Pharmaceutical Interventions for Emotionalism After Stroke
Disturbances of emotional behavior, referred to as emotionalism, manifest by easy crying, or less often laughing, is a common complication of stroke, affecting at least one quarter of survivors. In most cases the disorder is mild and transient, but when severe it may cause distress and embarrassment both to the patient and their friends and family, leading to the avoidance of social contact and reduced quality of life. Our objective in this study was to assess the effectiveness of pharmacological interventions, as compared with placebo, for emotionalism after stroke.
We searched the Cochrane Stroke Group Trials Register (last searched in June 2003), Cochrane Controlled Trials Register (Cochrane Library, Issue 3, 2002), MEDLINE (1966 to September 2002), EMBASE (1980 to September 2002), CINAHL (1982 to September 2002), PsychINFO (1967 to September 2002), Applied Science and Technology Plus (1986 to September 2002), Arts and Humanities Index (1991 to September 2002), Biological Abstracts (1969 to September 2002), General Science Plus (1994 to September 2002), Science Citation Index (1992 to September 2002), Social Sciences Citation Index (1991 to September 2002), and Sociofile (1974 to September 2002). We checked all relevant articles and textbooks and wrote to authors of known studies and pharmaceutical companies who manufacture psychotropic medications.
We considered all truly randomized (and quasi-randomized) controlled trials comparing psychotropic medication to placebo in patients with clinically diagnosed stroke and emotionalism, as defined by the authors in each study.
We identified 5 trials (published in 12 articles) with 103 participants at entry.1–5⇓⇓⇓⇓ Two trials were of crossover design1,4⇓ and outcome data were not available from the first phase (before crossover) in an appropriate format for inclusion. This review reports data from 3 trials with 75 participants.
The 3 trials showed treatment effects on the primary endpoint of emotionalism: 50% reduction in emotionalism,2 “reduced” self-reported tearfulness3 (Figure), and an 8-point reduction in scores on a Pathological Laughter and Crying scale.6 Although the point estimates were consistent with large treatment effects, the confidence intervals were wide, suggesting that the treatment may have had only a small positive effect, or in the case of the 1 trial,3 potentially a small negative effect on 2 endpoints.
Although this systematic review suggests that antidepressants reduce the frequency and severity of crying episodes among stroke patients, our conclusions are guarded by several methodological deficiencies in the studies. To begin with, there is no standard definition available for emotionalism, and no externally validated scales were used by any of the studies. In addition, multiple methods were used to report findings, both within and between trials, which made the pooling of emotionalism data inappropriate.
Trials included patients whose index stroke varied from 6 days to 13 years before randomization, and with coexistent depression, whereas the duration of treatment was short in most studies. Because the pathophysiology and response to treatment may vary with the duration from stroke onset, it may be difficult to generalize this evidence into clinical practice.
All the included trials were small, with only 1 trial2 reporting adequate concealment of the randomization sequence. Only 1 trial3 systematically recorded and reported all adverse events in the study, thus making an accurate presentation of the benefits and risks of treatment with antidepressants impossible.
In summary, our review provides suggestive, but not definitive, evidence that antidepressants can reduce the frequency of crying, sometimes even abolishing it altogether, an effect that does not seem specific to one drug or class of drugs.
Implications for Practice
Despite the lack of strong evidence, it seems reasonable to use antidepressants in a therapeutic trial in individual patients with persistent emotionalism that is sufficiently severe, and to accept the small risks associated with such agents in the elderly.
Implications for Research
Future trials need to address the following: (1) a validated scale is required for the presence and severity of emotionalism to facilitate future clinical studies; (2) studies of emotionalism should include a standard measure of depression, because it is a main confounder in assessing response to treatment; (3) trials with larger sample sizes are necessary to allow adjustment for concomitant depression and time since stroke; (4) trials with longer treatment duration would establish relapse rates or continued therapeutic benefit; (5) more careful assessment and reporting of adverse events would allow quantification of risks and benefits of treatment; and (6) trials should include 3 or 4 main outcomes for assessment, including 1 a priori primary hypothesis for testing, and should ensure that the measures are clinically relevant.
Note: The full text of this review is available in The Cochrane Library (for subscribers: www.update-softward.com/Cochrane). The full article should be cited as: House AO, Hackett ML, Anderson CS, Horrocks JA. Interventions for emotionalism after stroke (Cochrane Review). In: The Cochrane Library. Issue 2, 2004. Oxford, UK: Update Software, © Cochrane Library, John Wiley & Sons Ltd.
This review was supported in part by a grant from the Stroke Society of Australasia, with additional financial assistance provided by the Academic Unit of Psychiatry, The University of Leeds, and the Department of Clinical Neurosciences, The University of Edinburgh. M.H. held a Senior Health Research fellowship from The University of Auckland during the period of the review. The authors thank the Cochrane Stroke Group, particularly Brenda Thomas and Hazel Fraser, for assistance throughout the review process.
Section Editor: Graeme J. Hankey MD FRACP
- Received May 25, 2004.
- Accepted May 31, 2004.
Andersen G, Vestergaard K, Riis JO. Post-stroke pathological crying or emotional affect treated with citalopram a selective serotonin reuptake inhibitor. Acta Neurol Scand. 1994; 89: 151.
Ohkawa S, Mori E, Yamadori A. Treatment of pathological laughing with amitriptyline. Clin Neurol. 1989; 29: 1183–1185.In Japanese.
Parikh RM, Robinson RG, Lipsey JR, Price TR. Nortriptyline treatment of post-stroke emotional lability a double blind study. Neurology. 1989; 5 (suppl 1): 177. Abstract.