Cochrane: Library

Cochrane
Library
Cochrane Database of Systematic Reviews
Pharmaceutical interventions for emotionalism after stroke

(Review)
Allida S, House A, Hackett ML
Allida S, House A, Hackett ML.

Pharmaceutical interventions for emotionalism after stroke.
Cochrane Database of Systematic Reviews 2022, Issue 11. Art. No.: CD003690.
DOI: 10.1002/14651858.CD003690.pub5.
www.cochranelibrary.com
Pharmaceutical interventions for emotionalism after stroke (Review)

Copyright © 2022 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Cochrane Trusted evidence.
Informed decisions.
Library Better health. Cochrane Database of Systematic Reviews
TABLE OF CONTENTS
ABSTRACT..................................................................................................................................................................................................... 1
PLAIN LANGUAGE SUMMARY....................................................................................................................................................................... 2
SUMMARY OF FINDINGS.............................................................................................................................................................................. 4
BACKGROUND.............................................................................................................................................................................................. 6
OBJECTIVES.................................................................................................................................................................................................. 6
METHODS..................................................................................................................................................................................................... 6
RESULTS........................................................................................................................................................................................................ 10
Figure 1.................................................................................................................................................................................................. 11
Figure 2.................................................................................................................................................................................................. 13
Figure 3.................................................................................................................................................................................................. 14
DISCUSSION.................................................................................................................................................................................................. 16
AUTHORS' CONCLUSIONS........................................................................................................................................................................... 17
ACKNOWLEDGEMENTS................................................................................................................................................................................ 18
REFERENCES................................................................................................................................................................................................ 19
CHARACTERISTICS OF STUDIES.................................................................................................................................................................. 24
DATA AND ANALYSES.................................................................................................................................................................................... 37
Analysis 1.1. Comparison 1: Pharmaceutical interventions versus placebo, Outcome 1: Emotionalism........................................ 39
Analysis 1.2. Comparison 1: Pharmaceutical interventions versus placebo, Outcome 2: Emotionalism: mean scores at end of 39
treatment...............................................................................................................................................................................................
Analysis 1.3. Comparison 1: Pharmaceutical interventions versus placebo, Outcome 3: Depression: 1. Mean scores at end of 40
treatment...............................................................................................................................................................................................
Analysis 1.4. Comparison 1: Pharmaceutical interventions versus placebo, Outcome 4: Depression: 2. Average change in scores 40
between baseline and end of treatment.............................................................................................................................................
Analysis 1.5. Comparison 1: Pharmaceutical interventions versus placebo, Outcome 5: Cognitive functioning: mean scores at 40
end of treatment...................................................................................................................................................................................
Analysis 1.6. Comparison 1: Pharmaceutical interventions versus placebo, Outcome 6: Activities of daily living: 1. Mean scores 41
at end of treatment..............................................................................................................................................................................
Analysis 1.7. Comparison 1: Pharmaceutical interventions versus placebo, Outcome 7: Adverse events: 1. Death....................... 41
Analysis 1.8. Comparison 1: Pharmaceutical interventions versus placebo, Outcome 8: Adverse events: 2. All............................. 42
Analysis 1.9. Comparison 1: Pharmaceutical interventions versus placebo, Outcome 9: Adverse events: 3. Leaving the study 42
early (including death)..........................................................................................................................................................................
ADDITIONAL TABLES.................................................................................................................................................................................... 43
APPENDICES................................................................................................................................................................................................. 44
WHAT'S NEW................................................................................................................................................................................................. 51
HISTORY........................................................................................................................................................................................................ 51
CONTRIBUTIONS OF AUTHORS................................................................................................................................................................... 52
DECLARATIONS OF INTEREST..................................................................................................................................................................... 52
SOURCES OF SUPPORT............................................................................................................................................................................... 52
DIFFERENCES BETWEEN PROTOCOL AND REVIEW.................................................................................................................................... 52
INDEX TERMS............................................................................................................................................................................................... 53
Pharmaceutical interventions for emotionalism after stroke (Review) i

Informed decisions.
[Intervention Review]
Pharmaceutical interventions for emotionalism after stroke
Sabine Allida1, Allan House2, Maree L Hackett1
1Mental Health Program, The George Institute for Global Health, Faculty of Medicine, University of New South Wales, Sydney, Australia.
2Division of Psychological and Social Medicine, Leeds Institute of Health Sciences, University of Leeds, Leeds, UK
Contact: Maree L Hackett, [email protected].
Editorial group: Cochrane Stroke Group.

Publication status and date: New search for studies and content updated (no change to conclusions), published in Issue 11, 2022.
Citation: Allida S, House A, Hackett ML. Pharmaceutical interventions for emotionalism after stroke. Cochrane Database of Systematic
Reviews 2022, Issue 11. Art. No.: CD003690. DOI: 10.1002/14651858.CD003690.pub5.
ABSTRACT
Background
Antidepressants may be useful in the treatment of abnormal crying associated with stroke. This is an update of a Cochrane Review first
published in 2004 and last updated in 2019.
Objectives
To evaluate the benefits and harms of pharmaceutical treatment in people with emotionalism after stroke.
Search methods
We searched the Cochrane Stroke Group Register, CENTRAL, MEDLINE, Embase, four other databases, and three trials registers (May 2022).
Selection criteria
We included randomised controlled trials (RCTs) and quasi-RCTs comparing psychotropic medication to placebo in people with stroke
and emotionalism (also known as emotional lability, pathological crying or laughing, emotional incontinence, involuntary emotional
expression disorder, and pseudobulbar affect).
Data collection and analysis

Two review authors independently selected trials, assessed risk of bias, extracted data from all included trials, and used GRADE to assess
the certainty of the body of evidence. We calculated the mean difference (MD) or standardised mean difference (SMD) for continuous
data and the risk ratio (RR) for dichotomous data, with 95% confidence intervals (CIs). We assessed heterogeneity using the I2 statistic.
The primary emotionalism measures were the proportion of participants achieving at least a 50% reduction in abnormal emotional
behaviour at the end of treatment, improved score on the Center for Neurologic Study - Lability Scale (CNS-LS) or Clinician Interview-Based
Impression of Change (CIBIC), or diminished tearfulness.
Main results
We did not identify any new trials for this update. We included seven trials with a total of 239 participants. Two trials had a cross-over design,
but outcome data were not available from the first phase (precross-over) in an appropriate format for inclusion as a parallel randomised
controlled trial (RCT). Thus, the results of the review are based on five trials with a total of 213 participants. It is uncertain whether fluoxetine
increases the number of people who have a 50% reduction in emotionalism when compared to placebo (risk ratio (RR) 0.26, 95% CI 0.09
to 0.77; P = 0.02; 1 trial, 19 participants) because the certainty of evidence is very low.
Sertraline may lead to little to no difference in Center for Neurologic Study - Lability Scale (CNS-LS) scores and Clinician Interview-Based
Impression of Change (CIBIC) scores when compared to placebo (RR 0.20, 95% CI 0.03 to 1.50; P = 0.12; 1 trial, 28 participants; low-certainty
evidence). Antidepressants probably increase the number of people who experience a reduction in tearfulness (RR 0.32, 95% CI 0.12 to 0.86;
Pharmaceutical interventions for emotionalism after stroke (Review) 1
Informed decisions.
P = 0.02; 3 trials, 164 participants; moderate-certainty evidence). No trials were found that evaluated the impact of other pharmaceutical
interventions.
Only two trial authors systematically recorded and reported adverse events, resulting in limited data on the potential harms of treatment.
Six trials reported death as an adverse event and found no difference between the groups (antidepressants versus placebo) in the number
of deaths reported (RR 0.59, 95% CI 0.08 to 4.50; P = 0.61; 172 participants; moderate-certainty evidence).
This review provides very low- to moderate-certainty evidence that antidepressants may reduce the frequency and severity of
emotionalism. The included trials were small and had some degree of bias.
Authors' conclusions
Antidepressants may reduce the frequency and severity of crying or laughing episodes when compared to placebo, based on very low-
certainty evidence. Our conclusions must be qualified by several methodological deficiencies in the trials and interpreted with caution
despite the effect being very large. The effect does not seem specific to one drug or class of drugs. More reliable data are required
before appropriate conclusions can be made about the treatment of post-stroke emotionalism. Future trialists investigating the effect
of antidepressants in people with emotionalism after stroke should consider developing and using a standardised method to diagnose
emotionalism, determine severity, and assess change over time; provide treatment for a sufficient duration and follow-up to better assess
rates of relapse or maintenance; and include careful assessment and complete reporting of adverse events.
PLAIN LANGUAGE SUMMARY
Medicines for emotionalism after stroke
Review question
To evaluate the benefits and harms of medicine in people with emotionalism after stroke.
Background
Emotionalism often occurs after stroke. Emotionalism means that the person has difficulty controlling their emotional behaviour. After
stroke, people may suddenly start crying or, less commonly, laughing for no apparent reason. This is distressing for that particular person
and their carers. Antidepressants, known to be helpful in people with depression, may be an effective treatment for emotionalism after
stroke, but there have been very few randomised controlled trials (RCTs) in this area. (RCTs are studies in which participants are assigned
randomly to 2 or more treatment groups. This is the best way to ensure that groups of participants are similar, and that investigators and
participants do not know who is in which group.)
Search date
We identified studies through searches conducted on 26 May 2022.
Study characteristics
We included 7 randomised controlled trials, involving 239 people with emotionalism, in the review, which reported on the use of
antidepressants for treating emotionalism. The numbers of people included in the studies ranged from 10 to 92. The mean/median ages
ranged from 57.8 years to 73 years. Studies were from Europe (UK: 1; Denmark: 1; Scotland: 1; and Sweden: 1); Asia (South Korea: 1 and
Japan: 1); and the USA (1).
Key results
We did not identify any new studies in our search update. We included 7 studies involving 239 people. However, 2 studies were not available
in the appropriate format to include in the analysis. Therefore, we included only 5 studies, with 213 people, in our analyses. It is uncertain
whether antidepressants affect the number of people with at least a 50% reduction in emotional displays, tearfulness, and scores on the
questionnaires that measure uncontrollable laughing or crying, but they had no effect on lability (frequency of mood changes) scores and
whether a clinician thought there had been an improvement when compared to placebo. No studies of other types of medicine were found.
Six studies reported death and found no differences between those who were on antidepressants and those who were not.
What are the limitations of the evidence?
Our confidence in the evidence is moderate to very low, and the results of further research could differ from the results of this review. Our
confidence is limited because the included studies were small, and there was no consistency in how emotionalism was measured across
the studies.
Conclusion

Informed decisions.
It is uncertain whether antidepressant medicines reduce outbursts of crying or laughing. More studies with systematic assessment and
reporting of unwanted or harmful effects of a treatment are needed to ensure that any benefits outweigh the risks.

SUMMARY OF FINDINGS
Summary of findings 1. Pharmaceutical interventions compared to placebo for emotionalism after stroke
Library
Cochrane
Pharmaceutical interventions compared to placebo for emotionalism after stroke
Patient or population: emotionalism after stroke

Setting: inpatient
Intervention: pharmaceutical interventions
Better health.
Informed decisions.
Trusted evidence.
Comparison: placebo
Outcomes Anticipated absolute effects* (95% CI) Relative ef- Number of Certainty of Comments
fect participants the evidence
Risk with Risk with antidepressants (95% CI) (studies) (GRADE)
placebo
Emotionalism 50% reduction in emo- 0 per 1000 777 per 1000 RR 0.26 19 ⊕⊝⊝⊝ -
(primary out- tionalism (7 to 9) (0.09 to 0.77) (1 RCT) Very lowa,b
come)
Improved score on Cen- 643 per 1000 926 per 1000 RR 0.20 28 ⊕⊕⊝⊝ -
ter for Neurologic Study (611 to 1000) (0.03 to 1.50) (1 RCT) Lowa
- Lability Scale (CNS-LS)
Clinician Inter- 643 per 1000 926 per 1000 RR 0.20 28 ⊕⊕⊝⊝ -
view-Based Impression (611 to 1000) (0.03 to 1.50) (1 RCT) Lowa
of Change - improved
score
Diminished tearfulness 329 per 1000 718 per 1000 RR 0.32 164 ⊕⊕⊝⊝ -
(425 to 1000) (0.12 to 0.86) (3 RCTs) Moderatec
Emotionalism: mean scores at end of treatment - Pathologi- MD 8.40 lower - 28 ⊕⊕⊝⊝ (High score = worse

cal Laughter and Crying Scale (11.56 lower to 5.24 lower) (1 RCT) Lowa emotionalism)
Depression: Mean scores at end of treatment SMD 0.82 lower - 72 ⨁◯◯◯ Hamilton Depression
(2.14 lower to 0.51 higher) (2 RCTs) Very lowb,d Rating Scale; Mont-
gomery Åsberg De-
pression Rating Scale
(high score = more de-
pressed)
4
Cognitive functioning: Mean scores at end of treatment MD 0.3 lower - 28 ⨁⨁◯◯ Mini-Mental State Ex-
(3.27 lower to 2.67 higher) (1 RCT) Lowb,c amination (low score =
cognitive impairment)
Library
Cochrane
Activities of daily living: 1. Mean scores at end of treatment MD 1.4 lower - 28 ⨁⨁◯◯ Johns Hopkins Func-
(5.22 lower to 2.42 higher) (1 RCT) Lowb,c tioning Inventory (high
score = worse func-
tion)
Better health.
Informed decisions.
Trusted evidence.
Adverse events: 1. death: at end of treat- 0 per 1000 0 per 1000 RR 0.59 172 ⊕⊕⊕⊝ -
ment (2 to 92) (0.08 to 4.50) (6 RCTs) Moderateb
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and
its 95% CI).
CI: confidence interval; GRADE: Grades of Recommendation, Assessment, Development and Evaluation; MD: mean difference; RCT: randomised controlled trial; RR: risk ra-
tio; SMD: standardised mean difference
GRADE Working Group grades of evidence

High certainty: we are very confident that the true effect lies close to that of the estimate of the effect.
Moderate certainty: we are moderately confident in the effect estimate; the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is
substantially different.
Low certainty: our confidence in the effect estimate is limited; the true effect may be substantially different from the estimate of the effect.
Very low certainty: we have very little confidence in the effect estimate; the true effect is likely to be substantially different from the estimate of effect.
aWe downgraded the certainty of evidence as < 50 participants in total, and only 1 study contributed to the analysis.
bWe downgraded the certainty of evidence due to imprecision.
cWe downgraded the certainty of evidence as we rated one of the studies at high risk for attrition bias.
dHeterogeneity is considerable (I2 statistic = 84%).

5
Informed decisions.
BACKGROUND and also to lithium (Massey 1981), L-dopa (Udaka 1984; Wolf 1997),
and venlafaxine (Smith 2003).
Description of the condition
How the intervention might work
Disturbances of emotional behaviour, such as difficulty controlling
crying or laughing, are common after stroke (House 1989). Poeck Classic tricyclic antidepressants commonly used to treat
1969 distinguished two main types of disturbed emotionality emotionalism are thought to block the reuptake of monoamine
associated with brain lesions: one he called pathological crying and neurotransmitters, such as serotonin and norepinephrine.
laughing, and the other, emotional lability. The main differences Selective serotonin reuptake inhibitors (SSRIs) only reduce the
were that in the former, provocation was by non-emotive or reuptake of serotonin, which in turn increases serotonin activity,
incongruous stimuli and the emotional display was socially and have little effect on other neurotransmitters, such as
abnormal and unstable, while in the latter, emotional behaviour norepinephrine. However, the effect of these drugs on the disorder
was more socially familiar and provoked by typically emotive of emotionality remains unclear. It does not appear to be the result
stimuli. The terminology is used inconsistently in the literature of a simple antidepressant action, although amelioration of co-
(Allman 1989), and the evidence to support Poeck's dichotomy is existing depression, which is exacerbating the emotionalism, may
not strong. In order to avoid prejudging the issue, we have preferred be the mechanism of action for some. Recovery occurs in people
a general term for all such disorders of emotionality and have without a depressive disorder; at times, it occurs in a dramatic
called the problem 'emotionalism': the habit of weakly yielding to fashion, within 24 to 48 hours of starting a low dose, and abnormal
emotion (House 1989). laughter may also respond to treatment (Lauterbach 1991; Schiffer
1983). There are case reports suggesting that withdrawal of
The essential feature of emotionalism is an increase in emotional antidepressants leads to re-emergence of emotionalism, while
behaviour (usually crying, but sometimes laughing) that the person reinstatement leads to resolution (Schiffer 1983; Seliger 1989).
reports as being outside normal control, so that they cry or However, drug treatment is not always effective and may be
laugh in situations that would not previously have provoked such complicated by common unacceptable side effects. We have been
behaviour. Onset of episodes is often reported as being more able to find two case reports of psychological therapy (Brookshire
sudden and unpredictable than usual, but even so, most report 1970; Sacco 2008).
precipitants, which are usually congruent with their emotional
response (House 1989). Why it is important to do this review
There are other disturbances of emotional behaviour after stroke, Emotionalism is an under-recognised and under-treated condition
especially irritable or anxious behaviours. By convention, however, that adversely affects individuals post-stroke worldwide.
they are not included in the present category, which is restricted Treatment options include the off-label use of a range of
to crying and laughing. Emotionalism is associated with an antidepressants. However, there are uncertainties surrounding the
increase in depressive symptoms. Nonetheless, most people with benefits versus the risks associated with their use. We undertook
emotionalism do not have a diagnosable depressive disorder, and a systematic review of all randomised controlled trials (RCTs) and
many do not have significant depressive symptoms at all (Calvert quasi-RCTs (published) of pharmaceutical agents for the treatment
1998; Kim 2000). of emotionalism associated with stroke. We did not find any RCTs
evaluating non-drug (psychological) interventions; we will consider
The reported prevalence of emotionalism varies across trials. any future trials in this area in a separate review.
Results of one systematic review, which included 15 English-
language trials involving 3391 participants, indicated that OBJECTIVES
emotionalism affects about 17% of those who have experienced
a strokein the first month, 20% between one and six months, and To evaluate the benefits and harms of pharmaceutical treatment in
12% more than six months after stroke (Gillespie 2016). It tends to people with emotionalism after stroke.
decline in frequency and severity over the first year, and a few have
METHODS
persistent severe problems. People with emotionalism describe
distress and embarrassment and thereby social avoidance and Criteria for considering studies for this review
impaired quality of contact with friends and family.
Types of studies
Description of the intervention
We considered all relevant randomised controlled trials (RCTs) in
Until recently, there had been no U.S. Food and Drug Administration people with a clinical diagnosis of stroke where a pharmaceutical
(FDA)-approved drug to treat emotionalism. However, there are agent used specifically for the treatment of emotionalism was
several classes of drugs consistently used by clinicians in an off- compared with placebo. We excluded trials in which the allocation
label manner. These drugs are mainly used to treat various central to treatment or placebo was not random. We excluded trials that
nervous system conditions, targeting serotonin, monoamine, or compared two or more antidepressant drugs but did not include
dopamine receptors. There are case reports and case series a placebo group. We identified RCTs and cross-over trials in all
suggesting that disorders of emotionality (variously named and languages. There was no restriction on eligibility of RCTs on the
defined) may respond to mirtazapine (Kim 2005), imipramine basis of sample size, duration of follow-up, or publication status.
(Allman 1992a), amitriptyline (Schiffer 1983), doxepin (Schiffer
1983), nomifensine (Sandyk 1985), fluoxetine (Hanger 1993; Nahas When we found trials that met all criteria for inclusion but did
1998; Panzer 1992; Sloan 1992), sertraline (Benedek 1995; Mukand not present any outcome data (and such data were not available
1996; Muller 1999; Nahas 1998; Tan 1996), paroxetine (Muller 1999), from the authors) and could therefore not contribute to any pooled
estimate of effect, we regarded these trials as 'dropouts' rather than
Informed decisions.
ineligible, and they are listed in Table 1 to indicate that we did not • Improved score on the Center for Neurologic Study - Lability
overlook these trials. Scale (CNS-LS) (Moore 1997).
• Improved score on the Testing Emotionalism After Recent Stroke
Types of participants – Questionnaire (TEARS-Q) (Broomfield 2021).
All participants had to have established emotionalism at the time • Clinician Interview-Based Impression of Change (CIBIC)
of entry into the trial. The essential feature of emotionalism is an (Knopman 1994).
increase in emotional behaviour (usually crying, but sometimes • Diminished tearfulness.
laughing) that the person reports as being outside normal control,
so that they cry or laugh in situations that would not previously Secondary outcomes
have provoked such behaviour. Onset of episodes is often reported
• Emotionalism: mean scores at end of treatment. There are
as being more sudden and unpredictable than usual, but even so,
a number of measures reported in the published literature
most people report precipitants, which are usually congruent with
(Allman 1992b; Lawson 1969; Moore 1997; Newsom-Davis
their emotional response.
1999; Robinson 1993a), but similar to frequency of emotional
We included all participants with a confirmed history of stroke, behaviour, there is no widely accepted standardised measure
at any stage after onset, where there was an explicit intention to of severity of emotionalism that could be used to measure this
provide a pharmacological agent to treat emotionalism associated outcome as a continuous variable
with stroke. We defined stroke according to standard clinical • Depression: mean scores at end of treatment
criteria: the criteria included cerebral infarction, intracerebral • Depression: average change in scores between baseline
haemorrhage, and uncertain pathological subtypes, but excluded and end of treatment. Depression as measured on scales
subarachnoid haemorrhage, which has a different natural history such as the Hamilton Depression Rating Scale (HDRS)
and management strategy to other stroke subtypes. (Hamilton 1960), Montgomery Åsberg Depression Rating Scale
(MADRS) (Montgomery 1979), Geriatric Depression Scale (GDS)
There were no restrictions on the basis of age, sex, or other (Gompertz 1993), Beck Depression Inventory (BDI) (Beck
characteristics. We excluded trials that included mixed populations 1961), and Hospital Anxiety and Depression Scale (Depression
(such as stroke and head injury or other central nervous system subscale, HADS-D) (Zigmond 1983), or as measured on
disorders) unless we could identify separate results for those with composite scales, such as the General Health Questionnaire
stroke. We included people with a diagnosed mood disorder or a (GHQ) (Goldberg 1972)
mood score above the standard cut-off scores at baseline, provided
• Cognitive functioning: mean scores at end of treatment.
it was clear that they also met criteria for emotionalism. We
Cognition as measured on scales such as the Mini-Mental State
excluded those who were being treated primarily for a stroke-
Examination (MMSE) (Folstein 1975)
associated pain syndrome or for stroke-associated depression,
even if emotionalism was measured as a secondary (post-hoc) • Activities of daily living: mean scores at end of treatment.
outcome. Activities of daily living as measured on scales such as the
Barthel Index (BI) (Mahoney 1965)
Types of interventions • Disadvantages of treatment were recorded as:
We included any trial that attempted to evaluate a ◦ adverse events: death;
comparison between a pharmacological agent and placebo ◦ adverse events: all; and
for the treatment of emotionalism following stroke. Specific ◦ adverse events: leaving the study early (including death).
pharmacological agents that we considered included the following: • We identified additional endpoints for use in further reviews,
tricyclic antidepressants (e.g. nortriptyline, imipramine, and where measured.
clomipramine), selective serotonin reuptake inhibitors (SSRIs) (e.g. ◦ Proportion who scored above accepted cut-offs for
fluvoxamine, fluoxetine, citalopram, sertraline, and paroxetine), identifying mood disorders, using mood-rating scales.
monoamine oxidase inhibitors (MAOIs) (e.g. moclobemide), ◦ Proportion who met the standard psychiatric diagnostic
and other antidepressant medications. We found no trials of criteria for major depression or dysthymia (Diagnostic and
psychostimulants (e.g. methylphenidate), mood stabilisers (e.g. Statistical Manual of Mental Disorders (DSM): DSM-IIIR, DSM-
lithium), benzodiazepines, or combined preparations. We will IV) (APA 1987; APA 1994).
include future trials of psychostimulants but will analyse them
separately. Search methods for identification of studies
Types of outcome measures This review is an update of a previously published Cochrane Review
(Allida 2019). The first version of this review was published in 2004
Primary outcomes
(House 2004). For this update, we only searched the Specialised
Emotionalism: despite widespread acknowledgement of the Register of Cochrane Stroke and the Cochrane Central Register of
importance of the problem, there is no widely accepted Controlled trials (CENTRAL) from 2008 until May 2022.
standardised set of diagnostic criteria for emotionalism. The
primary endpoint emotionalism measures were as follows: Electronic searches
Cochrane Stroke's Information Specialist searched the following
• The proportion of participants achieving at least a 50%
databases:
reduction in abnormal emotional behaviour at the end of
treatment. • Cochrane Stroke Group Register;

Informed decisions.
• Cochrane Central Register of Controlled Trials (CENTRAL; 2021, Data extraction and management
Issue 7) in the Cochrane Library (searched 26 May 2022)
Two review authors (SA, MH) independently extracted study
(Appendix 1);
characteristics and outcome data from included trials on specially
• MEDLINE Ovid (1966 to 26 May 2022) (Appendix 2); designed forms. We cross-checked and entered the data into
• Embase Ovid (1980 to 26 May 2022) (Appendix 3); Review Manager 5 (Review Manager). We obtained missing
• CINAHL EBSCO (Cumulative Index to Nursing and Allied Health information from the study authors when possible. We resolved
Literature; 1982 to 26 May 2022) (Appendix 4); disagreements by discussion or through consultation with a third
• PsycINFO Ovid (1967 to 26 May 2022) (Appendix 5); review author (AH). If outcome data were not reported in a usable
• BIOSIS Previews (Web of Science) (January 2002 to 26 May 2022) way, we reported this in the notes in the Characteristics of included
(Appendix 6); and studies table.
• Science Citation Index Expanded (SCI-EXPANDED), Social We collected data on the following:
Sciences Citation Index (SSCI), and Arts & Humanities Citation
Index (A&HCI) within Web of Science (January 2002 to 26 May • the report: author, year, and source of publication;
2022) (Appendix 7). • the study: sample characteristics, social demography, definition
and criteria used for emotionalism;
We also searched the following resources using "emotion" or
• the participants: stroke sequence (first-ever versus recurrent),
"laughing" or "tearful" or "pseudobulbar affect" and "stroke" or
social situation, time since stroke onset, history of psychiatric
"cerebral hemorrhage" or "brain ischemia" from inception.
illness, current neurological status, current treatment for
• US National Institutes of Health Ongoing Trials Register depression, coronary artery disease;
ClinicalTrials.gov (www.clinicaltrials.gov; searched 26 May • the research design and features: sampling mechanism,
2022); treatment assignment mechanism, adherence rates, non-
• World Health Organization International Clinical Trials Registry response rates, length of follow-up;
Platform (apps.who.int/trialsearch; searched 26 May 2022); and • the intervention: type, duration, dose, timing, mode of delivery;
• ProQuest Dissertations & Theses Global database and
(www.proquest.com; searched 26 May 2022). • the effect size: sample size, nature of outcome, estimate, and
standard error.
The full search strategies for other resources are in Appendix 8.
To allow an intention-to-treat analysis, we sought data irrespective
Searching other resources of adherence, whether the participants were subsequently deemed
Personal communication ineligible or otherwise excluded from treatment or follow-up.
We contacted the study authors for information on ongoing trials or We checked all of the extracted data for agreement between review
to request additional study data. authors. To avoid introducing bias, we requested this unpublished
information in writing then entered it into Review Manager 5
Reference lists (Review Manager).
We searched the reference lists of relevant trials and systematic Assessment of risk of bias in included studies
reviews and reviewed chapters in books on the prevention and
treatment of depression and management of stroke, including Two review authors (SA, MH) independently assessed risk of bias for
but not limited to, reviews of the management of stroke, books each study using the criteria outlined in the Cochrane Handbook for
specifically directed at the treatment or prevention of depression, Systematic Reviews of Interventions (Higgins 2017). We resolved any
and those on stroke and old age. disagreements by discussion or by involving another author (AH).
We assessed the risk of bias according to the following domains.
Data collection and analysis
• Random sequence generation.
Selection of studies
• Allocation concealment.
Two review authors (SA, MH) discarded irrelevant citations based • Blinding of participants and personnel.
on the title of the publication and its abstract. If we thought • Blinding of outcome assessment.
that an article could possibly be relevant, we retrieved the full-
• Incomplete outcome data.
length article for further assessment. Two review authors (SA, MH)
independently selected the trials for inclusion in the review from • Selective outcome reporting.
the culled citation list. We obtained translations of potentially • Other bias.
relevant non-English language articles. We resolved disagreements
by discussion, and one review author (AH) confirmed the final In accordance with RoB 1, we graded potential sources of bias as
list and adjudicated any persisting differences. We present the high, low, or unclear and provided a quote from the study report
selection process in a PRISMA flow diagram (Page 2021). We listed together with justification for our judgement in the risk of bias
the included trials under Characteristics of included studies, and we table. We summarised the risk of bias judgements across different
listed the trials that we ultimately excluded under Characteristics of trials for each of the domains listed.
excluded studies and provided the primary reasons for exclusion.
When considering treatment effects, we took into account the risk
of bias for the trials that contributed to that outcome.

Informed decisions.
Measures of treatment effect Data synthesis

Dichotomous data We analysed data using Review Manager 5 and pooled data for
meta-analysis when trials assessed similar treatment and had
The primary outcomes of interest were the proportion of
similar outcomes (Review Manager). We conducted a meta-analysis
participants with improved emotionalism scores (reduced
using available or calculated SMDs for continuous outcomes, and
symptoms) at the end of follow-up on the CNS-LS (Moore 1997),
risk ratios (RRs) for dichotomous outcomes. We included in the
TEARS-Q (Broomfield 2021), and CIBIC (Knopman 1994), and
results measures of uncertainty, such as 95% CIs and estimates of
diminished tearfulness, and the proportion who showed a 50%
the I2 statistic.
reduction in abnormal emotional behaviour (crying or laughing) at
the end of treatment. For all dichotomous outcomes, we calculated Subgroup analysis and investigation of heterogeneity
Mantel-Haenszel risk ratios (RRs) with 95% confidence intervals
(CIs) where appropriate, using random-effects analyses. If there were at least two trials that reported the same outcomes,
we reviewed the data for appropriateness of pooling. If the
Continuous data heterogeneity remained high, we combined the trials using
random-effects analyses with cautious interpretation, or did not
For continuous outcomes, if ordinal scale data appeared to be
combine them at all.
approximately normally distributed or if the analysis suggested
parametric tests were appropriate, we treated the outcome Sensitivity analysis
measures as continuous. If there were at least two trials that
reported the same outcomes, then we calculated a mean difference We explored the sensitivity of the combined estimate to individual
(MD) across the trials. Where different outcome measures were trials by leaving one study out due to high risk of bias and
used, we planned to calculate a standardised mean difference methodological differences. We then calculated the combined
(SMD). effect of the remaining trials and compared the results with the
combined effect based on all the trials.
Unit of analysis issues
Summary of findings and assessment of the certainty of the
No unit of analysis issues were anticipated in the trials we included. evidence
Dealing with missing data We created a summary of findings table using the primary
outcome emotionalism, measured as the proportion of participants
We wrote to the authors of all included trials requesting data
achieving at least a 50% reduction in abnormal emotional
that were unavailable or ambiguous in the published articles.
behaviour at the end of treatment, improved score on the CNS-
Three authors responded with the requested additional data
LS, improved CIBIC, and diminished tearfulness. Other outcomes
(Andersen 1993; Murray 2005; Robinson 1993b). Another study
included depression, cognitive functioning, activities of daily living,
author responded stating that the requested additional data were
and adverse events.
unavailable (Burns 1999). We did not receive responses from the
remaining study authors. We used the five GRADE considerations (study limitations,
consistency of effect, imprecision, indirectness, and publication
Assessment of heterogeneity
bias) to assess the certainty of the body of evidence as it
We assessed clinical and methodological heterogeneity by relates to the trials that contributed data to the meta-analyses
examining the study characteristics of trials. We used the I2 statistic for the prespecified outcomes. We used the methods and
to measure heterogeneity amongst the trials in each analysis recommendations described in Chapter 12 of the Cochrane
(Deeks 2021). We interpreted the heterogeneity of each analysis Handbook for Systematic Reviews of Interventions (Schünemann
based on the following cut-offs: 2019), and we created the tables using GRADEpro GDT software
(GRADEpro GDT).
• 0% to 40%: might not be important;
• 30% to 60%: may represent moderate heterogeneity; By treating all pharmaceutical interventions as one comparator,
we pooled the data from the trials comparing pharmaceutical
• 50% to 90%: may represent substantial heterogeneity; and
interventions to placebo into a single meta-analysis and reported
• 75% to 100%: considerable heterogeneity. these in Summary of findings 1. We documented in the footnotes
of the tables all decisions to downgrade the certainty of trials, and
If there was definite evidence of heterogeneity (I2 statistic >
we made comments to aid the reader's understanding of the review
50%), we explored the potential reasons for the differences by
where necessary.
performing subgroup analyses. We reported similarities between
interventions, participants, design, and outcomes in the Included An experienced review author (SA) made judgements about the
studies subsection. certainty of the evidence; another author (MLH) independently
checked these judgements, and a third author (AH) resolved any
Assessment of reporting biases
disagreements. We justified, documented, and incorporated these
We did not assess publication bias as fewer than 10 trials were judgements into our reporting of results for each outcome.
available. We attempted to avoid language bias by including trials
irrespective of language of publication.

Informed decisions.
RESULTS duplicates, we screened 5422 titles and abstracts and excluded

5377 irrelevant records. We retrieved full-text reports for the
Description of studies remaining 45 trials. After reading the full texts, we excluded 35 trials
as they did not meet the review eligibility criteria (see Figure 1). We
Results of the search
have provided primary reasons for exclusion in the Characteristics
In total, we identified 6754 records; of these, we retrieved 6202 of excluded studies tables and in Figure 1. Although one trial met
records through database searching. We found 541 additional the inclusion criteria, baseline evaluation of the trial occurred at
references by searching other resources. After removing 1332 one month post-treatment (Kim 2017b); we considered this trial a
'dropout' (Table 1).

Informed decisions.
Figure 1.

Informed decisions.
Figure 1. (Continued)
In the previous published version of this review (Hackett 2010), we treatment with a 'stomach medicine' to disguise taste and smell,
identified two trials that met the inclusion criteria (Aizawa 1977; with the control group receiving the stomach medicine only
Ohtomo 1985). However, both trials included participants with (Ohkawa 1989).
cerebral arteriosclerosis, and neither presented outcome data by
those diagnosed with emotionalism at entry. We considered these Outcomes
two trials as 'dropouts'. See Table 1 for more detailed information Primary outcome: emotionalism
on these trials.
No standard criteria were used to define emotionalism at
Included studies entry across the trials. Emotionalism was measured in seven
different ways in the seven trials (Andersen 1993; Brown
From the previous update of this review, there were a total of seven 1998; Burns 1999; Choi-Kwon 2006; Murray 2005; Ohkawa 1989;
included trials with 239 participants. Two trials had a cross-over Robinson 1993b), and no more than two trials used the same
design (Andersen 1993; Ohkawa 1989), and outcome data were method of assessment. Andersen 1993 and Brown 1998 assessed
not available from the first phase (precross-over) in an appropriate emotionalism using a semi-structured interview modified from
format for inclusion as a parallel randomised controlled trial (RCT). Lawson and Macleod (Lawson 1969). In Burns 1999, the presence
Therefore, this review primarily reports data from five trials with a or absence of emotionalism was assessed using seven questions
total of 213 participants (see Characteristics of included studies). Of based on the trial by House and colleagues (House 1989). Other
these, we had included five trials in the first version of this review trials confirmed the presence of emotionalism through clinical
in 2004 (Andersen 1993; Brown 1998; Burns 1999; Ohkawa 1989; diagnosis made by a psychiatrist (Robinson 1993b), or asking the
Robinson 1993b), and we added two trials to the previous update participants and their relatives whether increased tearfulness or
in 2010 (Choi-Kwon 2006; Murray 2005). For this present review, we inappropriate laughing has occurred at any time (Choi-Kwon 2006),
found no new trials. or on more than two occasions (Murray 2005).
Participants Secondary outcomes
All trials in this review included men and women. The mean or A variety of additional outcomes were assessed in each trial. Several
median age of participants ranged from 57.8 to 73 years. Five trials trials assessed, but did not report, outcome data for depression
reported the time between stroke and randomisation into the trial, (Andersen 1993; Brown 1998; Burns 1999; Ohkawa 1989), activities
with recruitment occurring in and out of hospital and the range of daily living (Burns 1999), and cognitive functioning (Burns 1999;
covering three days to 13 years (Andersen 1993; Burns 1999; Murray Ohkawa 1989). Only three trials reported that they systematically
2005; Ohkawa 1989; Robinson 1993b). measured and reported adverse events (Burns 1999; Choi-Kwon
2006; Murray 2005).
Interventions
Five trials assessed the efficacy of selective serotonin reuptake Funding sources
inhibitors (SSRIs): citalopram (Andersen 1993), fluoxetine (Brown Pharmaceutical companies provided the antidepressant and
1998; Choi-Kwon 2006), and sertraline (Burns 1999; Murray 2005); placebo in four trials (Lundbeck (Andersen 1993), Pfizer (Brown
two assessed tricyclic antidepressants: amitriptyline (Ohkawa 1998; Murray 2005), and Sandoz (Robinson 1993b)) and additional
1989), and nortriptyline (Robinson 1993b). Duration of treatment funding, including statistical advice, for Andersen 1993. Funding
ranged from 10 to 182 days. from the Ministry of Health supported a further trial (Choi-Kwon
2006), and the source of funding was unclear for two trials (Brown
Comparator intervention(s)
1998; Ohkawa 1989).
Only one trial did not compare the active intervention with
a placebo-matched control, but instead combined the active

Informed decisions.
Excluded studies (Lawson 1969), and the trial assessed prevention rather than
treatment of emotionalism (NCT01278498 2011).
We have listed the 35 excluded trials and the reasons for their
exclusion in the Characteristics of excluded studies tables. The Ongoing studies
main reasons for exclusion were no placebo used (Atarashi 1988;
Bassi 1984; Muller 1999), participants had not experienced stroke We identified no ongoing trials.
or the number with stroke was unclear (Aizawa 1977; Ohtomo
Studies awaiting classification
1985; Schiffer 1985; Seliger 1992; Udaka 1984), participants had
emotionalism and dementia or Alzheimer's (Doody 2014; Sauve We identified no trials awaiting classification.
2017), not a RCT (Allen 2018; Chen 2010; Colamonico 2012; Kim
2017a; Manzo 1998; NCT01799941 2013; Schoedel 2014; Work New studies found for inclusion in this update
2011; Yang 2015), open-label trial (D'Amico 2017; Formella 2017a; We identified no new trials in this update.
Formella 2017b), emotionalism was not investigated either at
baseline (EUCTR2004-004803-39-DE 2006; EUCTR2011-005541-12- Risk of bias in included studies
DK 2012; EUCTR2011-006130-16-SE 2014; EUCTR2013-002253-30-
DK 2013; Kim 2017b; Lorwatanapongsa 2020; Rasmussen 2000) or In Figure 2, we present a graphical summary of the risk of bias
at end of treatment (Moller 2007; Narushima 2002; Otomo 1984), assessments we performed for the seven included trials (based on
baseline evaluation of emotionalism occurred at one month post- the seven risk of bias domains). Figure 3 provides a summary of
treatment (Kim 2017b), method of randomisation was ineffectual risk of bias for each included trial. We have provided the reasons
for judgements in the risk of bias tables. For clarification, we have
provided quotes in these tables.
Figure 2. Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages
across all included studies.
Random sequence generation (selection bias)

Allocation concealment (selection bias)
Blinding of participants and personnel (performance bias): All outcomes
Blinding of outcome assessment (detection bias): All outcomes
Incomplete outcome data (attrition bias): All outcomes
Selective reporting (reporting bias)
Other bias
0% 25% 50% 75% 100%
Low risk of bias Unclear risk of bias High risk of bias

Informed decisions.
Figure 3. Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
Blinding of participants and personnel (performance bias): All outcomes

Blinding of outcome assessment (detection bias): All outcomes
Incomplete outcome data (attrition bias): All outcomes
Random sequence generation (selection bias)
Allocation concealment (selection bias)
Selective reporting (reporting bias)

Other bias
Andersen 1993 + + ? ? + + ?
Brown 1998 ? + + + + + ?
Burns 1999 + ? ? ? + + ?
Choi-Kwon 2006 + ? + + − + ?
Murray 2005 + + ? ? + + ?
Ohkawa 1989 ? ? + + + + ?
Robinson 1993b + + + + + + ?

Informed decisions.
Allocation improvement (reduction) in tearfulness when compared to placebo

(RR 0.32, 95% CI 0.12 to 0.86; P = 0.02; 3 RCTs, 164 participants;
We rated three trials as having a low risk of selection bias and four as
moderate-certainty evidence; Burns 1999; Choi-Kwon 2006; Murray
having an unclear risk. Two of the trials did not report the method
2005; Analysis 1.1, subgroup 1.1.4).
used for sequence generation (Brown 1998; Ohkawa 1989). One
trial randomised participants using a random number allocation Sertraline may have little to no effect on the Center for Neurologic
list (Burns 1999), and three trials randomised participants using Study - Lability Scale (CNS-LS) and Clinician Interview-Based
a computer-generated number sequence (Andersen 1993; Choi- Impression of Change (CIBIC) scores compared to placebo (RR 0.20,
Kwon 2006; Murray 2005). One trial administered the intervention 95% CI 0.03 to 1.50; P = 0.12; 1 RCT, 28 participants) because the
after random number assignment (Robinson 1993b). Four trials certainty of the evidence is low (Burns 1999; Analysis 1.1). The point
clearly reported allocation concealment (Andersen 1993; Brown estimates were consistent with large treatment effects for all five
1998; Murray 2005; Robinson 1993b). trials; however, the CIs were wide for three trials (Brown 1998; Burns
1999; Robinson 1993b), so it is possible that the treatment may
Blinding
have had only a small beneficial effect, or in the case of one trial
We assessed four trials as having a low risk of performance (Burns 1999), the treatment may even have had a small harmful
and detection bias (Brown 1998; Choi-Kwon 2006; Ohkawa 1989; effect on two endpoints (see subgroups 1.1.2 and 1.1.4).
Robinson 1993b). Although three trials reported that they were
double-blind, they failed to report exactly who was blinded Secondary outcomes
(Andersen 1993; Burns 1999; Murray 2005). Emotionalism scores
Incomplete outcome data It is uncertain whether nortriptyline lowers scores (fewer
emotionalism symptoms) on the Pathological Laughter and Crying
We judged six of the trials as having a low risk of bias for outcome
Scale (PLCS) when compared with placebo (mean difference (MD)
reporting (Andersen 1993; Brown 1998; Burns 1999; Murray 2005;
-8.40, 95% CI -11.56 to -5.24; P < 0.00001; 1 RCT, 28 participants)
Ohkawa 1989; Robinson 1993b). One trial was at high risk, as the
because the certainty of the evidence is low (Robinson 1993b;
total number of participants stated was inconsistent. We rated trials
Analysis 1.2, subgroup 1.2.1).
as having a high risk of bias if dropout rates were uneven between
groups and the reason for dropout was suspected to be related to Depression
group allocation. We also rated trials as high risk if investigators
did not report how dropout was dealt with (e.g. intention-to-treat The moderate reduction in depression that was observed between
analysis and last observation carried forward). treatment groups at the end of one trial (Robinson 1993b; Analysis
1.3, subgroup 1.3.1) has been mediated by the addition of a
Selective reporting second trial (Murray 2005), and is not apparent after controlling for
differences in depression between groups at baseline (standardised
We rated risk of bias from selective reporting as low risk for all the
mean difference (SMD) 0.82, 95% CI 2.14 to 0.51; P = 0.01; 2 RCTs,
trials (Andersen 1993; Brown 1998; Burns 1999; Choi-Kwon 2006;
72 participants; very low-certainty evidence) (see Analysis 1.4,
Murray 2005; Ohkawa 1989; Robinson 1993b).
subgroup 1.4.1, where the CI includes 0 and the possibility of a
Other potential sources of bias negative treatment effect).
We judged the risk of other potential sources of bias as Cognitive functioning

unclear for all the trials as trial authors either did not report
One trial provided data concerning this outcome; nortriptyline had
the source of funding or the reported funding source was a
no effect on cognitive functioning when compared with placebo
pharmaceutical company with no associated statement indicating
(MD 0.30, 95% CI -3.27 to 2.67; P = 0.84; 1 RCT, 28 participants; low-
their involvement in trial conduct or analysis.
certainty evidence; Robinson 1993b; Analysis 1.5).
Effects of interventions Activities of daily living
See: Summary of findings 1 Pharmaceutical interventions One trial addressed activities of daily living and found no treatment
compared to placebo for emotionalism after stroke effect on activities of daily living (MD 1.40, 95% CI -5.22 to 2.42;
P = 0.47; 1 RCT, 28 participants; low-certainty evidence; Robinson
For details of the comparisons made for trials with outcome data,
1993b; Analysis 1.6).
please refer to the Data and analyses section. Labelling of the x-axis
on the forest plots changes because of variation in the method of Disadvantages of treatment
measurement and direction of the outcome scale.
Two trials systematically recorded and reported adverse events
Primary outcome (Burns 1999; Murray 2005). Other trials provided selected data on
some adverse events (Robinson 1993b), or those leaving the trial
Emotionalism
early (Brown 1998; Choi-Kwon 2006).
It is uncertain whether antidepressants (fluoxetine) increase the
number of people who have a 50% reduction in emotionalism Death
when compared to placebo (risk ratio (RR) 0.26, 95% confidence We saw no differences between groups with the six trials reporting
interval (CI) 0.09 to 0.77; P = 0.02; 1 randomised controlled trial deaths (RR 0.59, 95% CI 0.08 to 4.50; P = 0.61; 6 RCTs, 172
(RCT), 19 participants) because the certainty of the evidence is very participants; moderate-certainty evidence; Andersen 1993; Brown
low (Brown 1998; Analysis 1.1, subgroup 1.1.1). We observed an 1998; Burns 1999; Murray 2005; Ohkawa 1989; Robinson 1993b;
Informed decisions.
Analysis 1.7). Deaths occurred in two trials: one in each group individuals. This review included trials with participants whose
(Burns 1999), and one death only in the placebo group (Murray index stroke varied from six days to 13 years before randomisation.
2005). It may not be appropriate to consider that the response to
treatment is consistent across such mixed populations, as the
All adverse events aetiology (and underlying pathology) of emotionalism may differ
We saw no difference in central nervous system events, between people early on after acute stroke and in the long term.
gastrointestinal effects, or other adverse events between groups. For example, people in the acute phase post-stroke have recently
Confidence intervals were extremely wide and all included unity experienced a potentially life-threatening event and are coping
(see Analysis 1.8). with the psychological consequences, as well as recovering from
the disabling effects of the stroke itself. On the other hand,
Leaving the trial early people who had a stroke several months previously are adjusting
Two trials showed that participants who were allocated active to the prospects of long-term disability and changes in social
treatment were less likely to leave the trials early (Choi-Kwon 2006; and financial circumstances, as well as the cumulative effects of
Murray 2005); three other trials showed that participants who were cerebrovascular disease or vascular dementia, or both. The natural
allocated active treatment were more likely to leave the trials early, history of emotionalism after stroke is for spontaneous resolution
giving a pooled estimate of no effect. However, CIs were extremely over a few months, whereas it is reasonable to suppose that it
wide (Brown 1998; Burns 1999; Choi-Kwon 2006; Robinson 1993b), may be a more chronic state in those with chronic cerebrovascular
and all included unity (see Analysis 1.9). disease. Therefore, since the balance of risks and benefits and the
effectiveness of treatment for emotionalism may change with time
DISCUSSION from the onset of stroke, mixing individuals at very different stages
after stroke makes interpretation difficult. Future trials should
Summary of main results include homogeneous groups of participants with respect to time
from the onset of stroke, or sufficient numbers of participants in the
This review provides very low- to moderate-certainty evidence early and late stages after stroke.
that antidepressants may reduce the frequency and severity of
crying episodes. No harms and no trials of other pharmaceutical Defining the disorder of interest is key to the conduct of a clinical
interventions were identified. We have downgraded the certainty trial. A widely agreed definition usually exists for most clinical
of the evidence due to several methodological deficiencies. conditions. However, no such standard definition is available
These deficiencies include the type of participants included, the for emotionalism. For example, it does not feature in a clearly
definition and diagnosis of emotionalism used, the inclusion of defined way in the Diagnostic and Statistical Manual of Mental
some comorbidities, the small number of trials and participants Disorders (DSM) (APA 1987; APA 1994; APA 2017), or International
contributing to most endpoints, and the generally poor trial design Classification of Diseases (ICD) (ICD 10). In the clinical setting,
and reporting of results. the diagnosis is generally made during interview. The commonly
accepted criteria for emotionalism include: 1) sudden onset of
In summary, these trials appear to add little to case reports and case crying (and less commonly, laughing); 2) not under usual control (a
series. They provide suggestive, but not definitive, evidence that change in behaviour has occurred); and 3) the crying is not simply
antidepressants can reduce the frequency of crying (sometimes an expression of depression or grief. Given the resource-intensive
abolishing it altogether). The effect does not seem specific to one nature of conducting psychiatric interviews on all participants in
drug or class of drugs. clinical trials, we considered it acceptable to determine caseness
during a psychiatric interview and to measure frequency and
Overall completeness and applicability of evidence severity using a validated questionnaire. Two standardised scales
The present review included seven trials with a total of 239 were used to assess emotionalism: the Pathological Laughter and
participants (Andersen 1993; Brown 1998; Burns 1999; Choi-Kwon Crying Scale (PLCS) (Robinson 1993b), and the Lawson and Macleod
2006; Murray 2005; Ohkawa 1989; Robinson 1993b). Two trials had Scale (Lawson 1969). An attempt to validate the PLCS was made
a cross-over design (Andersen 1993; Ohkawa 1989), and outcome (Robinson 1993b), but neither scale has been externally validated
data were not available from the first phase (precross-over) in using traditional methods. The 'severity' score on the PLCS
an appropriate format for inclusion as a parallel randomised includes items recording the quality of crying, not just frequency
controlled trial (RCT). Thus, the results of the review were based of occurrence. In the absence of a validated questionnaire, the
on five trials with a total of 213 participants. Overall, there were most appropriate method to diagnose and determine severity
no standard criteria for defining emotionalism at entry and no of emotionalism is likely to be a simple and easily replicable
standard measures of emotionalism across all of the included trials. assessment of the frequency of crying episodes, or laughing
We considered three trials as 'dropouts' as outcome data were not episodes (assessing these elements separately), combined with
available in the format appropriate for a parallel RCT and baseline an a priori cut-off score for entry into the trial. The nature of
was completed post-treatment. As a result, there were a small precipitants should be assessed in separate questions to those
number of trials and participants contributing to most endpoints. assessing frequency of crying and laughing.
The accuracy of the findings of this systematic review and meta-
analysis is based on the studies that met the eligibility criteria. We Emotionalism is known to be confounded by depression (House
will incorporate new data in future updates. 1989). Tearfulness can be the result of an underlying depressive
disorder. The inclusion of participants with depression and
In general, clinical trials are carried out on selected groups of emotionalism limits our ability to draw conclusions regarding the
individuals, while the usefulness of the information derived lies treatment of emotionalism alone. While all included trials assessed
primarily in the ability to generalise the data to a wide range of depression, only two trials reported results (Murray 2005; Robinson

Informed decisions.
1993b). Ideally, future trials should limit inclusion to participants Imprecision

with emotionalism alone, or recruit sufficient participants to allow
We downgraded the certainty of the evidence for 50% reduction
adjustment for depression in the results, and report the results from
in emotionalism and death by one level owing to wide confidence
all questionnaires administered.
intervals.
The main weakness of this review is the inadequate reporting of
Publication bias
some of the trials, which has precluded classification of risk of bias
as either low or high risk. This led us to rate some of the trials across We did not downgrade the certainty of the evidence for publication
the categories as unclear risk of bias, with a small number of trials bias in any of the outcomes, as we did not detect publication bias.
and participants contributing to most endpoints.
Potential biases in the review process
Quality of the evidence
This review has rigorously adhered to the Cochrane methods
We rated the certainty of evidence for all comparisons using the for performing systematic reviews and their respective updates.
five GRADE considerations (study limitations, consistency of effect, During the review process, we have tried to avoid and minimise
indirectness, imprecision, and publication bias; Schünemann any biases. We undertook extensive searches of databases and
2019). We created a summary of findings table. Our rating of the additional resources. We did not apply any language restrictions
certainty of the evidence ranged from very low to moderate. within the search process. Thus, we believe that we have identified
and included all potentially relevant trials in this review. We
As is often noted in reviews of the literature, there were several arranged for any relevant and non-relevant non-English language
deficiencies in trial methods that further limit our findings. First, full-text trials to be translated into English, to finalise the eligibility
most trials (except Ohtomo 1985) were small, with only three trials process. Furthermore, at least two review authors independently
reporting adequate concealment of the randomisation sequence extracted and managed the data.
(Brown 1998; Choi-Kwon 2006; Murray 2005). The duration of
treatment was short for most trials. Observational studies suggest Agreements and disagreements with other studies or
that many cases of emotionalism resolve over the first months reviews
after stroke, so the impact of short-term therapy may be difficult to
assess. Furthermore, case histories report relapse in emotionalism To date, there are no similar meta-analyses investigating the use
upon withdrawal of treatment, so it would be useful to have of pharmacological interventions in reducing the frequency of
information on longer-term results and relapse rates in future emotional displays in people with emotionalism after stroke, with
trials. We also included cross-over trials if the washout period was which to compare our findings. Most of the literature in this area
deemed appropriate. While the benefits and risks of treatment are focused on lesion network-symptom mapping (Klingbeil 2021),
with antidepressants appear to be balanced, only three trials clinical and neuroimaging markers (Gillespie 2022), and qualitative
systematically recorded and reported all adverse events in the investigation (Fitzgerald 2021; McAleese 2021).
study, which makes an accurate presentation of the benefits and
risks difficult (Burns 1999; Choi-Kwon 2006; Murray 2005). AUTHORS' CONCLUSIONS
Limitations in study design or execution Implications for practice

Concerning the comparison of pharmaceutical interventions with Antidepressants may reduce the frequency of emotionalism after
placebo for diminished tearfulness, we downgraded the certainty stroke. However, there is continued uncertainty about who might
of the evidence by one level for strong suspicions of attrition bias, benefit the most from treatment amongst those who meet the
related to the inconsistency in the total number of participants clinical features indicative of emotionalism. There are no data to
included in the analysis. guide recommendations about how long people should remain
on treatment or what side effects may be expected. Given the
Inconsistency of results limited evidence, clinicians and people who have had a stroke may
consider a therapeutic trial of antidepressants when emotionalism
We downgraded the certainty of the evidence by two levels for
is persistent and severe.
the following outcomes: 50% reduction in emotionalism, improved
score on the Center for Neurologic Study - Lability Scale (CNS-LS), Implications for research
Clinician Interview-Based Impression of Change (CIBIC), and PLCS
scores, as there were fewer than 50 participants and only one trial We recommend that future trials investigating the effect of
contributing to the analysis for each of these outcomes. antidepressants in people with emotionalism after stroke should:
Indirectness of evidence • use a standardised method to diagnose emotionalism,

determine severity, and assess change over time: this would
All included trials addressed the main review question (PICO): the assume development of a standard definition of emotionalism;
effect of pharmaceutical interventions when compared to placebo
in reducing the frequency of emotionalism and emotional displays • use a standard measure of depression as the major confounder
to be considered in analyses;
post-stroke. Thus, we did not downgrade any outcomes in any
comparisons for indirectness of evidence. • recruit an adequate number of participants so that variables
such as 'concomitant depression' and 'time passed between
stroke and recruitment' can be controlled;
• provide treatment for a sufficient duration and follow-up, so that
rates of relapse or maintenance of remission can be assessed;
Informed decisions.
• include careful assessment and complete reporting of adverse ACKNOWLEDGEMENTS

events;
• limit the number of outcomes to three or four and report results The review was supported by a grant from the Stroke Society of
for all outcomes; Australasia, with additional financial assistance provided by the
Academic Unit of Psychiatry, University of Leeds, and the Centre
• make the outcomes relevant to the individual participant by, for
for Clinical Brain Sciences, University of Edinburgh. The review
example, being clear whether reduction in crying frequency or
authors thank Cochrane Stroke, particularly Brenda Thomas, for
change in crying behaviour represented a satisfactory outcome
previously searching the Specialised Register of Cochrane Stroke
for the participant;
(now called Cochrane Stroke Group Register) and assisting with the
• ensure that the theoretical or biological rationale, or development of the search strategies. We also thank Hazel Fraser
both, and sequential development of interventions follow a for assistance throughout the review process. We would also like
recommended framework for development, and if a framework to acknowledge Craig S Anderson, Judith A Horrocks, and Michelle
is followed, then it must be reported in the main study Yang who contributed to earlier versions of this review.
publications (Walker 2017);
• ensure complete intervention descriptions to increase research For this update, Cochrane Stroke is grateful to the following peer
usability, replicability, and development of standardised reviewers for their time and comments:
interventions, as an identified priority for the world stroke
agenda (Walker 2017); and • Alex Todhunter-Brown: Co-ordinating Editor, Cochrane Stroke;
• adopt an implementation fidelity model or framework at the • Dr Amanda Barugh: Associate Editor, Cochrane Stroke;
design stage (Walker 2017). • Aryelly Rodriguez: Edinburgh Clinical Trials Unit (ECTU) at the
University of Edinburgh; and
• also to the peer reviewer who chose not to be publicly
acknowledged.

Informed decisions.
REFERENCES
References to studies included in this review Choi-Kwon S, Kwon SU, Kang DW, Kim JS. Fluoxetine improves
the quality of life in patients with post-stroke emotional
Andersen 1993 {published data only}
disturbances. Stroke 2009;40(4):e282.
Andersen G, Riis JO. Citalopram for post-stroke pathological
crying. Lancet 1993;342(8875):837-9. Murray 2005 {published and unpublished data}
Murray V, Von Arbin M, Asberg M, Bartfai A, Berggren A,
Andersen G, Vestergaard K, Riis JO. Citalopram for post-stroke
Landtblom A, et al. Double-blind placebo comparison of
pathological crying. Journal of Neurology 1994;241 Suppl 1:S81.
sertraline and placebo in stroke patients with depression. Data
Andersen G, Vestergaard K, Riis JO. Pathological crying and on file 2003.
emotional lability - a controlled study of citalopram - a SSRI.
* Murray V, Von Arbin M, Bartfai A, Berggren AL, Landtblom AM,
Canadian Journal of Neurological Sciences 1993;20 Suppl
Lundmark J, et al. Double-blind comparison of sertraline
4:S115.
and placebo in stroke patients with minor depression and
* Andersen G, Vestergaard K, Riis JO. Post-stroke pathological less severe major depression. Journal of Clinical Psychiatry
crying or emotional affect treated with citalopram: a selective 2005;66(6):708-16. [PMID: 15960563]
serotonin reuptake inhibitor. Acta Neurologica Scandinavica
Murray V, Von Arbin M, Varelius R, Olsson JE, Terent A,
1994;89:151.
Samuelsson M, et al. Sertraline in poststroke depression: a
Flicker C, Anderson G. Citalopram treatment of poststroke controlled study. Stroke 2002;33(1):292.
patients: improvement of uncontrolled crying. Journal of the
Ohkawa 1989 {published data only}
American Geriatrics Society 1998;46(9):S67.
Ohkawa S, Mori E, Yamadori A. Treatment of pathological
Brown 1998 {published data only} laughing with amitriptyline. Rinsho Shinkeigaku
Brown KW, Sloan RL, Pentland B. Fluoxetine as a treatment 1989;29(9):1183-5.
for post-stroke emotionalism. Acta Psychiatrica Scandinavica
Robinson 1993b {published data only}
1998;98(6):455-8.
* Parikh RM, Robinson RG, Lipsey JR, Price TR. Nortriptyline
Burns 1999 {published data only} treatment of post-stroke emotional lability: a double blind
* Burns A, Russell E, Stratton-Powell H, Tyrell P, O'Neill P, study. Neurology 1989;39 Suppl 1:177.
Baldwin R. Sertraline in stroke-associated lability of mood.
Robinson R. Additional data provided by author. Data on file
International Journal of Geriatric Psychiatry 1999;14(8):681-5.
2003.
Burns A, Russell E, Stratton-Powell H. A pilot study to evaluate
Robinson RG, Parikh RM, Lipsey JR, Starkstein SE, Price TR.
the efficacy of sertraline in the management of emotional
Pathological laughing and crying following stroke: validation
lability following stroke. In: 151st Annual Meeting of the
of a measurement scale and a double-blind treatment study.
American Psychiatric Association, Toronto, Ontario, Canada,
American Journal of Psychiatry 1993;150(2):286-93.
30th May to 4th June. Toronto, Ontario, Canada: American
Psychiatric Association, 1998.
References to studies excluded from this review
Choi-Kwon 2006 {published data only}
Choi-Kwon S, Choi J, Kwon SU, Kang DW, Kim JS. Fluoxetine Aizawa 1977 {published data only}
improves the quality of life in patients with poststroke Aizawa T, Kase M, Kutsuzawa T, Hasegawa T, Sekimoto H,
emotional disturbances. Cerebrovascular Diseases Omae T. Clinical evaluation on the effect of cyclandelate in
2008;26(3):266-71. [PMID: 18648199] the treatment of cerebrovascular diseases: double-blind
comparative placebo-controlled clinical study with cinnarizine
Choi-Kwon S, Choi J, Kwon SU, Kang DW, Kim JS. Fluoxetine is as a basic treatment. Rinsho Hyoka 1977;5(3):659-86.
not effective in the treatment of poststroke fatigue: a double-
blind, placebo-controlled study. Cerebrovascular Disease Allen 2018 {published data only}
2007;23(2-3):103-8. Allen C, Zarowitz B, O'Shea T, Peterson E, Yonan C, Waterman F.
Identification of pseudobulbar affect symptoms in the nursing
Choi-Kwon S, Choi JM, Kwon DW, Kim JS. Fluoxetine improves
home setting: development and assessment of a screening tool.
QoL in patients with post-stroke emotional disturbances.
Geriatric Nursing 2018;39(1):54-9.
International Journal of Stroke 2008;3(Suppl 1):405-6.
Atarashi 1988 {published data only}
* Choi-Kwon S, Han SW, Kwon SU, Kang DW, Choi JM, Kim JS.
Fluoxetine treatment in poststroke depression, emotional Atarashi J, Ohtomo E, Kogure K, Hirai S, Tazaki Y, Araki G, et
incontinence, and anger proneness: a double-blind, placebo- al. Clinical utility of HYG-FAS in treatment of cerebrovascular
controlled study. Stroke 2006;37(1):156-61. [PMID: 16306470] disorders: multi-center double-blind study in comparison with
hydergine tablet 2mg. Rinsho Hyoka 1988;16(3):425-86.

Informed decisions.
Bassi 1984 {published data only} www.www.clinicaltrialsregister.eu/ctr-search/

Bassi S, Albizzati MG, Sbacchi M, Frattola L. Chronic trial/2013-002253-30/DK (first received 9 September 2013).
cerebrovascular disorders: clinical study with cyclandelate.
Formella 2017a {published data only}
British Journal of Clinical Practice 1984;38:344-9.
Formella AE, Alexander DN, Cutler AJ, D'Amico S, Hammond FM,
Chen 2010 {published data only} Sauve W, et al. Dextromethorpha/quinidine improved
Chen YR, Huang YP, Lin SJ, Kuan TH, Lin CH. Effectiveness symptoms of pseudobulbar affect irrespective of concomitant
of quetiapine for poststroke pathological laughing: case antidepressant use. CNS Spectrums 2017;22(1):64-5.
report and review of literature. Clinical Neuropharmacology
Formella 2017b {published data only}
2010;33(6):319-22.
Formella AE, Alexander DN, Cutler AJ, D'Amico S, Hammond FM,
ChiCTR-IPR-15007651 2015 {published data only} Sauve W, et al. Dextromethorphan/quinidine improved
ChiCTR-IPR-15007651. Comparison of effects of fluoxetine symptoms of pseudobulbar affect irrespective of concomitant
in different dose on the secondary prevention of patients antidepressant use. Neurology 2017;88 Suppl 1:1-2.
in acute ischemic stroke: a randomized controlled trial.
Kim 2017a {published data only}
www.chictr.org.cn/showprojen.aspx?proj=12909 (first received
26 December 2015). Kim JS. Management of post-stroke mood and emotional
disturbances. Expert Review of Neurotherapeutics
Colamonico 2012 {published data only} 2017;17(12):1179-88.
Colamonico J, Formella A, Bradley W. Psuedobulbar
Kim 2017b {published data only}
affect: burden of illness in the USA. Advances in Therapy
2012;29(9):775-98. * Kim JS, Lee EJ, Chang DI, Park JH, Ahn SH, Cha JK, et al. The
efficacy of early administration of escitalopram on depressive
D'Amico 2017 {published data only} and emotional symptoms and neurological dysfunction after
D'Amico S, Alexander DN, Cutler AJ, Zorowitz RD, Davis CS, stroke: a multicentre, double-blind, randomised, placebo-
Shin P, et al. Safety and tolerability of dextromethorphan/ controlled study. Lancet Psychiatry 2017;4(1):33-41.
quinidine in older patients. American Journal of Geriatric
NCT01278498. The preventive effect of escitalopram on
Psychiatry 2017;25(3 Suppl 1):S142.
depression and related emotional disorders in acute stroke
Doody 2014 {published data only} patients. clinicaltrials.gov/ct2/show/NCT01278498 (first
received 19 January 2011).
Doody RS, D'Amico S, Cutler AJ, Shin P, Ledon F, Yonan C, et
al. Safety, tolerability and effectiveness of dextromethorphan/ Lawson 1969 {published data only}
quinidine for pseudobulbar affect in patients with Alzheimer's
Lawson IR, MacLeod RD. The use of imipramine ("Tofranil")
Disease/Dementia: PRISM-II. Annals of Neurology 2014;76(Suppl
and other psychotropic drugs in organic emotionalism. British
18):S99-S100.
Journal of Psychiatry 1969;115(520):281-5.
EUCTR2004-004803-39-DE 2006 {published data only}
Lorwatanapongsa 2020 {published data only}
EUCTR2004-004803-39-DE. SAINT II (Stroke - Acute Ischemic
Lorwatanapongsa S, Chompoonuch S, Pongpakdee S,
- NXY Treatment) A double blind, randomized, placebo
Mayotarn S, Piyatanont K, Awaiwanont A. Randomized
controlled, parallel group, multicenter, phase IIb/III study to
controlled trial of fluoxetine or placebo on quality of life after
assess the efficacy and safety of intravenous NXY-059 in acute
acute ischemic stroke (RTAF-QoL). In: 36th Annual Meeting, The
ischemic stroke - SAINT-II. www.clinicaltrialsregister.eu/ctr-
Royal College of Physicians of Thailand. 2020.
search/trial/2004-004803-39/results (first received 18 August
2005). Manzo 1998 {published data only}
EUCTR2011-005541-12-DK 2012 {published data only} Manzo JF, Health RL, Blonder LX. The interpersonal
management of crying among survivors of stroke. Sociological
EUCTR2011-005541-12-DK. Efficacy of escitalopram treatment
Spectrum 1998;18(2):161-84.
in acute stroke and the role of specific genotypes in stroke.
www.clinicaltrialsregister.eu/ctr-search/trial/2011-005541-12/ Moller 2007 {published data only}
DK (first received 27 February 2012).
Moller M, Andersen G, Gjedde A. Serotonin 5HT1A receptor
EUCTR2011-006130-16-SE 2014 {published data only} availability and pathological crying after stroke. Acta
Neurologica Scandinavica 2007;116(2):83-90.
EUCTR2011-006130-16-SE. Establishing the effect(s) and safety
of fluoxetine in non-depressed stroke patients initiated in the Muller 1999 {published data only}
acute phase of stroke. www.clinicaltrialsregister.eu/ctr-search/
Muller U, Murai T, Bauer-Wittmund T, Cramon DY. Paroxetine
trial/2011-006130-16/SE (first received 16 June 2014).
versus citalopram treatment of pathological crying after brain
EUCTR2013-002253-30-DK 2013 {published data only} injury. Brain Injury 1999;13(10):805-11.
EUCTR2013-002253-30-DK. Efficacy of citalopram
treatment in acute stroke [Clinical Trials Register].

Informed decisions.
Narushima 2002 {published data only} Schiffer 1985 {published data only}
Narushima P, Kosier J, Robinson R. Preventing poststroke Schiffer RB, Herndon RM, Rudick RA. Treatment of pathologic
depression: a 12-week double-blind randomized treatment trial laughing and weeping with amitriptyline. New England Journal
and 21-month follow-up. Journal of Nervous and Mental Disease of Medicine 1985;312(23):1480-2.
2002;190(5):296-303.
Schoedel 2014 {published data only}
NCT01278498 2011 {published data only} Schoedel KA, Morrow SA, Sellers EM. Evaluating the safety and
NCT01278498. The preventive effect of escitalopram on efficacy of dextromethorphan/quinidine in the treatment of
depression and related emotional disorders in acute stroke pseudobulbar affect. Neuropsychiatric Disease and Treatment
patients. clinicaltrials.gov/ct2/history/NCT01278498 (first 2014;10:1161-74.
received 18 January 2011).
Seliger 1992 {published data only}
NCT01799941 2013 {published data only} Seliger GM, Hornstein A, Flax J, Herbert J, Schroeder K.
NCT01799941. Safety, tolerability and effectiveness of nuedexta Fluoxetine improves emotional incontinence. Brain Injury
in the treatment of pseudobulbar affect (PBA). clinicaltrials.gov/ 1992;6(3):267-70.
ct2/show/NCT01799941 (first received 27 February 2013).
Udaka 1984 {published data only}
Ohtomo 1985 {published data only} Udaka F, Yamao S, Nagata H, Nakamura S, Kameyama M.
Ohtomo E, Kutsuzawa T, Araki G, Hirai S, Terashi A, Kuzuya F, Pathologic laughing and crying treated with levodopa. Archives
et al. Clinical usefulness of tiapride on psychiatric symptoms of Neurology 1984;41(10):1095-6.
caused by cerebrovascular disorders: a multi-center double-
blind study in comparison with inactive placebo. Rinsho Hyoka Work 2011 {published data only}
1985;13(2):295-332. Work SS, Colamonico JA, Bradley WG, Kaye RE. Pseudobulbar
affect: an under-recognized and under-treated neurological
Otomo 1984 {published data only} disorder. Advances in Therapy 2011;28(7):586-601.
Otomo E, Saso S, Araki G, Hirai S, Atarashi J, Hasegawa K, et al.
Clinical evaluation of amantadine hydrochloride (Symmetrel) Yang 2015 {published data only}
in the treatment of cerebrovascular disorders with psychiatric Yang LPH, Deeks ED. Dextromethorphan/quinidine: a
symptoms: multi-center double-blind study in comparison with review of its use in adults with pseudobulbar affect. Drugs
placebo. Rinsho Hyoka 1984;12(2):321-67. 2015;75(1):83-90.
Rasmussen 2000 {published data only}

Rasmussen A, Klysner R, Mellerup E. A double-blind, placebo- Additional references
controlled study of the prophylactic effect of sertraline in post Allman 1989
stroke emotional lability. Conference poster 2000.
Allman P. Crying and laughing after brain damage: a confused
Rasmussen A, Lunde M, Poulsen DL, Sorensen K, Qvitzau S, nomenclature. Journal of Neurology, Neurosurgery, and
Bech P. A double-blind, placebo-controlled study of sertraline in Psychiatry 1989;52(12):1439-40.
the prevention of depression in stroke patients. Psychosomatics
2003;44(3):216-21. Allman 1992a
Allman P. Drug treatment of emotionalism following
* Rasmussen A. A double blind controlled study of the influence brain damage. Journal of the Royal Society of Medicine
of sertraline on the incidence of post stroke depression and 1992;85(7):423-4.
on emotional lability [PhD Thesis]. Copenhagen (Denmark):
University of Copenhagen, 2000. Allman 1992b
Allman P, Marshall M, Hope A, Fairburn C. Emotionalism
* Rasmussen A. Depression and stroke. Nordic Journal of following stroke: development and reliability of a semi-
Psychiatry 2001;55(4):288. structured interview. International Journal of Methods in
Rasmussen A. Post-stroke emotional lability. A double-blind Psychiatric Research 1992;2:125-31.
psychopharmacological treatment with sertralin. Nordic Journal APA 1987
of Psychiatry 2000;54:35.
American Psychiatric Association. Diagnostic and Statistical
Sauve 2017 {published data only} Manual of Mental Disorders: DSM-IIIR. Washington, DC:
Sauve W, Alexander DN, Cutler AJ, D'Amico S, Hammond FM, American Psychiatric Association, 1987.
Zorowitz RD, et al. Neurological and psychiatric comorbidities APA 1994
assessment in the PRISM II study of dextromethorphan/
quinidine for treatment of pseudobulbar affect. CNS Spectrums American Psychiatric Association. Diagnostic and Statistical
2017;22(1):104. Manual of Mental Disorders: DSM-IV. Washington, DC: American
Psychiatric Association, 1994.

Informed decisions.
APA 2017 Goldberg 1972

American Psychiatric Association. Diagnostic and Statistical Goldberg DP. The Detection of Psychiatric Illness by
Manual of Mental Disorders. 5th edition. Arlington VA: American Questionnaire. Vol. 21. Oxford: Oxford University Press, 1972.
Psychiatric Publishing, 2017.
Gompertz 1993
Beck 1961 Gompertz P, Pound P, Ebrahim S. The reliability of stroke
Beck AT, Ward C, Mendelson M. An inventory for measuring outcome measurement. Clinical Rehabilitation 1993;7(4):290-6.
depression. Archives of General Psychiatry 1961;4:561-71.
GRADEpro GDT [Computer program]
Benedek 1995 GRADEpro GDT. Hamilton (ON): McMaster University (developed
Benedek D, Peterson KA. Sertraline for treatment of by Evidence Prime), accessed 28 April 2022. Available at
pathological crying. American Journal of Psychiatry gradepro.org.
1995;152(6):953-4.
Hamilton 1960
Brookshire 1970 Hamilton M. Rating scale for depression. Journal of Neurology,
Brookshire R. Control of "involuntary" crying behaviour emitted Neurosurgery and Psychiatry 1960;23:56-62.
by a multiple sclerosis patient. Journal of Communication
Disorders 1970;3(3):171-6. Hanger 1993
Hanger HC. Emotionalism after stroke. Lancet
Broomfield 2021 1993;342(8881):1235-6.
Broomfield NM, West R, House A, Munyombwe T, Barber M,
Gracey F, et al. Psychometric evaluation of a newly developed Higgins 2017
measure of emotionalism after stroke (TEARS-Q). Clinical Higgins JP, Green S, editor(s). Cochrane Handbook for
Rehabilitation 2021;35(6):894-903. Systematic Reviews of Interventions Version 5.2.0 (updated
June 2017). The Cochrane Collaboration, 2017. Available from
Calvert 1998 training.cochrane.org/handbook/archive/v5.2.
Calvert T, Knapp P, House A. Psychological associations with
emotionalism after stroke. Journal of Neurology, Neurosurgery House 1989
and Psychiatry 1998;65(6):928-9. House A, Dennis M, Molyneux A, Warlow C, Hawton K.
Emotionalism after stroke. BMJ 1989;298(6679):991-4.
Deeks 2021
Deeks JJ, Higgins JPT, Altman DG. Chapter 10: Analysing data ICD 10
and undertaking meta-analyses. In: Higgins JP, Thomas J, World Health Organization. The ICD-10 Classification of Mental
Chandler J, Cumpston M, Li T, Page MJ, Welch VA, editor(s). and Behavioural Disorders: Clinical Descriptions and Diagnostic
Cochrane Handbook for Systematic Reviews of Interventions Guidelines. Geneva: World Health Organization, 1992.
Version 6.2 (updated February 2021). Cochrane, 2021. Available
from training.cochrane.org/handbook. Kim 2000
Kim JS, Choi-Kwon S. Post stroke depression and emotional
Fitzgerald 2021 incontinence. Neurology 2000;54(9):1805-10.
Fitzgerald S, Gracey F, Broomfield N. Post-stroke emotionalism
(PSE): a qualitative longitudinal study exploring individuals' Kim 2005
experience with PSE. Disability and Rehabilitation 2021 [Epub Kim SW, Shin IS, Kim JM, Lim SY, Yang SJ, Yoon JS. Mirtazapine
ahead of print]. [DOI: 10.1080/09638288.2021.2002439] treatment for pathological laughing and crying after stroke.
Clinical Neuropharmacology 2005;28(5):249-51.
Folstein 1975
Folstein MF, Folstein SE, McHugh PR. 'Mini-Mental State': a Klingbeil 2021
practical method for grading the cognitive state of patients for Klingbeil J, Wawrzyniak M, Stockert A, Brandt M-L, Schneider
the clinician. Journal of Psychiatric Research 1975;12(3):189-98. H-R, Metelmann M, et al. Pathological laughter and crying:
insights from lesion network-symptom-mapping. Brain
Gillespie 2016 2021;144(10):3264-76.
Gillespie D, Cadden AP, Lees R, West RM, Broomfield NM.
Prevalence of psuedobulbar affect following stroke: a Knopman 1994
systematic review and meta-analysis. Journal of Stroke and Knopman D, Knapp MJ, Gracon SI, Davis CS. The Clinician Based
Cerebrovascular Diseases 2016;25(3):688-94. Impression (CIBI): a clinician’s global change rating scale in
Alzheimer’s disease. Neurology 1994;44:2315–21.
Gillespie 2022
Gillespie DC, Halai AD, West RM, Dickie DA, Walters M, Lauterbach 1991
Broomfield NM. Demographic, clinical and neuroimaging Lauterbach E, Schweri M. Amelioration of pseudobulbar
markers of post-stroke emotionalism: a preliminary affect by fluoxetine. Journal of Clinical Psychopharmacology
investigation. Journal of the Neurological Sciences 1991;11:392-3.
2022;436:120229.
Informed decisions.
Lawson 1969 Review Manager [Computer program]

Lawson JR, Macleod RD. The use of imipramine ("Tofranil") Review Manager 5 (RevMan 5). Version 5.4. Copenhagen: Nordic
and other psychotropic drugs in organic emotionalism. British Cochrane Centre, The Cochrane Collaboration, 2020.
Journal of Psychiatry 1969;115:281-5.
Robinson 1993a
Mahoney 1965 Robinson RG, Parikh RM, Lipsey JR, Starkstein SE, Price TR.
Mahoney FI, Barthel DW. Functional evaluation: the Barthel Pathological laughing and crying following stroke: validation
Index. Maryland State Medical Journal 1965;14:61-5. of a measurement scale and a double-blind treatment study.
American Journal of Psychiatry 1993;150(2):286-93.
Massey 1981
Massey E, Lowe S. Lithium carbonate in pseudobulbar palsy. Sacco 2008
Annals of Neurology 1981;9(1):97. Sacco S, Sarà M, Pistoia F, Conson M, Albertini G, Carolei A.
Management of pathologic laughter and crying in patients
McAleese 2021 with locked-in syndrome: a report of 4 cases. Archives Physical
McAleese N, Guzman A, O'Rourke SJ, Gillespie DC. Post-stroke Medicine and Rehabilitation 2008;89(4):775-8.
emotionalism: a qualitative investigation. Disability and
Rehabilitation 2021;43(2):192-200. Sandyk 1985
Sandyk R, Gillman MA. Nomifensine for emotional incontinence
Montgomery 1979 in the elderly. Clinical Neuropharmacology 1985;8(4):377-8.
Montgomery SA, Asberg M. A new depression scale designed
to be sensitive to change. British Journal of Psychiatry Schiffer 1983
1979;134:382-9. Schiffer R, Cash J, Herndon R. Treatment of emotional lability
with low dosage tricyclic antidepressants. Psychosomatics
Moore 1997 1983;24:1094-6.
Moore S, Gresham L, Bromberg M, Kasarkis E, Smith R. A self-
report measure of emotional lability. Journal of Neurology, Schünemann 2019
Neurosurgery and Psychiatry 1997;63(1):89-93. Schünemann HJ, Vist GE, Higgins JP, Santesso N, Deeks JJ,
Glasziou P, et al. Chapter 15: Interpreting results and drawing
Mukand 1996 conclusions. In: Higgins JP, Thomas J, Chandler J, Cumpston
Mukand J, Kaplan M, Senno RG, Bishop DS. Pathological crying M, Li T, Page MJ, Welch VA, editor(s). Cochrane Handbook for
and laughing: treatment with sertraline. Archives of Physical Systematic Reviews of Interventions Version 6.0 (updated July
Medicine and Rehabilitation 1996;77(12):1309-11. 2019). Cochrane, 2019. Available from training.cochrane.org/
handbook/archive/v6.
Nahas 1998
Nahas Z, Arlinghaus KA, Kotrla KJ, Clearman RR, George MS. Seliger 1989
Rapid response of emotional incontinence to selective Seliger GM, Hornstein A. Serotonin, fluoxetine, and
serotonin reuptake inhibitor. Journal of Neuropsychiatry and pseudobulbar affect. Neurology 1989;39(10):1400.
Clinical Neurosciences 1998;10:453-5.
Sloan 1992
Newsom-Davis 1999 Sloan RL, Brown KW, Pentland B. Fluoxetine as a treatment for
Newsom-Davis I, Abraham S, Goldstein L, Leigh P. The emotional lability after brain injury. Brain Injury 1992;6(4):315-9.
emotional lability questionnaire: a measure of emotional
lability in amyotrophic lateral sclerosis. Journal of the Smith 2003
Neurological Sciences 1999;169(1-2):22-5. Smith AG, Montealegre-Orjuela M, Douglas JE, Jenkis EA.
Venlafaxine for pathological crying after stroke. Journal of
Page 2021 Clinical Psychiatry 2003;64(6):731-2.
Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC,
Mulrow CD, et al. The PRISMA 2020 statement: an updated Tan 1996
guideline for reporting systematic reviews. BMJ 2021;372:n71. Tan I, Dorevitch M. Emotional incontinence: a dramatic
response to paroxetine. Australian and New Zealand Journal of
Panzer 1992 Medicine 1996;26(6):844.
Panzer MJ, Mellow AM. Antidepressant treatment of
pathological laughing or crying in elderly stroke patients. Walker 2017
Journal of Geriatric Psychiatry and Neurology 1992;5(4):195-9. Walker MF, Hoffmann TC, Brady MC, Dean CM, Eng JJ,
Farrin AJ, et al. Improving the development, monitoring
Poeck 1969 and reporting of stroke rehabilitation research: Consensus-
Poeck K. Pathophysiology of emotional disorders associated based core recommendations from the Stroke Recovery and
with brain damage. In: Vinken PJ, Bruyn GW, editors(s). Rehabilitation Roundtable. Neurorehabilitation and Neural
Handbook of Clinical Neurology. Holland: Holland Publishing Repair 2017;31(10-11):877-84.
Co, 1969:343-67.

Informed decisions.
Wolf 1997 Cochrane Database of Systematic Reviews 2010, Issue 2. Art. No:
Wolf J, Santana H, Thorpy M. Treatment of 'emotional CD003690. [DOI: 10.1002/14651858.CD003690.pub3]
incontinence' with levodopa. Neurology 1997;29:1935-6.
House 2002
Zigmond 1983 House AO, Anderson CS, Hackett ML, Horrocks JA. Interventions
Zigmond AS, Snaith RP. The Hospital Anxiety and Depression for emotionalism after stroke. Cochrane Database of
Scale. Acta Psychiatrica Scandanavica 1983;67(6):361-70. Systematic Reviews 2002, Issue 3. Art. No: CD003690. [DOI:
10.1002/14651858.CD003690]
References to other published versions of this review House 2004

House AO, Hackett ML, Anderson CS, Horrocks JA.
Allida 2019
Allida S, Patel K, House A, Hackett ML. Pharmaceutical Cochrane Database of Systematic Reviews 2004, Issue 2. Art. No:
interventions for emotionalism after stroke. Cochrane Database CD003690. [DOI: 10.1002/14651858.CD003690]
of Systematic Reviews 2019, Issue 3. Art. No: CD003690. [DOI:
10.1002/14651858.CD003690.pub4]
* Indicates the major publication for the study
Hackett 2010
Hackett ML, Yang M, Anderson CS, Horrocks JA, House A.
CHARACTERISTICS OF STUDIES
Characteristics of included studies [ordered by study ID]
Andersen 1993
Methods Study design: randomised, cross-over design

Number of arms: 2
Experimental arm: citalopram 20 mg/day if under 66 years old, 10 mg/day if older
Control arm: placebo
Analysis: per protocol: 1 withdrawn (placebo), excluded from analysis; 2 early dropouts (placebo), last
value carried forward
Participants Geographical location: Denmark

Setting: Aalborg Hospital
Number of participants: 16
Diagnosis: stroke, 6 to 913 days prior to randomisation
Inclusion criteria: not specified
Exclusion criteria: 1) aphasia, 2) if participants or spouse is unable to keep a diary
Age: median age 58.5 years, range 40 to 83 years. Mean not reported
Sex: 88% men
Emotionalism criteria: involuntary outbursts of crying
Interventions Treatment: citalopram 20 mg daily if under 66 years old, 10 mg daily for older participants
Control: matched placebo
Cross-over details: 7 days baseline; 21-day intervention; 7-day washout. 7-day baseline; 21-day inter-
vention
Duration: treatment continued for 21 days
Washout period: 7 days + 7 days baseline registration
Outcomes • Improvements in crying history

• Semi-structured interview (modified from Lawson and Macleod)
• Qualitative clinical evaluation of facial grimacing and concomitant crying
• Frequency of crying episodes (5-point scale; none to continuous)
• Context in which episodes occurred (3-point scale; non-specific to emotionally provoked)
• Recorded quantitative or qualitative data concerning crying (kept in a diary)
• Unwanted side effects UKU side effects rating scale)

Informed decisions.
Andersen 1993 (Continued)

• Compliance (examination of medical containers at the end of study)
Notes Able to use:
• Adverse events
Unable to use in analysis: interviewer assessed that the participant no longer met criteria for emotion-
alism - modified Lawson and Macleod scale, 50% reduction in emotionalism, HDRS, leaving the study
early, adverse events (data not reported in appropriate format)
Funding: "H Lundbeck A/ and Lundbeck Pharma A/S, Copenhagen, provided the citalopram and place-
bo tablets. Lundbeck Pharma A/ also paid the patients’ transportation expenses and for secretarial as-
sistance and statistical advice."
Risk of bias
Bias Authors' judgement Support for judgement
Random sequence genera- Low risk Comments: participants were randomised in blocks of 4; sequence generated
tion (selection bias) by Lundbeck.
Allocation concealment Low risk Comments: the method of concealment was centralised.
(selection bias)
Blinding of participants Unclear risk Quote: "We investigated the effect of the selective serotonin reuptake inhibitor
and personnel (perfor- citalopram on uncontrolled crying in stroke patients in a double-blind place-
mance bias) bo-controlled crossover study." (pp. 837)
All outcomes
Comments: the study authors stated that this was a double-blind trial, but did
not provide details of who were blinded.
Blinding of outcome as- Unclear risk Quote: "We investigated the effect of the selective serotonin reuptake inhibitor
sessment (detection bias) citalopram on uncontrolled crying in stroke patients in a double-blind place-
All outcomes bo-controlled crossover study." (pp. 837)
Incomplete outcome data Low risk Quote: "1 of the patients with classic pathological crying (no 8) was withdrawn
(attrition bias) from study during the initial treatment period (placebo) because of a gener-
All outcomes alised seizure on day 28; 2 others (no 3 and no 9) did not complete the second
treatment period (both placebo) because of lack of response to treatment af-
ter the first week." (pp. 838)
Quote: "In the latter cases the self-registered score in the first week was there-
fore used as the endpoint score." (pp. 838)
Comments: Dropouts or exclusions were detailed. The reasons for dropouts or

exclusions were not related to group allocation. The study authors provided
the method of how they dealt with the dropouts.
Selective reporting (re- Low risk Comments: all prespecified outcomes were reported. There was no trial proto-
porting bias) col available prior to randomisation of the first participant.
Other bias Unclear risk Comments: trial drug and placebo, statistical advice and secretarial support
provided by pharmaceutical company.

Informed decisions.
Brown 1998
Methods Study design: parallel, randomised controlled trial

Number of arms: 2
Experimental arm: fluoxetine 20 mg/day
Analysis: per protocol: 1 withdrawn (treatment), excluded from analysis
Participants Geographical location: Scotland

Setting: Astley Ainslie Hospital
Diagnosis: stroke, time from stroke to randomisation not reported
Inclusion criteria: 1) following a stroke who had a history of emotionalism of at least 4 weeks duration
Exclusion criteria: 1) cognitive impairment, 2) dysphasia, 3) major depressive disorder
Age: overall mean age not reported
Number of participants included in treatment group: 10 participants (55% men, mean age 61.4 years,
SD 8.6)
Number of participants included in control group: 10 participants (60% men, mean age 63.7 years, SD
5.4)
Emotionalism criteria: emotionalism of at least 4 weeks duration assessed during semi-structured in-
terview using a modified Lawson and MacLeod rating scale, in addition to frequency of outbursts
Interventions Treatment: fluoxetine 20 mg/day

Duration: treatment continued for 10 days
Outcomes • Emotional outburst grade (modified Lawson and Macleod gradings)

• Depression (HDRS)
• Cognitive functioning (MMSE)
• Side effects (24-item checklist of possible symptoms)
Notes Able to use in analysis:
• 50% reduction in frequency of emotionalism outbursts

• Leaving the study early
Unable to use in analysis: HDRS, Lawson and Macleod Scale, self-rating scales (mean and SD not pre-
sented), adverse events
Funding: information not available
Risk of bias
Random sequence genera- Unclear risk Quote: "The patients were randomly allocated by an independent statisti-
tion (selection bias) cian..." (pp.456)
Comments: participants were randomly allocated, but the method of alloca-

tion was not detailed.
Allocation concealment Low risk Quote: "The medication was repackaged so as to make the active and placebo
(selection bias) capsules identical to each other." (pp. 456)
Blinding of participants Low risk Quote: "The patients, nursing staff and rating clinicians were blinded to the al-
and personnel (perfor- location of active or placebo medication." (pp. 456)
mance bias)
All outcomes

Informed decisions.
Brown 1998 (Continued)
Blinding of outcome as- Low risk Quote: "The patients, nursing staff and rating clinicians were blinded to the al-
sessment (detection bias) location of active or placebo medication." (pp. 456)
All outcomes
Incomplete outcome data Low risk Quote: "One patient had to be withdrawn because he developed a generalised
(attrition bias) rash on active..." (pp. 456)
All outcomes
Comments: dropouts or exclusions were detailed. Reasons for dropouts or ex-
clusions were not related to group allocation.
Other bias Unclear risk Comments: it was unclear how the trial was funded.
Burns 1999

Number of arms: 2
Experimental arm: sertraline 50 mg/day
Analysis: intention-to-treat: 2 withdrawn and 1 death (treatment), 1 death (placebo), last value carried
forward
Participants Geographical location: UK

Setting: 3 hospitals in Manchester
Diagnosis: stroke, 1 to 156 months prior to randomisation
Inclusion criteria: 1) clinically documented stroke (with or without computed tomography evidence of
infarction), 2) presence of lability of mood observed by the referring clinician, 3) at least 1 month hav-
ing elapsed since stroke, 4) absence of depression and dementia according to DSM-III-R criteria
Exclusion criteria: 1) less than 1 month since stroke, 2) depression or dementia using DSM-III-R criteria
Number of participants included in treatment group: 14 (36% men, mean age 73 years, SD 9.1)
Number of participants included in control group: 14 (57% men, mean age 67.6 years, SD 8.5)
Emotionalism criteria: lability of mood observed by referring clinician
Interventions Treatment: sertraline 50 mg/day

Duration: treatment continued for 8 weeks
Outcomes Primary outcomes
• Emotionalism/lability of mood (Center for Neurologic Study - Lability Scale)

• Episodes of tearfulness (4-point rating scale; 0: 1 episode less than once per week; 1: episodes more
than once a week but less than once a day; 2: episodes up to 5 times a day; 3: episodes 6 or more times
a day)
• CIBIC rating scale
Secondary outcomes

• Depression (MADRS)
• Physical functional ability (Barthel Index)
• Severity of stroke (Scandinavian Stroke Scale)
Informed decisions.
Burns 1999 (Continued)

• Language dysfunction (Frenchay Aphasia Battery)
• Improved score on the Center for Neurologic Study - Lability Scale (House 1989)
• Improved score on CIBIC
• Diminished tearfulness
• Death
• Adverse events
Unable to use in analysis: MADRS, Barthel, MMSE (data not presented)
Funding: sponsored by Pfizer
Risk of bias
Random sequence genera- Low risk Quote: "After randomization (in blocks of four using a random number alloca-
tion (selection bias) tion list produced by the Department of Medical Statistics…)" (pp.683)
Allocation concealment Unclear risk Comments: not detailed

(selection bias)
Blinding of participants Unclear risk Quote: "The study was carried out according to a double-blind placebo con-
and personnel (perfor- trolled..." (pp. 683)
mance bias)
All outcomes Comments: the study authors stated that this was a double-blind trial, but did
Blinding of outcome as- Unclear risk Quote: "The study was carried out according to a double-blind placebo con-
sessment (detection bias) trolled..." (pp. 683)
All outcomes
Incomplete outcome data Low risk Quote: "Four patients did not complete the study. Two withdrew in the sertra-
(attrition bias) line group..." (pp. 683)
All outcomes
Quote: "Results are presented on an intention to treat basis, with the last ob-
servation carried forward..." (pp. 683)

clusions were not related to group allocation. The study authors provided the
method of how they dealt with the dropouts.
Other bias Unclear risk Comments: the pharmaceutical company provided the trial drug and placebo.
Choi-Kwon 2006
Methods Study design: parallel design

Informed decisions.
Choi-Kwon 2006 (Continued)

Number of arms: 2
Experimental arm: fluoxetine 20 mg/day
Analysis: intention-to-treat: 3 withdrawn (placebo), last value carried forward
Per protocol was also performed to investigate the consistency of the results
Participants Geographical location: South Korea

Setting: Asan Medical Centre
Diagnosis: stroke, time from stoke to randomisation not reported
Exclusion criteria: 1) did not undergo imaging (CT/MRI) studies, 2) had subarachnoid haemorrhage, 3)
had transient ischaemic attack without progression to stroke, 4) had communication problems (apha-
sia, dementia or dysarthria) severe enough as not to undergo a reliable interview, 5) were scored < 23
on MMSE, 6) had a history of being diagnosed as having depression or other psychiatric illnesses before
the onset of stroke, 7) had been already treated with psychiatric regimens including SSRI, and 8) lived
alone so that information from the relatives was not available
Age: mean age not reported
Number of participants included in treatment group: 44* - age and sex of participants with excessive
crying not reported
Number of participants included in control group: 48* - age and sex of participants with excessive cry-
ing not reported
*Number stated is inconsistent in the report - we have reported data on excessive/inappropriate crying
only, as represented in Table 3 of Choi-Kwon 2006
Emotionalism criteria: emotionalism is present if both participants and relatives agreed that ≥ 2 occa-
sions of excessive or inappropriate laughing or crying, or both, has occurred when compared with their
premorbid state.
Interventions Treatment: fluoxetine 20 mg/day

Duration: treatment continued for 3 months
Assessments performed at enrolment, 1 month, 3 months, and 6 months
Outcomes Primary outcomes
• Presence of post-stroke depression (BDI score)

• Intensity of post-stroke emotional incontinence (VAS)
• Presence of post-stroke anger (Spielberger Trait Anger Scale)
Secondary outcomes
• Percentage changes in BDI scores for post-stroke depression

• Percentage changes in VAS scores for emotional incontinence crying/laughing
• Percentage changes in post-stroke anger scores
• Patients' subjective responses as "aggravated", "no change", and "improved"
• Participants subjective responses of improvement in excessive/inappropriate crying

Unable to use in analysis: VAS for measuring extent of excessive or inappropriate laughing or crying
(data not presented in appropriate format). Percentage change of VAS between follow-ups (number of
emotionalism participants is inconsistent in report)
Adverse events (data for emotionalism participants not stated)
Funding: "supported by a grant from the Brain Research Center of the 21st Century Frontier Research
Program funded by the Ministry of Science and Technology of Korea (M103 KV01001006 K220101010)
and the Research Institute of Nursing Science at Seoul National University" and "by a research fund
from the Korean Ministry of Health and Welfare (03-PJ1-PG1-CH06 – 0001)."

Informed decisions.
Choi-Kwon 2006 (Continued)
Risk of bias
Random sequence genera- Low risk Quote: "Treatment allocation was based on a computer generated list of treat-
tion (selection bias) ment numbers." (pp.157)

(selection bias)
Blinding of participants Low risk Quote: "The patient, relatives and researchers were not aware of the drug be-
and personnel (perfor- ing given." (pp. 157)
mance bias)
All outcomes
Blinding of outcome as- Low risk Quote: "The patient, relatives and researchers were not aware of the drug be-
sessment (detection bias) ing given." (pp. 157)
All outcomes
Incomplete outcome data High risk Quote: "Among 152 patients, 27 dropped out before completing the 3-month
(attrition bias) treatment protocol (15 received fluoxetine, and 12 received placebo), leaving
All outcomes 125 patients. Although there was no difference in the dropout rate between
the 2 groups..." (pp. 157)
Comments: the number of total participants stated was inconsistent.
Murray 2005

Number of arms: 2
Experimental arm: sertraline 50 mg/day to 100 mg/day
Analysis: intention-to-treat: 8 withdrawn (treatment), 9 withdrawn and 1 death (placebo), last value
carried forward
Participants Geographical location: Sweden

Setting: 4 stroke centres throughout Sweden
Diagnosis: stroke, and depression within 12 months after a stroke, 3 to 375 days prior to randomisation
(mean of 128 ± 97 days)
Inclusion criteria: 1) with major depressive episode according to DSM-IV criteria, 2) minor depressive
disorder according to DSM-IV research criteria and a MADRS score > 10
Exclusion criteria: 1) apparent difficulties in adhering to the study protocol, 2) severe impairment of the
ability to communicate, 3) acute myocardial infarction, 4) psychiatric illnesses other than depression,
5) significant risk of suicide, 6) on antidepressant drug treatment during the month before the study
start, 7) current use of any psychotropic medications (except small daytime doses of benzodiazepines
or zopiclone, zolpidem, or benzodiazepines for night sedation), and 8) current use of opiate analgesic
drugs
Number of participants included in treatment group: 24 (58% men, mean age 69.5 years, SD 9.4)
Informed decisions.
Murray 2005 (Continued)

Number of participants included in control group: 20 (30% men, mean age 65.9, SD 10.9)
Emotionalism criteria: increased tearfulness and pathological crying reported by participants and rela-
tives
Interventions Treatment: sertraline 50 mg to 100 mg daily (dosage was increased to 100 mg for participants with lack
of improvement after 4 weeks; intake reduced to the starting dose if side effects occurred)
Outcomes Primary outcome
• Change in MADRS
Secondary outcomes
• Response rate > 50% decrease in MADRS score

• Remission rate MADRS score < 1
• Presence or absence of emotionalism measured as a dichotomous variable

• MADRS
• Death
• Limited adverse events
Funding: "supported by an unrestricted grant, study drug, and placebo from Pfizer AB Sweden and
grants from AFA Insurances and Marianne and Marcus Wallenberg Foundation."
Risk of bias
Random sequence genera- Low risk Quote: "A centralized randomization procedure was applied." (pp. 709)
tion (selection bias)
Allocation concealment Low risk Quote: "Each centre pharmacy received a consecutive series of pre-sealed
(selection bias) treatment packages. Patient received double-blind identical capsules of either
sertraline..." (pp. 709)
Blinding of participants Unclear risk Quote: "This 26-week, double-blind, placebo-controlled study of sertraline
and personnel (perfor- was carried out..." (pp. 709)
mance bias)
All outcomes Comments: the study authors stated that this was a double-blind trial, but did
Blinding of outcome as- Unclear risk Quote: "This 26-week, double-blind, placebo-controlled study of sertraline
sessment (detection bias) was carried out..." (pp. 709)
All outcomes
Incomplete outcome data Low risk Quote: "At week 6, 11 patients (18%) in the sertraline group and 6 patients
(attrition bias) (10%) in the placebo group had dropped out of the study (Figure 2). At week
All outcomes 26, an additional 13 patients (21%) in the sertraline group and 24 patients
(39%) in the placebo group had been withdrawn. Of the 54 patients prema-
turely withdrawn, 30 had a major depressive episode and 17 a minor depres-
sive disorder. Lack of antidepressant effect was the reason for exclusion in 38
cases and side effects in 13."

Informed decisions.
Murray 2005 (Continued)

clusions were not related to group allocation. The study authors provided the
method of how they dealt with the dropouts. Analysis was performed based on
the intention-to-treat principle.
Ohkawa 1989
Methods Study design: cross-over, randomised controlled trial

Number of arms: 2
Experimental arm: amitriptyline 50 mg/day
Control arm: bitter stomach medicine only
Analysis: per protocol: 2 withdrawn (treatment), 1 withdrawn (placebo), excluded from analysis
Participants Geographical location: Japan

Setting: hospital setting
Diagnosis: mixed vascular group with lacunar state, 1 month to 2 years prior to randomisation
Exclusion criteria: not specified
Age: mean age 63.3 years, SD 7.2, range 51 to 73 years
Sex: 86% male
Emotionalism criteria: compulsive laughter alone or with compulsive crying, definition unclear
Interventions Treatment: amitriptyline 50 mg daily, mixed with bitter stomach medicine to disguise taste and smell
Control: bitter stomach medicine only
Cross-over details: 3 weeks each of placebo and amitriptyline in random order
Outcomes • Frequency of occurrence of compulsive laughter (classified into 4 classes: 0 - none, 1 - mild, only once,
2 - moderate, a few times, and 3 - severe, frequent)
• Depression (Self-Rating Depression Scale)
Notes Able to use:
• Adverse events
Unable to use in analysis: no longer meet criteria for emotionalism, improved scores on frequency of
compulsive laughter measure, leaving the study early, Self-Rating Depression Scale, MMSE (data not
presented in appropriate format)
Funding: information not available
Risk of bias
Random sequence genera- Unclear risk Quote: "...and placebo were dosed in a random order (determined by the con-
tion (selection bias) troller...)" (pp. 1184)

Informed decisions.
Ohkawa 1989 (Continued)

Comments: the method of randomisation was not detailed.

(selection bias)
Blinding of participants Low risk Quote: "At examination, either the patients or nursing family members were
and personnel (perfor- interviewed..." (pp. 1184)
mance bias)
All outcomes Comments: the study authors stated that it was a double-blind placebo-con-
trolled trial.
Blinding of outcome as- Low risk Quote: "The examiner (other author), to whom the dosage order is un-
sessment (detection bias) known..." (pp. 1184)
All outcomes
Quote: "Also a psychiatrist, to whom dosage is unknown, performed the self-
rating depression scale (SDS)..." (pp. 1184)
Incomplete outcome data Low risk Quote: "At first ten subjects was [sic] selected for the experiment, but three
(attrition bias) have been excluded..." (pp. 1184)
All outcomes
Robinson 1993b

Number of arms: 2
Experimental arm: nortriptyline
Analysis: per protocol: 3 dropouts (treatment), excluded from analysis
Participants Geographical location: USA

Setting: university hospital
Diagnosis: stroke, on average 8.1 (SD 9.9 treatment) to 15.7 (SD 13.5 control) months prior to randomi-
sation
Exclusion criteria: participants with decreased levels of consciousness or moderate to severe aphasia
with deficits in comprehension
Number of participants included in treatment group: 15 (60% men, mean age 57.8 years, SD 10.1)
Number of participants included in control group: 14 (40% men, mean age 58.5 years, SD 11.8)
Emotionalism criteria: score of > 12 on the PLCS
Interventions Treatment: nortriptyline, 1 week at 20 mg, 2 weeks at 50 mg, 1 week at 70 mg, and 2 weeks at 100 mg
Outcomes • Emotionalism (PLCS scores)

• Depression (HDRS)

Informed decisions.
Robinson 1993b (Continued)

• Activities of Daily Living (John Hopkins Functioning Inventory)
• PLCS
• HDRS
• MMSE
• John Hopkins Functioning Inventory
Funding: "supported by Research Scientist Award MH-00163 (to R.G.R.) and grant MH-40355 from NIMH,
a New Investigator Award from the National Alliance for Research in Schizophrenia and Affective Dis-
orders (to S.E.S.) and a grant from the University of Buenos Aires (to S.E.S.). Nortriptyline and placebo
capsules used in the study were supplied by the Sandoz Pharmaceutical Corporation."
Risk of bias
Random sequence genera- Low risk Quote: "...in a single daily dose at bedtime after random number assignmen-
tion (selection bias) t..." (pp. 287)
Allocation concealment Low risk Quote: "The 28 patients participating in the treatment study were given nor-
(selection bias) triptyline or placebo (in identical capsules)..." (pp. 287)
Blinding of participants Low risk Quote: "Both the patients and the examiners were unaware of which treat-
and personnel (perfor- ment was being given." (pp. 287)
mance bias)
All outcomes
Blinding of outcome as- Low risk Quote: "Both the patients and the examiners were unaware of which treat-
sessment (detection bias) ment was being given." (pp. 287)
All outcomes
Incomplete outcome data Low risk Quote: "There was only one patient who dropped out during the course of the
(attrition bias) study." (pp. 287)
All outcomes
Quote: "...dropped out between weeks 2 and 4 because of complaints of seda-
tion." (pp. 287)

BDI: Beck Depression Inventory

CIBIC: Clinician Interview-Based Impression of Change
CT: computerised tomography
DSM: Diagnostic and Statistical Manual of Mental Disorders
HDRS: Hamilton Depression Rating Scale
MADRS: Montgomery Åsberg Depression Rating Scale
MMSE: Mini-Mental State Examination
MRI: magnetic resonance imaging
PLCS: Pathological Laughter and Crying Scale
SD: standard deviation
SSRI: selective serotonin reuptake inhibitors
Informed decisions.
VAS: Visual Analogue Scale
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Aizawa 1977 Methods: double-blind RCT

Participants: cerebrovascular disorders (including arteriosclerosis)
Intervention: cyclandelate for 4 weeks
Outcomes: data not currently available for those with 'emotionalism' at baseline
Allen 2018 Methods: not a RCT, i.e. non-interventional, cross-sectional, case-control study
Participants: nursing home residents with documented diagnosis of pseudobulbar affect
Intervention: dextromethorphan/quinidine
Atarashi 1988 Methods: randomisation unclear

Participants: stroke, including cerebral arteriosclerosis
Intervention: no placebo comparison
Bassi 1984 Methods: non-random, open-label

Participants: chronic cerebrovascular disorders
Chen 2010 Methods: not a RCT, i.e. case report and literature review
Intervention: quetiapine
ChiCTR-IPR-15007651 2015 Participants: no emotionalism at entry

Outcome: emotionalism is not an outcome of the study
Colamonico 2012 Methods: not a RCT, i.e. survey to estimate the impact or burden of pseudobulbar affect
D'Amico 2017 Methods: non-random, open-label
Doody 2014 Participants: ineligible study population, i.e. participants were adults who had pseudobulbar affect
after being diagnosed with dementia/Alzheimer's Disease
EUCTR2004-004803-39-DE Participants: no emotionalism at trial entry

2006
EUCTR2011-005541-12-DK Participants: no emotionalism at trial entry

2012
EUCTR2011-006130-16-SE Participants: no emotionalism at trial entry

2014
EUCTR2013-002253-30-DK Participants: no emotionalism at trial entry

2013
Formella 2017a Methods: non-random, open-label
Formella 2017b Methods: non-random, open-label
Kim 2017a Methods: not RCT
Kim 2017b Outcomes: data not currently available for those with 'emotionalism' at baseline pretreatment
Lawson 1969 Methods: randomised

Participants: hypertensive or ischaemic cerebral disease (number with stroke unclear)

Informed decisions.
Study Reason for exclusion

Intervention: method of randomisation makes placebo comparison ineffectual - no appropriate
washout period
Lorwatanapongsa 2020 Participants: no emotionalism at trial entry
Manzo 1998 Methods: not a RCT, i.e. qualitative study of pseudobulbar affect
Moller 2007 Methods: randomised

Participants: people with stroke and pathological crying
Intervention: citalopram for 30 days
Outcomes: emotionalism not investigated
Muller 1999 Methods: quasi-randomised, 2 active treatments

Participants: brain injury
Narushima 2002 Methods: double-blind, randomised

Participants: post-stroke
Intervention: prevention of depression
NCT01278498 2011 Aim: prevent emotionalism
NCT01799941 2013 Method: open-label trial
Ohtomo 1985 Methods: double-blind, randomised

Participants: cerebrovascular disorders, including arteriosclerosis
Intervention: tiapride for 5 weeks
Outcomes: data not currently available for those with 'emotionalism' at baseline
Otomo 1984 Methods: double-blind, randomised

Participants: cerebrovascular disorders
Rasmussen 2000 Methods: double-blind, randomised

Participants: post-stroke without depression, emotionalism not assessed at baseline
Sauve 2017 Participants: ineligible study population, i.e. participants were adults who had pseudobulbar affect
after being diagnosed with dementia/Alzheimer's Disease
Schiffer 1985 Methods: double-blind, cross-over

Participants: ineligible study population, i.e. multiple sclerosis (not stroke)
Schoedel 2014 Method: not a RCT, i.e. literature review
Seliger 1992 Method: non-random, open-label

Participants: people with stroke or multiple sclerosis (not stroke) and emotional incontinence
Udaka 1984 Methods: non-random, open-label

Participants: ineligible study population, i.e. diffuse cerebrovascular disease (not stroke)
Work 2011 Methods: not a RCT, i.e. a survey to estimate the overall prevalence of pseudobulbar affect and
quantify the extent to which it is diagnosed and treated
Yang 2015 Methods: not a RCT, i.e. a literature review

Participants: adults with pseudobulbar affect
Intervention: dextromethorphan/quinidine

Informed decisions.
RCT: randomised controlled trial
DATA AND ANALYSES
Comparison 1. Pharmaceutical interventions versus placebo
Outcome or subgroup title No. of studies No. of partici- Statistical method Effect size
pants
1.1 Emotionalism 4 Risk Ratio (M-H, Random, Subtotals only

95% CI)
1.1.1 50% reduction in emotionalism 1 19 Risk Ratio (M-H, Random, 0.26 [0.09, 0.77]
95% CI)
1.1.2 Improved score on Center for Neu- 1 28 Risk Ratio (M-H, Random, 0.20 [0.03, 1.50]
rologic Study - Lability Scale 95% CI)
1.1.3 Clinician Interview-Based Impres- 1 28 Risk Ratio (M-H, Random, 0.20 [0.03, 1.50]
sion of Change - improved score 95% CI)
1.1.4 Diminished tearfulness 3 164 Risk Ratio (M-H, Random, 0.32 [0.12, 0.86]
95% CI)
1.2 Emotionalism: mean scores at end of 1 Mean Difference (IV, Fixed, Subtotals only
treatment 95% CI)
1.2.1 Pathological Laughter and Crying 1 28 Mean Difference (IV, Fixed, -8.40 [-11.56,
Scale (high score = worse emotionalism) 95% CI) -5.24]
1.3 Depression: 1. Mean scores at end of 2 72 Std. Mean Difference (IV, -0.82 [-2.14, 0.51]
treatment Random, 95% CI)
1.3.1 Hamilton Depression Rating Scale 1 28 Std. Mean Difference (IV, -1.53 [-2.39,
(high score = more depressed) Random, 95% CI) -0.67]
1.3.2 Montgomery Åsberg Depression 1 44 Std. Mean Difference (IV, -0.18 [-0.77, 0.42]
Rating Scale (high score = more de- Random, 95% CI)
pressed)
1.4 Depression: 2. Average change in 2 72 Std. Mean Difference (IV, -0.05 [-0.72, 0.62]
scores between baseline and end of Random, 95% CI)
treatment
1.4.1 Hamilton Depression Rating Scale 1 28 Std. Mean Difference (IV, -0.43 [-1.18, 0.32]
(high score = more depressed) Random, 95% CI)
1.4.2 Montgomery Åsberg Depression 1 44 Std. Mean Difference (IV, 0.26 [-0.34, 0.85]
Rating Scale (high score = more de- Random, 95% CI)
pressed)
1.5 Cognitive functioning: mean scores 1 Mean Difference (IV, Fixed, Totals not select-
at end of treatment 95% CI) ed
1.5.1 Mini-Mental State Examination 1 Mean Difference (IV, Fixed, Totals not select-
(low score = cognitive impairment) 95% CI) ed

Informed decisions.
Outcome or subgroup title No. of studies No. of partici- Statistical method Effect size
pants
1.6 Activities of daily living: 1. Mean 1 Mean Difference (IV, Fixed, Subtotals only
scores at end of treatment 95% CI)
1.6.1 Johns Hopkins Functioning Inven- 1 28 Mean Difference (IV, Fixed, -1.40 [-5.22, 2.42]
tory (high score = worse function) 95% CI)
1.7 Adverse events: 1. Death 6 Risk Ratio (M-H, Random, Subtotals only
95% CI)
1.7.1 At end of treatment 6 172 Risk Ratio (M-H, Random, 0.59 [0.08, 4.50]
95% CI)
1.8 Adverse events: 2. All 2 Risk Ratio (M-H, Random, Subtotals only
95% CI)
1.8.1 Central nervous system events 2 56 Risk Ratio (M-H, Random, 1.00 [0.11, 9.08]
(e.g. confusion, sedation, tremor) 95% CI)
1.8.2 Gastrointestinal effects (e.g. con- 1 28 Risk Ratio (M-H, Random, 0.33 [0.01, 7.55]
stipation, diarrhoea) 95% CI)
1.8.3 Other events not listed above (e.g. 1 28 Risk Ratio (M-H, Random, 5.00 [0.26, 95.61]
dysuria, eye discomfort) 95% CI)
1.8.4 Recurrent stroke 1 28 Risk Ratio (M-H, Random, 1.00 [0.07, 14.45]
95% CI)
1.9 Adverse events: 3. Leaving the study 5 Risk Ratio (M-H, Random, Subtotals only
early (including death) 95% CI)
1.9.1 All dropouts and withdrawals 5 216 Risk Ratio (M-H, Random, 1.17 [0.38, 3.58]
95% CI)

Informed decisions.
Analysis 1.1. Comparison 1: Pharmaceutical interventions versus placebo, Outcome 1: Emotionalism
Intervention Placebo Risk Ratio (Non-event) Risk Ratio (Non-event)

Study or Subgroup Events Total Events Total Weight M-H, Random, 95% CI M-H, Random, 95% CI
1.1.1 50% reduction in emotionalism

Brown 1998 7 9 0 10 100.0% 0.26 [0.09 , 0.77]
Subtotal (95% CI) 9 10 100.0% 0.26 [0.09 , 0.77]
Total events: 7 0
Heterogeneity: Not applicable
Test for overall effect: Z = 2.43 (P = 0.02)
1.1.2 Improved score on Center for Neurologic Study - Lability Scale

Burns 1999 13 14 9 14 100.0% 0.20 [0.03 , 1.50]
Subtotal (95% CI) 14 14 100.0% 0.20 [0.03 , 1.50]
Total events: 13 9
1.1.3 Clinician Interview-Based Impression of Change - improved score

Burns 1999 13 14 9 14 100.0% 0.20 [0.03 , 1.50]
Subtotal (95% CI) 14 14 100.0% 0.20 [0.03 , 1.50]
Total events: 13 9
1.1.4 Diminished tearfulness

Burns 1999 14 14 9 14 10.0% 0.09 [0.01 , 1.50]
Choi-Kwon 2006 37 44 14 48 42.3% 0.22 [0.11 , 0.45]
Murray 2005 13 24 4 20 47.7% 0.57 [0.35 , 0.93]
Subtotal (95% CI) 82 82 100.0% 0.32 [0.12 , 0.86]
Total events: 64 27
Heterogeneity: Tau² = 0.46; Chi² = 7.21, df = 2 (P = 0.03); I² = 72%
0.002 0.1 1 10 500

Favours treatment Favours placebo
Analysis 1.2. Comparison 1: Pharmaceutical interventions versus

placebo, Outcome 2: Emotionalism: mean scores at end of treatment
Intervention Placebo Mean Difference Mean Difference

Study or Subgroup Mean SD Total Mean SD Total Weight IV, Fixed, 95% CI IV, Fixed, 95% CI
1.2.1 Pathological Laughter and Crying Scale (high score = worse emotionalism)
Robinson 1993b 1.2 2 14 9.6 5.7 14 100.0% -8.40 [-11.56 , -5.24]
Subtotal (95% CI) 14 14 100.0% -8.40 [-11.56 , -5.24]
Test for overall effect: Z = 5.20 (P < 0.00001)
-20 -10 0 10 20

Informed decisions.
Analysis 1.3. Comparison 1: Pharmaceutical interventions versus

placebo, Outcome 3: Depression: 1. Mean scores at end of treatment
Intervention Placebo Std. Mean Difference Std. Mean Difference

Study or Subgroup Mean SD Total Mean SD Total Weight IV, Random, 95% CI IV, Random, 95% CI
1.3.1 Hamilton Depression Rating Scale (high score = more depressed)

Robinson 1993b 2.1 2.4 14 11.3 7.9 14 47.3% -1.53 [-2.39 , -0.67]
Subtotal (95% CI) 14 14 47.3% -1.53 [-2.39 , -0.67]
1.3.2 Montgomery Åsberg Depression Rating Scale (high score = more depressed)
Murray 2005 7.67 6.56 24 9.09 9.14 20 52.7% -0.18 [-0.77 , 0.42]
Subtotal (95% CI) 24 20 52.7% -0.18 [-0.77 , 0.42]
Total (95% CI) 38 34 100.0% -0.82 [-2.14 , 0.51]

Test for overall effect: Z = 1.21 (P = 0.23) -10 -5 0 5 10
Test for subgroup differences: Chi² = 6.45, df = 1 (P = 0.01), I² = 84.5% Favours treatment Favours placebo
Analysis 1.4. Comparison 1: Pharmaceutical interventions versus placebo, Outcome

4: Depression: 2. Average change in scores between baseline and end of treatment
Intervention Placebo Std. Mean Difference Std. Mean Difference

Study or Subgroup Mean SD Total Mean SD Total Weight IV, Random, 95% CI IV, Random, 95% CI
1.4.1 Hamilton Depression Rating Scale (high score = more depressed)

Robinson 1993b -10 5.43 14 -7.3 6.69 14 44.3% -0.43 [-1.18 , 0.32]
Subtotal (95% CI) 14 14 44.3% -0.43 [-1.18 , 0.32]
1.4.2 Montgomery Åsberg Depression Rating Scale (high score = more depressed)
Murray 2005 -10.71 6.31 24 -12.61 8.206 20 55.7% 0.26 [-0.34 , 0.85]
Subtotal (95% CI) 24 20 55.7% 0.26 [-0.34 , 0.85]
Total (95% CI) 38 34 100.0% -0.05 [-0.72 , 0.62]

Test for overall effect: Z = 0.14 (P = 0.89) -10 -5 0 5 10
Test for subgroup differences: Chi² = 1.98, df = 1 (P = 0.16), I² = 49.5% Favours treatment Favours placebo
Analysis 1.5. Comparison 1: Pharmaceutical interventions versus placebo,

Outcome 5: Cognitive functioning: mean scores at end of treatment

Study or Subgroup Mean SD Total Mean SD Total IV, Fixed, 95% CI IV, Fixed, 95% CI
1.5.1 Mini-Mental State Examination (low score = cognitive impairment)

Robinson 1993b 26.5 4.8 14 26.8 3 14 -0.30 [-3.27 , 2.67]
-10 -5 0 5 10

Informed decisions.

Outcome 6: Activities of daily living: 1. Mean scores at end of treatment

Study or Subgroup Mean SD Total Mean SD Total Weight IV, Fixed, 95% CI IV, Fixed, 95% CI
1.6.1 Johns Hopkins Functioning Inventory (high score = worse function)

Robinson 1993b 3.7 5 14 5.1 5.3 14 100.0% -1.40 [-5.22 , 2.42]
Subtotal (95% CI) 14 14 100.0% -1.40 [-5.22 , 2.42]
-10 -5 0 5 10
Analysis 1.7. Comparison 1: Pharmaceutical interventions versus placebo, Outcome 7: Adverse events: 1. Death
Intervention Placebo Risk Ratio Risk Ratio

1.7.1 At end of treatment

Andersen 1993 0 16 0 16 Not estimable
Brown 1998 0 10 0 10 Not estimable
Burns 1999 1 14 1 14 58.1% 1.00 [0.07 , 14.45]
Murray 2005 0 24 1 20 41.9% 0.28 [0.01 , 6.52]
Ohkawa 1989 0 10 0 10 Not estimable
Robinson 1993b 0 14 0 14 Not estimable
Subtotal (95% CI) 88 84 100.0% 0.59 [0.08 , 4.50]
Total events: 1 2
0.01 0.1 1 10 100


Informed decisions.
Analysis 1.8. Comparison 1: Pharmaceutical interventions versus placebo, Outcome 8: Adverse events: 2. All

1.8.1 Central nervous system events (e.g. confusion, sedation, tremor)

Burns 1999 0 14 1 14 50.0% 0.33 [0.01 , 7.55]
Robinson 1993b 1 14 0 14 50.0% 3.00 [0.13 , 67.91]
Subtotal (95% CI) 28 28 100.0% 1.00 [0.11 , 9.08]
Total events: 1 1
1.8.2 Gastrointestinal effects (e.g. constipation, diarrhoea)

Burns 1999 0 14 1 14 100.0% 0.33 [0.01 , 7.55]
Subtotal (95% CI) 14 14 100.0% 0.33 [0.01 , 7.55]
Total events: 0 1
1.8.3 Other events not listed above (e.g. dysuria, eye discomfort)
Burns 1999 2 14 0 14 100.0% 5.00 [0.26 , 95.61]
Subtotal (95% CI) 14 14 100.0% 5.00 [0.26 , 95.61]
Total events: 2 0
1.8.4 Recurrent stroke

Burns 1999 1 14 1 14 100.0% 1.00 [0.07 , 14.45]
Subtotal (95% CI) 14 14 100.0% 1.00 [0.07 , 14.45]
Total events: 1 1
0.005 0.1 1 10 200


Outcome 9: Adverse events: 3. Leaving the study early (including death)

1.9.1 All dropouts and withdrawals

Brown 1998 1 10 0 10 10.7% 3.00 [0.14 , 65.90]
Burns 1999 3 14 1 14 18.6% 3.00 [0.35 , 25.46]
Choi-Kwon 2006 0 44 3 48 11.6% 0.16 [0.01 , 2.93]
Murray 2005 8 24 10 20 46.8% 0.67 [0.33 , 1.36]
Robinson 1993b 4 18 0 14 12.2% 7.11 [0.41 , 121.88]
Subtotal (95% CI) 110 106 100.0% 1.17 [0.38 , 3.58]
Total events: 16 14
0.01 0.1 1 10 100


Informed decisions.
ADDITIONAL TABLES
Table 1. Characteristics of 'dropout' studies

Study ID Methods Participants Interventions Outcomes Notes
Aizawa 1977 Study design: Geographical location: Japan Arm 1: cyclan- • Global im- Unable to use:
randomised, delate 900 provement all data (data
parallel de- Setting: 50 institutes across Japan mg/day and rating not presented
sign cinnarizine 75 • Improve- by 'emotion-
Number of mg/day ment rating alism at base-
arms: 2 Diagnosis: stroke over 1 month ago of subjective line', unable
Arm 2: symptoms to exclude
Arm 1: cyclan- Inclusion criteria: 1) inpatients and outpa- matched people with
delate 900 placebo and • Improve-
tients who had cerebral infarct, intracra- ment rating cerebral arte-
mg/day + cin- nial bleeding, transient cerebral ischaemia, cinnarizine 75 riosclerosis
of psychi-
narizine 75 and cerebral arteriosclerosis mg/day and transient
atric symp-
mg/day ischaemic at-
Duration: toms
Exclusion criteria: 1) expectant mothers; 2) tack)
Arm 2: place- with glaucoma, and 3) severe concomitant treatment • Improve-
bo + cinnar- diseases continued for ment rating
izine 75 mg/ 4 weeks of neurolog-
day Age: overall mean age not reported ical symp-
Analysis: per toms
protocol Number of participants included in Arm 1: • Global utility
188 (68% men, age details unclear) rating
Numbers of participants included in Arm 2: • Overall safe-
190 (68% men, age details unclear) ty rating
Ohtomo 1985 Study design: Geographical location: Japan Arm 1: • Severity of Unable to use:
randomised, tiapride 75 psychiatric all data (data
parallel de- Setting: unclear mg/day for 1 symptoms not presented
sign week, dose • Activities of by 'emotion-
Number of escalation to daily living alism at base-
arms: 2 Diagnosis: cerebral haemorrhage, sub- 150 mg to 225 • Somatic line', unable
arachnoid haemorrhage, cerebral infarc- mg/day for 5 complaints to exclude
Arm 1: weeks accord- people with
tion, cerebral apoplexy sequelae, cerebral
tiapride 75 ing to clinical cerebral arte-
arteriosclerosis
mg/day for 1 response riosclerosis)
week, dose Inclusion criteria: 1) people with cerebral
escalation to arteriosclerosis Arm 2:
150 mg to 225 matched
mg/day for 5 Exclusion criteria: 1) severe aphasia, 2) se- placebo
weeks accord- vere dementia, 3) drug dependence, 4) in-
ing to clinical adequate conditions for the study Duration:
response treatment
Age: overall mean age not reported continued for
Arm 2: place- 6 weeks
bo + cinnar- Number of participants included in Arm 1:
izine 75 mg/ 141 (54% men, age details unclear)
day
Analysis: per Number of participants included in Arm 2:
protocol 147 (61% men, age details unclear)
Kim 2017b Study design: Geographical location: South Korea Arm 1: esci- Primary out- Unable to use:
randomised, talopram 10 comes: all data (data
parallel de- Setting: 17 hospitals across South Korea mg/day presented by
sign Arm 2: • Occurrence 'emotionalism
Number of participants: 478 of moder-
Number of matched at baseline' 1
arms: 2 placebo ate or se- month post-
vere depres- treatment)
Informed decisions.
Table 1. Characteristics of 'dropout' studies (Continued)

Arm 1: esci- Diagnosis: stroke, had an acute ischaemic Duration: 12 sive symp-
talopram 10 stroke or intracerebral haemorrhage within weeks toms
mg/day the previous 21 days (confirmed by MRI or
CT) Secondary out-
Arm 2: place- comes:
bo Inclusion criteria: 1) > 20 years, 2) had
Analysis: per an acute ischaemic stroke or intracere- • Occurrence
protocol bral haemorrhage within the previous 21 of emotion-
days (confirmed by MRI or CT), 3) modified al inconti-
Rankin Scale score > 2 nence (Kim's
criteria)
Exclusion criteria: 1) history of diagnosed • Anger prone-
depression or other psychiatric diseases ness (mod-
before the index stroke; 2) severe demen- ified Spiel-
tia, 3) aphasia, 4) exhibited strong suicidal berger Trait
thoughts Anger Scale/
National In-
Age: overall mean age not reported stitutes of
Number of participants included in Arm 1: Health
210 (57% men, mean age 64 (13) years Stroke Scale
scores)
Number of participants included in Arm 2: • Modified
195 (65% men, mean age 64 (12) years Rankin Scale
• Barthel In-
dex
CT: computerised tomography

MRI: magnetic resonance imaging
APPENDICES
Appendix 1. CENTRAL
ID Search
#1 [Cerebrovascular Disorders] this term only
#2 [Basal Ganglia Cerebrovascular Disease] explode all trees
#3 [Brain Ischemia] explode all trees
#4 [Carotid Artery Diseases] explode all trees
#5 [Intracranial Arterial Diseases] explode all trees
#6 [Intracranial Arteriovenous Malformations] explode all trees
#7 [Intracranial Embolism and Thrombosis] explode all trees
#8 [Intracranial Hemorrhages] explode all trees
#9 [Stroke] this term only
#10 [Hemorrhagic Stroke] this term only

Informed decisions.
(Continued)
#11 [Ischemic Stroke] explode all trees
#12 [Ischemic Stroke] explode all trees
#13 [Stroke, Lacunar] this term only
#14 [Vasospasm, Intracranial] this term only
#15 [Vertebral Artery Dissection] this term only
#16 [Stroke Rehabilitation] this term only
#17 (stroke or poststroke or post-stroke or cerebrovasc* or brain next vasc* or cerebral next vasc* or
cva* or apoplex* or SAH):ti,ab,kw (Word variations have been searched)
#18 ((brain* or cerebr* or cerebell* or intracran* or intracerebral) near/5 (isch*emi* or infarct* or

thrombo* or emboli* or occlus*)):ti,ab,kw (Word variations have been searched)
#19 ((brain* or cerebr* or cerebell* or intracerebral or intracranial or subarachnoid) near/5 (haemor-

rhage* or hemorrhage* or haematoma* or hematoma* or bleed*)):ti,ab,kw (Word variations have
been searched)
#20 {or #1-#19}
#21 [Laughter] this term only
#22 [Crying] this term only
#23 [Emotions] explode all trees
#24 [Affective Symptoms] this term only
#25 emotion* or laugh* or cry* or weep* or tearful* or pseudobulbar affect:ti,ab,kw (Word variations
have been searched)
#26 {or #21-#25}
#27 #20 and #26 with Publication Year from 2008 to 2017
Appendix 2. MEDLINE
Search strategy for MEDLINE, May 2022
1. cerebrovascular disorders/ or exp basal ganglia cerebrovascular disease/ or exp brain ischemia/ or exp carotid artery diseases/ or exp
intracranial arterial diseases/ or exp "intracranial embolism and thrombosis"/ or exp intracranial hemorrhages/ or stroke/ or exp brain
infarction/ or vasospasm, intracranial/ or vertebral artery dissection/
2. (stroke or poststroke or post-stroke or cerebrovasc$ or brain vasc$ or cerebral vasc$ or cva$ or apoplex$ or SAH).tw.
3. ((brain$ or cerebr$ or cerebell$ or intracran$ or intracerebral) adj5 (isch?emi$ or infarct$ or thrombo$ or emboli$ or occlus$)).tw.
4. ((brain$ or cerebr$ or cerebell$ or intracerebral or intracranial or subarachnoid) adj5 (haemorrhage$ or hemorrhage$ or haematoma
$ or hematoma$ or bleed$)).tw.
5. 1 or 2 or 3 or 4
6. crying/ or laughter/
7. affective symptoms/ or exp emotions/
8. (laugh$ or cry$ or weep$ or emotional$ or pseudobulbar affect).tw.
9. 6 or 7 or 8

Informed decisions.
10.randomized controlled trial.pt.

11.controlled clinical trial.pt.
12.placebo.ab.
13.clinical trials as topic.sh.
14.(random$ or RCT or RCTs).tw.
15.trial.ti.
16.or/10-15
17.exp animals/ not humans.sh.
18.16 not 17
19.5 and 9 and 18
Appendix 3. Embase
Search strategy for Embase, May 2022
1. cerebrovascular disease/ or brain disease/ or exp basal ganglion hemorrhage/ or exp brain hemangioma/ or exp brain hematoma/ or
exp brain hemorrhage/ or exp brain infarction/ or exp brain ischemia/ or exp carotid artery disease/ or exp cerebral artery disease/ or exp
cerebrovascular accident/ or exp cerebrovascular malformation/ or exp intracranial aneurysm/ or exp occlusive cerebrovascular disease/
or exp vertebrobasilar insufficiency/
2. (stroke$ or poststroke or apoplex$ or cerebral vasc$ or brain vasc$ or cerebrovasc$ or cva$ or SAH).tw.
3. ((brain or cerebr$ or cerebell$ or vertebrobasil$ or hemispher$ or intracran$ or intracerebral or infratentorial or supratentorial or middle
cerebral artery or MCA$ or anterior circulation or posterior circulation or basilar artery or vertebral artery or space-occupying) adj5 (isch?
emi$ or infarct$ or thrombo$ or emboli$ or occlus$ or hypoxi$)).tw.
4. ((brain$ or cerebr$ or cerebell$ or intracerebral or intracran$ or parenchymal or intraparenchymal or intraventricular or infratentorial
or supratentorial or basal gangli$ or putaminal or putamen or posterior fossa or hemispher$ or subarachnoid) adj5 (h?emorrhag$ or h?
ematoma$ or bleed$)).tw.
5. 1 or 2 or 3 or 4
6. crying/ or pathological crying/ or laughter/ or pathological laughter/ or nonverbal communication/
7. emotion/ or affective neurosis/
8. emotional stability/ or emotionality/
9. (emotion$ or laugh$ or cry$ or weep$ or tearful$ or pseudobulbar affect).tw.
10. 6 or 7 or 8 or 9
11. Randomized Controlled Trial/ or "randomized controlled trial (topic)"/
12. Randomization/
13. Controlled clinical trial/ or "controlled clinical trial (topic)"/
14. control group/ or controlled study/
15. clinical trial/ or "clinical trial (topic)"/ or phase 1 clinical trial/ or phase 2 clinical trial/ or phase 3 clinical trial/ or phase 4 clinical trial/
16. Crossover Procedure/
17. Double Blind Procedure/
18. Single Blind Procedure/ or triple blind procedure/
19. placebo/ or placebo effect/
20. (random$ or RCT or RCTs).tw.
21. (controlled adj5 (trial$ or stud$)).tw.
22. (clinical$ adj5 trial$).tw.
23. ((control or treatment or experiment$ or intervention) adj5 (group$ or subject$ or patient$)).tw.
24. (quasi-random$ or quasi random$ or pseudo-random$ or pseudo random$).tw.
25. ((control or experiment$ or conservative) adj5 (treatment or therapy or procedure or manage$)).tw.
26. ((singl$ or doubl$ or tripl$ or trebl$) adj5 (blind$ or mask$)).tw.
27. (cross-over or cross over or crossover).tw.
28. (placebo$ or sham).tw.
29. trial.ti.
30. (assign$ or allocat$).tw.
31. controls.tw.
32. or/11-31
33. (exp animals/ or exp invertebrate/ or animal experiment/ or animal model/ or animal tissue/ or animal cell/ or nonhuman/) not (human/
or normal human/ or human cell/)
34. 5 and 10 and 32
35. 34 not 33
Search results: 644

Informed decisions.
Appendix 4. CINAHL
Search strategy for CINAHL, May 2022
# Query
S1 (MH "Cerebrovascular Disorders") OR (MH "Basal Ganglia Cerebrovascular Disease+") OR (MH

"Carotid Artery Diseases+") OR (MH "Cerebral Ischemia+") OR (MH "Cerebral Vasospasm") OR (MH
"Intracranial Arterial Diseases+") OR (MH "Intracranial Embolism and Thrombosis") OR (MH "In-
tracranial Hemorrhage+") OR (MH "Stroke") OR (MH "Vertebral Artery Dissections")
S2 (MH "Stroke Patients") OR (MH "Stroke Units")
S3 TI (stroke* or poststroke or apoplex* or cerebral vasc* or brain vasc* or cerebrovasc* or cva* or

SAH ) or AB ( stroke* or poststroke or apoplex* or cerebral vasc* or brain vasc* or cerebrovasc* or
cva* or SAH)
S4 TI (brain or cerebr* or cerebell* or vertebrobasil* or hemispher* or intracran* or intracerebral or in-

fratentorial or supratentorial or middle cerebral artery or MCA* or anterior circulation or posterior
circulation or basilar artery or vertebral artery or space-occupying ) or AB ( brain or cerebr* or cere-
bell* or vertebrobasil* or hemispher* or intracran* or intracerebral or infratentorial or supratentor-
ial or middle cerebral artery or MCA* or anterior circulation or posterior circulation or basilar artery
or vertebral artery or space-occupying)
S5 TI (ischemi* or ischaemi* or infarct* or thrombo* or emboli* or occlus* or hypoxi* ) or AB ( ischemi*

or ischaemi* or infarct* or thrombo* or emboli* or occlus* or hypox*)
S6 S4 and S5
S7 TI (brain* or cerebr* or cerebell* or intracerebral or intracran* or parenchymal or intraparenchy-

mal or intraventricular or infratentorial or supratentorial or basal gangli* or putaminal or putamen
or posterior fossa or hemispher* or subarachnoid) or AB (brain* or cerebr* or cerebell* or intrac-
erebral or intracran* or parenchymal or intraparenchymal or intraventricular or infratentorial or
supratentorial or basal gangli* or putaminal or putamen or posterior fossa or hemispher* or sub-
arachnoid)
S8 TI (haemorrhage* or hemorrhage* or haematoma* or hematoma* or bleed* ) or AB ( haemorrhage*

or hemorrhage* or haematoma* or hematoma* or bleed*)
S9 S7 and S8
S10 S1 OR S2 OR S3 OR S6 OR S9
S11 (MH "Emotions+") OR (MH "Affective Symptoms+")
S12 (MH "Laughter") OR (MH "Crying") OR (MH "Nonverbal Communication")
S13 TI ((laugh* or cry* or weep* or emotional* or pseudobulbar affect) ) OR AB ( (laugh* or cry* or

weep* or emotional* or pseudobulbar affect))
S14 S11 OR S12 OR S13
S15 (MH "Randomized Controlled Trials") or (MH "Random Assignment") or (MH "Random Sample+")
S16 (MH "Clinical Trials") or (MH "Intervention Trials") or (MH "Therapeutic Trials")
S17 (MH "Control (Research)") or (MH "Control Group") or (MH "Placebos") or (MH "Placebo Effect")

Informed decisions.
(Continued)
S18 (MH "Crossover Design") OR (MH "Quasi-Experimental Studies")
S19 (MH "Crossover Design") OR (MH "Quasi-Experimental Studies")
S20 PT (clinical trial or randomized controlled trial)
S21 TI (random* or RCT or RCTs) or AB (random* or RCT or RCTs)
S22 TI (controlled N5 (trial* or stud*)) or AB (controlled N5 (trial* or stud*))
S23 TI (clinical* N5 trial*) or AB (clinical* N5 trial*)
S24 TI ((control or treatment or experiment* or intervention) N5 (group* or subject* or patient*)) or AB

((control or treatment or experiment* or intervention) N5 (group* or subject* or patient*))
Appendix 5. PsycINFO
Search strategy for PsycINFO, May 2022
1. cerebrovascular disorders/ or cerebral hemorrhage/ or exp cerebral ischemia/ or cerebral small vessel disease/ or cerebrovascular
accidents/ or subarachnoid hemorrhage/
2. (stroke$ or poststroke or apoplex$ or cerebral vasc$ or brain vasc$ or cerebrovasc$ or cva$ or SAH).tw.
3. ((brain or cerebr$ or cerebell$ or vertebrobasil$ or hemispher$ or intracran$ or intracerebral or infratentorial or supratentorial or middle
cerebral artery or MCA$ or anterior circulation or posterior circulation or basilar artery or vertebral artery or space-occupying) adj5 (isch?
emi$ or infarct$ or thrombo$ or emboli$ or occlus$ or hypoxi$)).tw.
4. ((brain$ or cerebr$ or cerebell$ or intracerebral or intracran$ or parenchymal or intraparenchymal or intraventricular or infratentorial
or supratentorial or basal gangli$ or putaminal or putamen or posterior fossa or hemispher$ or subarachnoid) adj5 (h?emorrhag$ or h?
ematoma$ or bleed$)).tw.
5. hemiparesis/ or hemiplegia/
6. (hemipleg$ or hemipar$ or paresis or paraparesis or paretic).tw.
7. 1 or 2 or 3 or 4 or 5 or 6
8. exp emotions/ or emotional adjustment/ or emotional stability/ or emotional instability/ or "resilience (psychological)"/ or exp emotional
responses/ or "emotionality (personality)"/ or emotional states/ or emotional adjustment/ or emotional control/ or emotionally disturbed/
9. "crying/ or laughter/ or nonverbal communication/ or distress/"
10. (emotion$ or laugh$ or cry$ or weep$ or tearful$ or pseudobulbar affect).tw.
11. 8 or 9 or 10
12. clinical trials/ or treatment effectiveness evaluation/ or placebo/
13. (random$ or RCT or RCTs).tw.
14. (controlled adj5 (trial$ or stud$)).tw.
15. (clinical$ adj5 trial$).tw.
16. ((control or treatment or experiment$ or intervention) adj5 (group$ or subject$ or patient$)).tw.
17. (quasi-random$ or quasi random$ or pseudo-random$ or pseudo random$).tw.
18. ((control or experiment$ or conservative) adj5 (treatment or therapy or procedure or manage$)).tw.
19. ((singl$ or doubl$ or tripl$ or trebl$) adj5 (blind$ or mask$)).tw.
20. (cross-over or cross over or crossover).tw.
21. (placebo$ or sham).tw.
22. trial.ti.
23. (assign$ or allocat$).tw.
24. controls.tw.
25. or/12-24
26. 7 and 11 and 25
Appendix 6. BIOSIS Previews

Search strategy for BIOSIS, May 2022

Informed decisions.
# Query
S1 TS=(stroke or poststroke or post-stroke or cerebrovasc* or brain vasc* or cerebral vasc* or cva* or

apoplex* or SAH)
S2 TS=((brain* or cerebr* or cerebell* or intracran* or intracerebral) NEAR/5 (isch$emi* or infarct* or

thrombo* or emboli* or occlus*))
S3 TS=((brain* or cerebr* or cerebell* or intracerebral or intracranial or subarachnoid) NEAR/5 (haem-

orrhage* or hemorrhage* or haematoma* or hematoma* or bleed*))
S4 #3 OR #2 OR #1
S5 TS=(emotion* or laugh* or cry* or weep* or tearful* or pseudobulbar affect)
S6 TS=(random* or RCT or RCTs)
S7 TS=(controlled NEAR/5 (trial* or stud*))
S8 TS=(clinical* NEAR/5 trial*)
S9 TS=((control or treatment or experiment* or intervention) NEAR/5 (group* or subject* or patient*))
S10 TS=(quasi-random* or quasi random* or pseudo-random* or pseudo random*)
S11 TS=((control or experiment* or conservative) NEAR/5 (treatment or therapy or procedure or man-

age*))
S12 TS=((singl* or doubl* or tripl* or trebl*) NEAR/5 (blind* or mask*))
S13 TS=(cross-over or cross over or crossover)
S14 TS=(placebo* or sham)
S15 TS=trial
S16 TS=(assign* or allocat*)
S17 TS=controls
S18 #17 OR #16 OR #15 OR #14 OR #13 OR #12 OR #11 OR #10 OR #9 OR #8 OR #7 OR #6
S19 #18 AND #5 AND #4
Appendix 7. Web of Science

Search strategy for Web of Science, May 2022
The following indexes Science Citation Index Expanded (SCI-EXPANDED), Social Sciences Citation Index (SSCI), and Arts & Humanities
Citation Index (A&HCI) within Web of Science were searched from January 2002 to May 2018.
# Query

Informed decisions.
(Continued)
1 TS=(stroke or poststroke or post-stroke or cerebrovasc* or brain vasc* or cerebral vasc* or cva* or

apoplex* or SAH)
Indexes=SCI-EXPANDED, SSCI, A&HCI Timespan=2002-2018
2 TS=((brain* or cerebr* or cerebell* or intracran* or intracerebral) NEAR/5 (isch$emi* or infarct* or

thrombo* or emboli* or occlus*))
3 TS=((brain* or cerebr* or cerebell* or intracerebral or intracranial or subarachnoid) NEAR/5 (h?em-

orrhage* or haematoma* or hematoma* or bleed*))
4 #3 OR #2 OR #1
5 TS=(emotion* or laugh* or cry* or weep* or tearful* or pseudobulbar affect)

6 TS=(random* or RCT or RCTs)

7 TS=(controlled NEAR/5 (trial* or stud*))

8 TS=(clinical* NEAR/5 trial*)

9 TS=((control or treatment or experiment* or intervention) NEAR/5 (group* or subject* or patient*))

10 TS=(quasi-random* or quasi random* or pseudo-random* or pseudo random*)

11 TS=((control or experiment* or conservative) NEAR/5 (treatment or therapy or procedure or man-

age*))
12 TS=((singl* or doubl* or tripl* or trebl*) NEAR/5 (blind* or mask*))

13 TS=(cross-over or cross over or crossover)

14 TS=(placebo* or sham)
15 TS=trial
16 TS=(assign* or allocat*)
17 TS=controls
18 #17 OR #16 OR #15 OR #14 OR #13 OR #12 OR #11 OR #10 OR #9 OR #8 OR #7 OR #6


Informed decisions.
(Continued)
19 #18 AND #5 AND #4

Appendix 8. Other resources

Additional searches
We searched online clinical trials and research registers in May 2022.
• US National Institutes of Health Ongoing Trials Register ClinicalTrials.gov (clinicaltrials.gov/)

◦ (emotion OR laughing OR cry OR weep OR tearful OR pseudobulbar affect ) AND Intracranial Hemorrhages OR Carotid Artery Diseases
OR Brain Ischemia OR Cerebral Hemorrhage OR Cerebrovascular Disorders OR Stroke [DISEASE]
• World Health Organization International Clinical Trials Registry Platform (who.int/ictrp/search/en/)
◦ Title: (stroke AND emotion OR stroke AND laughing OR stroke AND cry OR stroke AND weep OR stroke AND tearful OR stroke AND
pseudobulbar affect)
ProQuest Dissertations and Theses Database was also searched on May 2022.
WHAT'S NEW
Date Event Description
26 May 2022 New search has been performed The searches have been updated and an error in the GRADE table
(absolute effects for emotionalism: diminished tearfulness) cor-
rected. We found no new trials for inclusion, so the total number
of included studies remains at seven, with 239 participants.
26 May 2022 New citation required but conclusions No change to the conclusions
have not changed
HISTORY
Protocol first published: Issue 3, 2002
Review first published: Issue 2, 2004
22 October 2018 New citation required but conclusions Conclusions not changed
have not changed
14 May 2018 New search has been performed The searches and risk of bias tables have been updated and a
GRADE table added. We found no new trials for inclusion, so the
total number of included studies remains at seven, with 239 par-
ticipants. Two trials were of cross-over design and outcome da-
ta were not available from the first phase (precross-over) in an
appropriate format for inclusion as a parallel randomised con-
trolled trial (RCT). Thus, the results of the review were based on
five trials with 213 participants. One trial appears to meet the in-
clusion criteria for the review, but data are not available in a for-
mat suitable for including in the analyses (Kim 2017b).
25 September 2009 New citation required but conclusions The first author has changed and there is also a new author for
have not changed this version of the review.

Informed decisions.
20 August 2009 New search has been performed This is a substantive amendment. The searches have been up-
dated. Two new trials have been added, making a total of sev-
en trials with 239 participants. Two trials appear to meet the re-
view inclusion criteria, but information is not available in a for-
mat suitable for pooling. Three further trials have been excluded.
14 April 2008 Amended Converted to new review format
CONTRIBUTIONS OF AUTHORS
SA: contributed to writing the review, completed title screening and inclusion/exclusion review, extracted any additional data needed, and
updated the risk of bias and summary of findings tables.
AH: contributed to writing the protocol and reviewed each version of this review.
MH: wrote the protocol, completed title/abstract screening, extracted data, entered data, contributed to writing the review, and oversaw
each version of this review.
DECLARATIONS OF INTEREST
SA: has declared that they have no conflict of interest.
AH: has declared that they have no conflict of interest.
MH: has declared that they have no conflict of interest.
SOURCES OF SUPPORT
Internal sources
• The George Institute for Global Health, Australia
In kind
External sources
• Stroke Society of Australasia, Australia
Overseas Study Scholarship

• Academic Unit of Psychiatry, University of Leeds, UK
In kind
• Division of Clinical Neurosciences, University of Edinburgh, UK
In kind
• Clinical Trials Research Unit, University of Auckland, New Zealand
In kind
DIFFERENCES BETWEEN PROTOCOL AND REVIEW

We planned the primary emotionalism endpoint to be the proportion of participants who, at the end of treatment, met the criteria for
emotionalism that the study authors applied in recruiting to the trial. However, data for this endpoint were not available.
We added a new method to measure emotionalism to the primary outcomes (see Primary outcomes).
There were methodological changes to the protocol and the review. We split the sensitivity analysis section into Subgroup analysis and
investigation of heterogeneity and Sensitivity analysis. We planned to perform subgroup analyses to examine the influence of date of
publication, sample size, duration of follow-up, treatment type, stroke severity, high (over 20%) number of dropouts, and blinded versus
unblinded outcome assessors. We did not perform these because of the small number of trials and participants.
When the protocol for this review was written (2004), we had intended to include outcomes relating to function (such as those measured
by the Modified Rankin Scale) and quality of life. We removed these outcomes because the Cochrane Review 'Selective serotonin reuptake

Informed decisions.
inhibitors (SSRIs) for stroke recovery' covers outcomes relating to function, and quality of life has not been a focus of included trials since
2004.
For all dichotomous outcomes, we changed odds ratios (ORs) with 95% confidence intervals (CIs) to risk ratios (RRs) with 95% CIs (see Data
and analyses).
INDEX TERMS
Medical Subject Headings (MeSH)

Antidepressive Agents [therapeutic use]; Crying [psychology]; Emotions; Pharmaceutical Preparations; Randomized Controlled Trials
as Topic; *Stroke [drug therapy] [psychology]
MeSH check words

Humans


Cochrane: Library

Uploaded by

Copyright:

Available Formats

Cochrane: Library

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Cochrane: Library

Uploaded by

Copyright:

Available Formats

Cochrane

Pharmaceutical interventions for emotionalism after stroke

Allida S, House A, Hackett ML

Allida S, House A, Hackett ML.

Pharmaceutical interventions for emotionalism after stroke (Review)

Pharmaceutical interventions for emotionalism after stroke (Review) i

Pharmaceutical interventions for emotionalism after stroke

Sabine Allida1, Allan House2, Maree L Hackett1

Contact: Maree L Hackett, [email protected].

Editorial group: Cochrane Stroke Group.

Data collection and analysis

PLAIN LANGUAGE SUMMARY

Medicines for emotionalism after stroke

We identified studies through searches conducted on 26 May 2022.

What are the limitations of the evidence?

Pharmaceutical interventions for emotionalism after stroke (Review) 2

Pharmaceutical interventions for emotionalism after stroke (Review) 3

Patient or population: emotionalism after stroke

Cochrane Database of Systematic Reviews

GRADE Working Group grades of evidence

Cochrane Database of Systematic Reviews

Pharmaceutical interventions for emotionalism after stroke (Review) 7

Pharmaceutical interventions for emotionalism after stroke (Review) 8

Measures of treatment effect Data synthesis

Pharmaceutical interventions for emotionalism after stroke (Review) 9

RESULTS duplicates, we screened 5422 titles and abstracts and excluded

Pharmaceutical interventions for emotionalism after stroke (Review) 10

Pharmaceutical interventions for emotionalism after stroke (Review) 11

Pharmaceutical interventions for emotionalism after stroke (Review) 12

Random sequence generation (selection bias)

0% 25% 50% 75% 100%

Low risk of bias Unclear risk of bias High risk of bias

Pharmaceutical interventions for emotionalism after stroke (Review) 13

Blinding of participants and personnel (performance bias): All outcomes

Selective reporting (reporting bias)

Pharmaceutical interventions for emotionalism after stroke (Review) 14

Allocation improvement (reduction) in tearfulness when compared to placebo

We judged the risk of other potential sources of bias as Cognitive functioning

Pharmaceutical interventions for emotionalism after stroke (Review) 16

1993b). Ideally, future trials should limit inclusion to participants Imprecision

Limitations in study design or execution Implications for practice

Indirectness of evidence • use a standardised method to diagnose emotionalism,

• include careful assessment and complete reporting of adverse ACKNOWLEDGEMENTS

Pharmaceutical interventions for emotionalism after stroke (Review) 18

Pharmaceutical interventions for emotionalism after stroke (Review) 19

Bassi 1984 {published data only} www.www.clinicaltrialsregister.eu/ctr-search/

Pharmaceutical interventions for emotionalism after stroke (Review) 20

Rasmussen 2000 {published data only}

Pharmaceutical interventions for emotionalism after stroke (Review) 21

APA 2017 Goldberg 1972

Lawson 1969 Review Manager [Computer program]

Pharmaceutical interventions for emotionalism after stroke (Review) 23

References to other published versions of this review House 2004

Characteristics of included studies [ordered by study ID]

Methods Study design: randomised, cross-over design

Participants Geographical location: Denmark

Outcomes • Improvements in crying history

Pharmaceutical interventions for emotionalism after stroke (Review) 24

Andersen 1993 (Continued)