Patients With Cerebrovascular Disease and Traumatic Brain Injury

Cognitive Impairment and Delirium

Approximately 30–40% of patients experience delirium in the first week after a stroke with those who suffered a hemorrhagic stroke at higher risk. Since delirium may be difficult to differentiate from cognitive deficits that result from focal brain lesions, it is essential to assess for waxing and waning symptomatology that is characteristic of delirious states. Risk factors for delirium include older age, impaired vision, impaired swallowing, and prior history of stroke. The presence of delirium post-stroke confers a poorer overall prognosis and is associated with longer length of hospitalization, increased risk of dementia, and increased overall mortality.

Cognitive impairment is also common post-stroke and occurs in about one-fourth of patients examined at 3 months after the event. The term “vascular cognitive impairment (VCI)” includes the full spectrum from mild to severe cognitive impairment both in people with cerebrovascular disease and those with vascular (post-stroke) dementia. Post-stroke risk factors of VCI include: severity of infarct, large size of lesion, older age, low education, history of diabetes or atrial fibrillation, and number of recurrent strokes. The first year after stroke confers the highest risk of developing dementia, with an estimated incidence of 20–30%.

Post-Stroke Depression

Post-stroke depression (PSD) is the prototypical acute psychiatric manifestation of stroke. It is common; approximately 30% of patients meet criteria for major depression in the first 3 months after stroke, and this often develops into a chronic remitting–relapsing condition. Stroke-associated depression may reduce survival and increase the risk of recurrent vascular events as well as of cognitive impairment. Risk factors for PSD include a history of depression, pre-stroke functional impairment, living alone, post-stroke social isolation, and possibly female gender.

Both biological and psychological theories of etiology have been studied. Biological hypotheses include lesion location (e.g., of the left frontal region and left basal ganglia), neurotransmitter mechanisms (decreased serotonin and norepinephrine), inflammatory cytokine-mediated (increased interleukin, IL-1β, IL-18, IL-6 and tumor necrosis factor, TNF-α), and gene polymorphism mechanisms (e.g., short variant of serotonin transporter gene-linked promotor region). A recent study also showed an association between elevated serum levels of neopterin both in the acute post-stroke phase and in those with PSD at 6-month follow up. Psychological factors, largely related to the various functional and personal losses associated with stroke, also contribute to the development of PSD. The correlation between location of stroke and the likelihood of developing PSD has been controversial; some studies have demonstrated positive correlations, and a large meta-analysis of 143 studies by Carson and co-workers found no correlation between lesion location and the risk of PSD.

PSD is associated with significant long-term negative effects on social function, motor abilities, and quality of life. Moreover, the negative effect of depression on functional impairment continues well beyond the period of abnormal mood symptoms. Such extended functional disability may be due to poor initial rehabilitation efforts by patients with PSD that limit the recovery of strength and mobility.

Diagnosis of PSD is straightforward in many cases, although certain situations can make diagnosis quite challenging. A number of non-depressive neurologic stroke sequelae may resemble symptoms of depression. Patients with expressive aprosodias have monotonous speech that may make them appear sad or withdrawn, and their affect may appear blunted. The presence of anosognosia (neurologically mediated unawareness of illness usually associated with right parietal lesions) may look like denial associated with depression, and this symptom can itself lead to frustration and anger when others insist that the patient has a problem that he or she simply cannot recognize. Finally, aphasias can make the diagnosis of depression—or any diagnosis—more difficult because of the difficulty of communicating with such patients. By being aware of these potential neurologic sequelae and by carefully using criteria from the Diagnostic and Statistical Manual of Mental Disorders , 5th edition (DSM-5), with particular attention paid to depressive symptoms that overlap less with concurrent medical and neurologic symptoms (e.g., feelings of guilt, worthlessness, hopelessness, suicidality), in most cases the psychiatrist can verify the presence or absence of PSD.

Despite the significant consequences of PSD (both because of under-diagnosis and the fear of intolerable side effects from antidepressant medications), it is often under-treated. However, early and effective treatment of depression is perhaps even more crucial in this patient population than it is in other populations, given the need for full mobilization for occupational and physical therapy and other functional re-training early in the course of recovery.

A number of placebo-controlled trials have demonstrated that antidepressants are effective in the treatment of PSD. SSRIs and nortriptyline have been shown to relieve symptoms of PSD; another study of nortriptyline found that treatment of depression resulted in improved cognitive outcome. A study by Robinson and associates found that nortriptyline was more effective than either fluoxetine or placebo in treating PSD and improving functional outcomes. Studies of PSD have found disruptions of both noradrenergic and serotonergic pathways; the effectiveness of venlafaxine was demonstrated in a case series by Kucukalic and colleagues, and only 2 of 30 patients studied had mild elevations in blood pressure, which would be a side effect of concern with venlafaxine in post-stroke patients. Finally, a 2008 systematic review of 12 pharmacotherapy trials found that antidepressant therapy is modestly beneficial for remission of PSD; however adverse events were more common with antidepressants. More studies are needed to fully assess the risks versus benefits in this regard.

Psychostimulants have also been used in the treatment of PSD. Retrospective studies using psychostimulants (methylphenidate and dextroamphetamine) to treat PSD found these medications to be effective, with response rates of 47–80%. Response to psychostimulants was rapid (usually within 48 hours), and adverse events were rare. However, unlike SSRIs and TCAs, psychostimulants have not been studied under placebo-controlled, double-blind conditions for the treatment of PSD.

ECT also appears to be an effective treatment for PSD, with high rates of response and low rates of medical complications. ECT is more extensively discussed in Chapter 37 . In addition to the somatic treatments of PSD, there is evidence to suggest that care management programs, which include education, antidepressant treatment guided by algorithm, and close monitoring of therapy may be more effective than somatic treatments alone. In addition to this, group and family psychotherapy have also been reported to safely and effectively treat PSD, but there are few randomized and controlled trials of individual psychotherapies for PSD. One study found that problem-solving therapy reduced the incidence of depression and also delayed the time to onset, compared with placebo, for post-stroke patients. Although there is some positive evidence for cognitive-behavioral therapy (CBT) as a treatment of PSD, other studies have found no significant difference between CBT, attention placebo, and standard care. More studies are needed in this area.

Administration of psychiatric medications to at-risk populations to prevent the onset of psychiatric illness is an increasingly popular area of study. Several recent studies have evaluated whether antidepressant medication can prevent PSD. Although initial studies demonstrated differences in rates of depression in post-stroke patients who received medication compared to placebo, they were not statistically significant and were under-powered. A more recent and methodologically sound study by Robinson and associates evaluated 176 non-depressed patients within 3 months of stroke and randomized them to three groups, a double-blinded escitalopram and placebo group and a non-blinded problem-solving therapy group. Rates of major and minor depression were statistically significantly lower for both escitalopram and problem-solving therapy, though escitalopram remained significant only with an intention-to-treat analysis. The potential clinical impact of these studies is impressive, but more studies are needed to definitely direct clinical care.

For most patients with mild to moderate PSD, SSRIs are the treatment of choice, given their proven efficacy, favorable side effect profile, and cardiovascular safety. However, TCAs, despite higher rates of side effects than SSRIs, might also be considered first-line agents for PSD because of their potentially superior efficacy. For more severe depression that impairs decision-making capacity, nutritional intake, or ability to participate in rehabilitation, psychostimulants should be strongly considered; methylphenidate or dextroamphetamine can be started at 2.5 to 5 mg in the morning, and a protocol for dosing and patient monitoring can be followed ( Table 20-2 ). ECT can also be considered in patients with incapacitating depression. Prophylactic treatment with antidepressants to prevent depression is supported by preliminary studies and may be prudent in patients with numerous risk factors; careful analysis of risks and benefits for the individual patient is still required.

Post-Stroke Apathy

Apathy is a disorder of diminished motivation and initiative that is characterized by restricted social engagement, lack of emotional response, and diminished cognitive abilities. In the post-stroke period, apathy was historically seen as a symptom of other syndromes, such as depression or dementia, but emerging evidence suggests that it might be a distinct entity. Post-stroke apathy (PSA) is as frequent as PSD and is also associated with poor functional recovery and low QoL. PSA has been associated with right hemispheric sub-cortical lesions with a particular focus in the basal ganglia and anterior cingulate circuit which is involved in motivation. Although more studies on treatment of PSA are needed, early work suggests the same treatment as PSD, including SSRIs and particularly psychostimulants.

Other Post-Stroke Psychiatric Phenomena

Other psychiatric syndromes that occur in the post-stroke period include anxiety, mania, and psychosis. Post-stroke anxiety is common, and usually arises in concert with PSD. Approximately one-fourth of patients meet criteria for generalized anxiety disorder (GAD) (except for duration criteria) in the acute post-stroke period; at least three-fourths of these patients with post-stroke GAD symptoms have co-morbid depression. Post-stroke anxiety has a negative impact on the functional recovery of stroke victims and has been associated with impairment in activities of daily living (ADL) up to 3 years after the event. The functional impairment of PSD and post-stroke GAD appear to be additive, insofar as patients with both GAD and PSD have greater ADL impairment at follow-up than those with isolated PSD.

Post-stroke mania occurs in fewer than 1% of patients. Symptoms of post-stroke mania are similar to those of primary mania (with flight of ideas, pressured speech, a decreased need for sleep, grandiosity, and associated psychotic symptoms). Lesions in the right orbitofrontal cortex, right basal temporal cortices, dorsomedial thalamus, and head of caudate appear to be associated most often with post-stroke mania. Stroke in the right hemisphere, compared with the left, has led to increased serotonin binding, and it is hypothesized that this may result in post-stroke mania.

Post-stroke psychosis is also uncommon, occurring at a rate of approximately 1–2%. Such patients usually have right temporoparietal lesions and a high rate of associated seizures. This suggests that temporal lobe damage that leads to complex partial seizures (CPS) and to associated psychosis may account for symptoms in a significant percentage of these patients.

Finally, there are two other clinical neuropsychiatric syndromes that are common in the post-stroke period. The first is termed catastrophic reaction , a collection of symptoms involving patient desperation and frustration. This syndrome is relatively common—especially in the acute phase; rates of post-stroke catastrophic reaction are 3–20%. Catastrophic reactions are strongly associated with PSD, with roughly three-quarters of patients with catastrophic reaction having PSD. Catastrophic reactions are also associated with a personal and family history of psychiatric disorders.

Catastrophic reactions also appear to be associated with anterior sub-cortical lesions and with left cortical lesions. Given the strong association of catastrophic reactions and depression, some feel that such a reaction is a behavioral symptom of depression (provoked by anterior sub-cortical damage) rather than a discrete syndrome. Others feel that catastrophic reactions result from damage to left hemispheric areas involved in the regulation of emotions related to social communication.

The second of these clinical syndromes is pseudobulbar affect , also termed pseudobulbar palsy . This syndrome (which consists of frequent and easily provoked spells of laughing or crying) occurs to some degree in approximately 15% of post-stroke patients. The pathophysiology is unknown but is thought to involve frontal release of brainstem emotional centers. It is usually seen in a mild form, with brief fits of crying or laughing linked with appropriate changes in mood; however, in more serious cases, it may involve frequent and spontaneous fits of laughing and crying inappropriate to the context. It can cause embarrassment, curtailment of social activities, and a decreased quality of life.

Treatment of these post-stroke phenomena generally parallels the treatment of primary psychiatric syndromes. Post-stroke anxiety can be treated like any primary anxiety syndrome. SSRIs are effective in the treatment of a variety of anxiety disorders, including GAD, and, given that most patients with post-stroke anxiety have co-morbid PSD, these agents are often the treatments of choice. Benzodiazepines can also be given for isolated anxiety, but they can lead to ataxia, sedation, and paradoxical disinhibition; therefore, they must be used with caution in this population. Furthermore, these agents do not treat co-morbid depression.

Treatment of post-stroke mania follows the same rules as does the treatment of primary mania. Mood stabilizers and adjunctive antipsychotic medications or benzodiazepines are used to control symptoms. Treatment studies of post-stroke mania have found lithium, valproic acid, carbamazepine, neuroleptics, and clonidine variably efficacious, though none of these treatments has been examined in placebo-controlled, double-blind trials for this condition. Post-stroke psychosis can be treated symptomatically with antipsychotics. However, anticonvulsants (especially valproic acid and carbamazepine) should be used when psychotic symptoms are the result of CPS, because psychotic symptoms should improve with better seizure control.

Finally, pseudobulbar affect and catastrophic reaction may respond well to antidepressants. A few trials of TCAs and SSRIs, some of which were placebo-controlled, have demonstrated efficacy in reducing symptoms of pseudo-bulbar affect. Additionally, the combination of dextromethorphan and quinidine has been shown to reduce laughing and crying spells due to pseudobulbar affect in patients with amyotrophic lateral sclerosis (ALS) and may also be of benefit in the post-stroke setting. Symptoms of catastrophic reaction may also improve with antidepressant treatment of co-morbid PSD.

In short, psychiatric symptoms after stroke are common and have a significant impact on the long-term outcome of post-stroke patients. Awareness of neurologic symptoms that may mimic psychiatric illness (e.g., anosognosia) and careful diagnostic interviews can allow accurate diagnosis and prompt treatment. In general, psychiatric symptoms secondary to stroke are treated in the same way as are non-stroke-related psychiatric syndromes with similar symptoms.