skills, thus increasing their senseof control over their current mood and their environment and helping prevent the recurrence of depressive episodes.85-87 The paucity of research on psychological interventions for PSD may reflect the inherent difficulty clinicians face when
661
Gordon
attempting to apply traditional psychotherapies to individuals who are cognitively compromised after stroke. We have reported a successfuladaptation of Beck’s” cognitive-behavioral psychotherapy to the treatment of PSD. Focused principles of cognitive therapy have also been developed as guidelines for successful application of this approach to the treatment of PSD.85*86~88,89 Clearly, further systematic investigation of the relative utility of psychotherapy alone or in combination with psychopharmacological agents is needed to advance our understanding of the treatment of PSD. SUMMARY This review of the literature on PSD has examined an extensive body of research. PSD remains a frequent sequela of stroke, but its actual prevalence remains unknown. This is due in large part to continued methodologic problems in the literature that stem from a lack of clarity regarding such issues as subject selection, tools used for assessment, and an apparent lack of specificity and sensitivity of traditional neuroendocrine markers of depression in individuals after stroke. While studies published in the early 1980s indicated a clear-cut relation between PSD and lesion location, the neuroanatomic correlates of PSD are more complex than initially anticipated. Thus, the early work of Robinson and colleagues’5*17*50 relating PSD to laterality of brain damage has evolved as more recent research has illustrated the importance of intrahemispheric loci of the lesion rather than the laterality of brain damage in the etiology of PSD. In addition, the extent of coexisting cognitive deficits and histories of psychiatric illness appear to play important roles in the presence and severity of the depression observed after stroke. Finally, this review has highlighted that treatment of PSD remains a relatively underexamined domain. Although a variety of antidepressant medications and psychostimulants appear to be effective in reducing the severity of PSD, these studies are short term, pay inadequate attention to the specific patient characteristics of those who respond to treatment, and fail to examine the long-term efficacy of these medications. Clearly, the cost benefits of ECT have not been adequately demonstrated in poststroke individuals. Furthermore, the demonstration that psychotherapy is a useful intervention for PSD, either alone or in combination with other therapeutic approaches, remains a challenge for future researchers. In summary, this review points out how much research has been done within the past decade, but we believe that little new information has been learned about PSD, which in turn has resulted in little forward movement in the field. Consequently much remains to be learned about the diagnosis and treatment of PSD. Acknowledgments: ments of Steven Flanagan, of this manuscript.
The authors appreciatethe constructivecomMD, and Jennifer Bodgany on earlier versions
References Labi MLC, Phillips TF, Gresham GE. Psychosocial disability in physically restored long-term stroke survivors. Arch Phys Med Rehabil 1980;61:561-5. Granger CV, Ottenbacher KJ, Fiedler C. The uniform data system for medical rehabilitation: report of first admissions for 1993. Am J Phys Med Rehabil 1995;74:62-9. Malec JF, Richardson JW, Sinaki M, O’Brien MW. Types of affective response to stroke. Arch Phys Med Rehabil 1990;71:27984. Shubert DS, Bums R, Para W, Sioson E. Increase of medical hospital length of stay by depression in stroke and amputation patients: a pilot study. Psychother Psychosom 1992;57:61-6.
Arch
Phys Med Rehebil
Vol78,
June
1997
662
POSTSTROKE
DEPRESSION,
5. Zalewshik C, Keller B, Bowers C, Miske P. Depressive symptomatology and post-stroke rehabilitation outcome. Clin Geront 1994; 14:62-76. 6. Par&h RM, Robinson RG, Lipsey JR. The impact of poststroke depression on recovery in activities of daily living over a two-year follow-up. Arch Nemo1 1990;47:785-9. 7. Morris PL, Raphael B, Robinson RG. Clinical depression is associated with impaired recovery from stroke. Med J Aust 1992; 157: 239-42. 8. Koivisto K, Viinamaki H, Reikkinen P. Poststroke depression and rehabilitation outcome. Nord J Psychiatry 1993;47:245-9. 9. Sharpe M, Hawton K, Seagroatt V, Bamford J, House A, Molyneux A, et al. Depressive disorders in long-term survivors of stroke. Associations with demographic and social factors, functional status, and brain lesion volume. Br J Psychiatry 1994; 160:380-6. 10. Morris PL, Robinson RG, Andrzejewski P, Samuels J, Price TR. Association of depression with lo-year poststroke mortality. Am J Psychiatry 1993; 150: 124-9. 11. Gordon WA, Diller L. Stroke: coping with a cognitive deficit. In: Burish E, Bradley LA, editors. Coping with chronic disease: research and applications 1983. New York: Academic Press, 1983: 113-35. 12. Astrom M, Asplund K, Astrom T. Psychosocial function and life satisfaction after stroke. Stroke 1992; 23:527-3 1. 13. Ruckdeschel-Hibbard M, Gordon WA, Diller L. Affective disturbances associated with brain damage. In: Filskov S, Boll T, editors. Handbook of clinical neuropsychology, vol 2. New York: John Wiley and Sons, 1986:457-72. 14. Gasparrini WG, Satz P, Heilman KM, Coolidge FL. Hemispheric asymmetries of affective processing as determined by Minnesota Multiphasic Personality Inventory. J Neurol Neurosurg Psychiatry 1978;41:470-3. 15. Robinson RG, Price TR. Post-stroke depressive disorders: a followup study of 103 patients. Stroke 1982;.13:635-41. 16. Finklestein S. Benowitz L. Baldessarini R. Arana GW, Levine D, Woo E, et al. Mood, vegetative disturbance and dexamethasone suppression test after stroke. Ann Neurol 1982; 12:463-8. 17. Robinson RG, Kubos KL, Starr LB, Rao K, Price TR. Mood changes in stroke patients: relationship to lesion location. Comp Psych 1983; 24:555-66. 18. Ruckdeschel-Hibbard M. Affective impairments in right brain damaged individuals [PhD dissertation]. New York: New York University, 1984. 19. Finset A, Goffeng L, Landro NI, Haakonsen M. Depressed mood and intra-hemispheric locations of lesion in right hemispheric stroke patients. Stand-J Rehabil Med 1989; 2 1: l-6. 20. Eaelko S. Simon D. Rilev E. Gordon WA. Ruckdeschel-Hibbard My Diller’L. First year after stroke: tracking cognitive and affective deficits. Arch Phys Med Rehabil 1989;70:297-302. 21. Gordon WA, Hibbard MR, Egelko S, Riley E, Simon D, Diller L, et al. Issues in the diagnosis of post-stroke depression. Rehabil Psych 1991;36:71-88. 22. Andersen G, Vestergaard K, Riis JO, Lauritzen L. Incidence of post-stroke depression during the first year in a large unselected stroke population determined using a valid standardized rating scale. Acta Psychiatr Stand 1994;90: 190-5. 23. Shima S, Kitagawa Y, Kitmaura T, Fujinawa A. Poststroke depression. Gen Hosp Psychiatry 1994; 16:286-9. 24. Morris PLP, Shields RB, Hopwood MJ, Robinson RG. Are there two depressive syndromes after stroke? J Nerv Ment Dis 1994; 182: 230-4. 25. Burvill AG, Johnson KD, Jamrozik D, Anderson CS, StewartWynne EG, Chakera TMH. Prevalence of depression after stroke: the Perth community stroke study. Br J Psychiatry 1995; 166:3207. 26. Gianotti G. Emotional behavior and hemispheric side of lesion. Cortex 1972;8:42-55. 27. McGlynn SM, Schacter DL. Unawareness of deficits in neuropsychological syndromes. J Clin Exp Neuropsychol 1989; 11:143-205. 28. Hibbard MR, Gordon WA, Stein PN, Grober S, Sliwinski MJ. Unawareness in stroke patients: A problem of the mind not the body [abstract]. Arch Phys Med Rehabil 1990;71:776. 29. Hibbard MR. Gordon WA, Stein PN, Grober S, Sliwinski M.
Arch
Phys Med
Rehabil
Vol78,
June
1997
30.
31.
32.
33.
34.
35.
36.
37.
38. 39.
40.
41.
42.
43.
44.
45.
46.
47.
48. 49.
50. 51. 52.
53.
Gordon
Awareness of disability in patients following stroke. Rehabil Psych 1992; 37: 103-20. Ross ED, Rush J. Diagnosis and neuroanatomical correlates of depression in brain-damaged patients. Arch Gen Psychiatry 1991;38: 1344-54. Hibbard MR, Gordon WG, Stein PN, Grober SW, Sliwinski M. A multi modal approach to the diagnosis of post-stroke depression. In: Gordon WA, editor. Advances in stroke rehabilitation. Andover (MA): Andover Medical Publishers, 1993:185-214. Fedoroff JP, Starkstein SE, Parikh RM, Price TR, Robinson RG. Are depressive symptoms nonspecific in patients with acute stroke? Am J Psychiatry 1991; 148:1172-6. Stein PN. Sliwinski M. Gordon WA. Hibbard MR. The disciminative properties of somatic and non-somatic symptoms for post-stroke depression. Clin Neuropsychol 1996; 10: 141-8. Toedter LJ, Schall R, Reese CA, Hyland DT, Berk SN, Dunn DS. Psychological measures: reliability in the assessment of stroke patients. Arch Phys Med Rehabil 1995;76:719-25. Heilman KM, Schwartz HD, Watson RT. Hypoarousal in patients with neglect syndrome and emotional indifference. Neurology 1978; 281229-32. Gordon WA, Hibbard MR, Egelko S, Diller L. The multifaceted nature of the cognitive deficits following stroke: an unexpected finding [abstract]. Arch Phys Med Rehabil 1985;66:538. Gordon WA. Methodlogical considerations in cognitive remediation. In: Meier M, Benton A, Diller L, editors. Neuropsychological rehabilitation. London: Churchill Livingston, 1987: 11 l-3 1. Ross ED. The aurosodias. Arch Neurol 1981:38:561-9. Starkstein SE, Federoff JP, Price TR, Leiguarda RC, Robinson RG. Neuropsychological and neuroradiologic correlates of emotional prosody comprehension. Neurology 1994;44:515-22. Carroll BJ, Feinberg M, Greden JF, Tarika J, Albala AA, Haskett RF, et al. A specific laboratory test for the diagnosis of melancholia: standardization, validation and clinical utility. Arch Gen Psychiatry 1981;38:15-22. Bauer M, Gans JS, Harley JP, Cobb W. Dexamethasone suppression test and depression in a rehabilitation setting. Arch Phys Med Rehabil 1983;64:421-2. Finkelstein S, Benowitz LI, Baldessarini RJ, Arana GW, Levine D, Woo E, et al. Mood, vegetative disturbance, and dexamethasone suppression test after stroke. Ann Neurol 1982; 12:463-8. Lipsey JR, Robinson RG, Pearlson GD, Rao K, Price TR. The dexamethasone suppression test and mood following stroke. Am J Psychiatry 1985; 142:318-23. Hamey JH, Fulton C, Ross ED, Rush AJ. Dexamethasone suppression test and onset of poststroke depression in patients with ischemic infarction. J Clin Psychiatry 1993;54:34-8. Grober SE, Gordon WA, Sliwinski M, Hibbard MR, Aletta EG, Paddison PL. Utility of the Dexamethasone Test in the diagnosis of poststroke depression. Arch Phys Med Rehabil 1991;72: 1076-9. Dam H, Pedersen HE, Damkjaer M, Ahlgren P. Dexamethasone suppression test in depressive stroke patients. Acta Neurol Stand 1991;84:14-7. Astrom M, Olsson T, Asplund K. Different linkage of depression to hypercortisolism early versus late after stroke. A 3-year longitudinal study. Stroke 1993;24:52-7. Black FW. Unilateral brain lesions and MMPI performance: a preliminary study. Percpt Mot Skills 1975;40:87-93. Gasparrini W, Satz P, Heilman K, Coolidge F. Hemispheric asymmetries of affective processing determined by Minnesota Multiphasic Personality Inventory. J Neurol Neurosug Psychiatry 1978;41: 473-80. Robinson RG, Szetela B. Mood change following left-hemisphere brain injury. Ann Neurol 198 1; 9:447-53. Robin AH. Are stroke patients more depressed that other disabled subjects? J Chron Dis 1976;22:479-82. Folstein M, Maiberger R, McHugh P. Mood disturbance as specific complication of stroke. J Neurol Neurosurg Psychiatry 1977;40: 1018-20. Egelko S, Gordon WA, Hibbard MR. The relationship among CT scans, neurological exam and neuropsychological test performance in right brain damage stroke patients. J Clin Exp Neuropsychol 1989; 10:539-64.
POSTSTROKE
DEPRESSION,
54. Dikman S, Reitan R. MMPI correlates of localization in cerebral lesions. Percept Mot Skills 1974;22:831-40. 55. Kerns R. Depression following stroke: self-evaluation neuropsychological valuation and laterality of lesion as predictor variables [dissertation]. Carbondale (IL): Southern Illinois University, 1980. 56. Sinyor D, Jacques P, Kaloupek D, Becker R, Goldenberg M, Coopersmith H. Post-stroke depression and lesion location. Brain 1986; 109:537-46. 57. Agrell B, Dehlin 0. Depression in stroke patients with right and left hemisphere lesions: a study in geriatric rehabilitation in-patients. Aging-Milan0 1994;6:49-56. 58. Sharpe M, Hawton K, House A, Molyneux A, Sandercock P, Bamford .I, et al. Mood disorders in long-term survivors of stroke: association with brain lesion location and volume. Psycho1 Med 1990; 20:815-28. 59. House A, Dennis M, Warlow C, Hawton K, Molyneux A. Mood disorders after stroke and their relation to lesion location. Brain 1990; 113:1113-29. 60. Schwartz JA, Speed NM, Brunberg JA, Brewer TL, Brown M, Greden JF. Depression in stroke rehabilitation. Biol Psychiatry 1993;33:694-9. 6 1. Downhill JE, Robinson RG. Longitudinal assessment of depression and cognitive impairment following stroke. J Nerv Ment Dis 1994; 182:425-31. 62. Grass0 MG, Pantano P, Ricci M, Intiso DF, Pace A, Padovani A, et al. Mesial temporal hypoperfusion is associated with depression in subcortical stroke. Stroke 1994;25:980-5. 63. Yamaguchi S, Kobayashi S, Koide H, Tsunematsu S. Logitudinal study of regional cerebral blood flow changes in depression after stroke. Stroke 1992;23:1716-22. 64. Stem RA, Bachman DL. Depressive symptoms following stroke. Am J Psychiatry 1991; 148:351-6. 65. Robinson RG, Starkstein SE. Current research in affective disorders following stroke. J Neuropsychiatry Clin Neurosci 1990; 2: 1 - 14. 66. Starkstein SE, Bryer JB, Berthier ML, Cohen B, Price TR, Robinson RG. Depression after stroke: the importance of cerebral hemisphere asymmetries. J Neuropsychiatry Clin Neurosci 1991;3:276-85. 67. Morris PL, Robinson RG, Raphael B. The relationship between risk factors for affective disorder and poststroke depression in hospitalized stroke patients. Aust N Z J Psychiatry 1992;26:208-17. 68. Starkstein SE, Fedoroff JP, Price TR, Leiguarda R, Robinson RG. Catastrophic reaction after cerebrovascular lesions: frequency, correlates and validation of a scale. J Neuropsychiatry Clin Neurosci 1993;5:189-94. 69. Finklestein SP, Weintraub RJ, Karmouz N, Askinazi C, Davar G, Baldessarini RJ. Anti-depressant drug treatment for post-stroke depression: a retrospective study. Arch Phys Med Rehabil 1987;68: 772-6. 70. Cunningham LA. Depression in the medically ill: choosing an antidepressant. J Clin Psychiatry 1994;55 Suppl A:90-7. 7 1. Lipsey JR, Robinson RG, Pearlson GD, Rao L, Price TR. Nortriptyline treatment of post-stroke depression: a double-blind study. Lancet 1984; 1:297-300.
72.
73. 74.
75.
76.
77.
78.
79. 80.
81.
82.
83. 84.
85.
86.
87. 88.
89.
663
Gordon
Federoff JP, Robinson RG. Tricyclic antidepressants in the treatment of poststroke depression. J Clin Psychiatry 1989;50 Suppl: 18-23. Balunov OA, Sadov OG, Alemasova AY. Therapy of depression in post-stroke patients. Alaska Med 1990;32:20-9. Robinson RG, Robert G, Parikh RM, Lipsey JT, Starkstein SE, et al. Pathological laughing and crying following stroke: validation of a measurement scale and a double-blind treatment study. Am J Psychiatry 1993; 150:286-93. Lauritzen L, Bendsen BB, Vilmar T, Bendsen EB, Lunde M, Beth P. Post-stroke depression: combined treatment with imipramine or desipramine and mianserin: a controlled clinical study. Psychopharmacology 1994; 114: 119-22. Reding MJ, Orto LA, Winter SW, Fortuna IM, Di Ponte P, McDowell FH. Antidepressant therapy after stroke. a double blind trial. Arch Neurol 1986;43:763-5. Andersen G, Vestergaard K, Lauritzen L. Effective treatment of poststroke depression with the selective serotonin reuptake inhibitor citalopram. Stroke 1994;25:1099-1104. Bryer JP, Starksein SE, Votypka V, Parikh RM, Price TR, Robinson RG. Reduction of CSF monoamine metabolites in poststroke depression: a preliminary report. J Neuropsychiatry Clin Neurosci 1992;4:440-2. Masand P, Murray G, Pickett P. Psycho stimulants in post-stroke depression. J Neuropsychiatry Clin Neurosci 1991;3:23-7. Lazarus LW, Moberg PJ, Langsley PR, Lingam VR. Methylphenidate and nortriptyline in the treatment of post stroke depression: a retrospective comparison. Arch Phys Med Rehabil 1994;75:403-6. Lazarus LW, Winemiller DR. Lingam VR, Newman I, Martman C, Abassian M, et al. Efficacy and side effects of methlyphenidate for post stroke depression. J Clin Psychiatry 1992;53:447-9. Johnson ML, Roberts MD, Ross AR, Written CM. Methylphenidate in stroke patients with depression. Am J Phys Med Rehabil 1992; 73:239-41. Murray GB, Shea VM, Conn DK. Electroconvulsive therapy for poststroke depression. J Clin Psychiat 1986;47:258-60. Currier MB, Murray GB, Welch CC. Electroconvulsive therapy for post-stroke depressed geriatric patients. J Neuropsychiatry Clin Neurosci 1992;4:140-4. Hibbard MR, Grober SE, Gordon WA, Aletta EG, Freeman A. Cognitive therapy and the treatment of poststroke depression. Top Geriatr Rehabil 1990; 5:43-55. Grober S, Hibbard MR, Gordon WA, Stein PN, Freeman A. The psychotherapeutic treatment of post-stroke depression with cognitive behavioral therapy. In: Gordon WA, editor. Advances in stroke rehabilitation. Andover (MA): Andover Medical Publishers, 1993: 217-41. Beck AT, Rush AJ, Shaw BF, Emery G. Cognitive therapy of depression. New York: Guilford Press, 1979. Hibbard MR, Grober SA, Gordon WA, Aletta EA. Modification of cognitive psychotherapy for the treatment of post-stroke depression. The Behav Therapist 1990; Jan: 15-7. Hibbard MR, Grober SE, Stein PN, Gordon WA. Post stroke depression. In: Freeman A, Dattilio FM, editors. Casebook in cognitive behavior therapy. New York: Plenum Press, 1992:303-10.
Arch
Phys Med
Rehabil
Vol 78, June
1997