The Psychiatric Management of Patients With Cardiac Disease

Overview

Caring for cardiac patients can present a host of dilemmas for the general

hospital psychiatrist. Patients with psychiatric conditions may exhibit cardiac

symptoms, psychotropic agents can result in electrocardiographic

abnormalities, and psychiatric manifestations may result from cardiac

conditions. Because the overlap between psychiatry and cardiology is so great,

knowledge of ways to manage specific problems can be of tremendous benefit.

For instance, knowing how to deal with chest pain in the face of a psychiatric

syndrome, an electrocardiographic complication from a psychotropic agent, or

delirium due to cerebral hypoperfusion, can facilitate comprehensive and

compassionate care.

This chapter focuses on three main psychiatric syndromes related to the

cardiac patient: anxiety, depression, and delirium. For each of these syndromes,

we will consider epidemiology, clinical manifestations, differential diagnosis,

psychopharmacologic approaches, and practical management strategies for

patients with cardiac disease in the general hospital. Additional information on

the interface between psychiatric and cardiac care will also be provided in other

chapters.

826

Anxiety in the Cardiac Patient

The assessment of anxiety in the cardiac patient in the general hospital is often

complex. First, it may be difficult to ascertain whether the patient is

experiencing distress as a result of a myocardial event, an acute confusional

state, a primary anxiety disorder, or a complex interaction among these factors.

Furthermore, there are many potential causes of anxiety for the cardiac patient,

from an adjustment reaction to a serious cardiac event to the anxiogenic effects

of cardiac medications administered to treat such events. Among inpatients, the

threshold for treatment of anxiety tends to be lower than it is in the outpatient

setting, insofar as the elevations in catecholamine levels and vital signs

associated with mild to moderate anxiety may have profound cardiovascular

effects in the patient who has recently experienced an acute coronary syndrome

(ACS; myocardial infarction [MI] or unstable angina), coronary artery bypass

grafting (CABG), or heart failure (HF).

Epidemiology

Anxiety Among Cardiac Patients

Anxiety is commonly experienced by patients with cardiovascular disease, such

as coronary artery disease (CAD) or HF. Following an ACS, 20% to 30% of

patients experience elevated levels of anxiety,1 and 10% to 14% have anxiety

levels higher than in the average psychiatric inpatient.2,3 While this anxiety

gradually improves in the months following ACS, 50% of those patients with

elevated anxiety following ACS continue to have elevated anxiety 1 year postevent,

1 suggesting that a significant portion of patients with stable CAD may

actually suffer from an anxiety disorder that warrants identification and

treatment. Similarly, clinically significant anxiety is present in up to 25% of

patients awaiting CABG, though in most cases this anxiety resolves in the three

months post-procedure.4 Anxiety is also highly prevalent in patients with more

chronic cardiac diseases, such as HF. In patients with HF, 28% experience

clinically significant anxiety, and 13% meet criteria for an anxiety disorder.5

Cardiac patients who are subjected to invasive technology, such as

implantable cardioverter defibrillators (ICDs) or left ventricular assist devices

(LVADs), may also experience anxiety, panic, and fear, oftentimes associated

with these devices. A recent systematic review suggests that clinically

significant anxiety is present in 27% to 63% of patients pre-implantation and 8%

to 59% of patients post-implantation of an ICD.6 Among patients who have

undergone ICD placement, having received a shock from the ICD appears to

increase the risk of anxiety,6–8 though studies have not universally found this to

be true.6 The prevalence of clinically significant anxiety among patients

receiving treatment with LVADs is somewhat lower, with 18% to 23% of

patients reporting anxiety symptoms post-LVAD implantation.9,10 This anxiety

appears to improve as time passes following implantation.9,11

827

Anxiety Disorders in Cardiac Patients

In addition to high levels of free-floating anxiety, cardiac patients also

experience elevated rates of formal anxiety disorders. Similar to the general

population, generalized anxiety disorder (GAD) is commonly encountered in

patients with cardiovascular disease. Among patients hospitalized for an acute

coronary syndrome (ACS, arrhythmia, or HF), GAD was equally prevalent with

clinical depression.12 GAD also affects patients with stable CAD, with

prevalence rates ranging from 5% to 24% in this population.13–15

Patients with cardiovascular disease also frequently experience panic

disorder (PD), with some studies suggesting that patients with CAD have PD at

approximately four times the rate of the general population.16 Furthermore,

approximately 20% of all patients who arrive at Emergency Departments with

chest pain meet criteria for PD,16 and up to 50% of patients who visit outpatient

cardiology clinics for evaluation of their chest pain experience panic attacks or

meet criteria for PD.17 While some patients with PD and chest pain may not

have underlying structural cardiac disease, some certainly do, and clinicians

must remain open to the possibility of co-morbid cardiac illness in this patient

population.18

Finally, cardiac patients who experience events as traumatic during their

hospitalization may exhibit symptoms of post-traumatic stress disorder (PTSD).

Recent studies have found that 8% to 16% of patients who have an MI develop

symptoms of PTSD;19–21 such PTSD symptoms also arise at a similar rate among

patients who undergo CABG.21,22 Studies of patients receiving intensive care for

burn injuries and acute respiratory distress syndrome suggest that PTSD may

be even more prevalent among cardiac patients in intensive care units

(ICUs).23,24 Finally, patients who have undergone placement of AICDs also

appear to be at higher risk for PTSD. In this population, 10% to 25% have

elevated PTSD symptoms,25,26 and approximately 8% likely meet criteria for

PTSD.27 In one study, having more than five shocks from an ICD predicted

higher post-traumatic stress symptoms at follow-up.25

Association Between Anxiety and Cardiac Illness

Anxiety and anxiety disorders may be associated with an increased risk for

cardiovascular disease and poor cardiac outcomes, though the evidence for

these relationships is not as strong as that for depression. Epidemiologic studies

suggest that cardiac illness may lead to increased anxiety and that anxiety may

also exacerbate cardiac illness. Acute and chronic emotional stress have been

linked to the development of ventricular arrhythmias and to the exacerbation of

silent myocardial ischemia.28

Among patients without pre-existing heart disease, anxiety has been

associated with the development of CAD. In a recent meta-analysis involving

249,846 healthy individuals, anxious persons were at significantly elevated risk

for the development of CAD and for cardiac-related mortality over the next 11

years, independent of health behaviors and sociodemographic and medical

covariates.29 This suggests that among healthy individuals, significantly

828

elevated anxiety may be associated with physiologic changes in the body that

predispose to a higher risk for the development and progression of CAD.

Among patients with CAD, the association between anxiety and

cardiovascular outcomes is less clear. In a recent meta-analysis of 44 studies and

30,527 individuals, anxiety was associated with an increased risk of poor

cardiovascular outcomes in unadjusted analyses; however, when controlling for

sociodemographic, medical, and psychological covariates (often depression),

many of these relationships became non-significant.30 This would suggest that

much of the relationship between anxiety and outcomes may be explained by

other medical and psychiatric variables, such as depression. When examining

specific subgroups of patients, anxiety was significantly associated with poor

outcomes in patients with stable CAD but not in patients who recently had

experienced an ACS.30 It may be that a certain amount of anxiety is to be

expected following ACS, and if this anxiety resolves quickly post-event, it may

not have a significant impact on future cardiac health. In contrast, anxiety

experienced in the setting of stable CAD may persist for longer periods of time

and therefore may have more clinically significant effects on heart health. The

links between free-floating anxiety and outcomes is equally unclear in patients

with HF. Four recent prospective, observational studies failed to find a

significant relationship between anxiety (as a symptom) and mortality, when

controlling for relevant medical and psychological covariates.31–34 However,

similar to the studies in patients with CAD, some of these studies did find

trends towards a significant relationship between anxiety and mortality in lesscontrolled

analyses.31,34 This suggests that while anxiety may be a useful marker

for poor outcomes in patients with HF, its relationship with mortality may be

accounted for by other psychiatric, medical, and sociodemographic factors.

While the relationship between anxiety and cardiac outcomes is not entirely

clear, there is evidence that specific anxiety disorders are associated with

adverse cardiac outcomes.35 When anxiety reaches the threshold of a disorder, it

is by definition more persistent, pervasive, and limiting, and carries a more

significant risk in terms of cardiac outcomes. GAD has also been associated

with higher rates of smoking, diabetes, and hypercholesterolemia, which may

increase the risk of developing cardiovascular diseases.36 Following MI, GAD

has been associated with higher rates of mortality and cardiac re-admissions.37

Similarly, in patients with stable CAD, GAD diagnosis predicted major cardiac

events in the subsequent 2 years.13

PD also has been associated with the development and progression of

cardiovascular disease. In a study of over 5,000 post-menopausal women,

patients with a history of panic attacks in the past 6 months were at higher risk

for subsequent MI, cardiac mortality, and all-cause mortality compared to

individuals without a history of panic attacks.38 Similarly, in a systematic

review and meta-regression involving over 1 million patients, PD was

associated with incident CAD, MI, and major adverse cardiac events.39 These

findings are supplemented by other studies that similarly have found PD to be

associated with incident CAD,40,41 though one study found that PD diagnosis

was associated with a reduced risk of cardiovascular mortality overall.41

829

While there is less research available related to the relationship between

PTSD and cardiac health, preliminary evidence suggests that PTSD may be

harmful for cardiac health. PTSD has been associated with an increased

incidence of CAD, independent of depression and other relevant factors.42 In

patients who have experienced an acute coronary syndrome, PTSD has been

linked to a greater risk of major adverse cardiac events and all-cause mortality.43

In sum, cardiac patients have high rates of situational anxiety and formal

anxiety disorders (e.g., PD, GAD, and PTSD). While the links between anxiety

(as a symptom) and cardiovascular outcomes is still unclear, anxiety disorders

have been associated with an increased risk of developing cardiac disease, as

well as worse cardiac outcomes and increased rates of mortality in patients with

established disease. This highlights the importance of accurately identifying

and treating these disorders when present in patients at risk for or with existing

cardiovascular disease.

Differential Diagnosis of Anxiety in the Cardiac Patient

Anxiety in the general hospital is often a primary psychiatric problem caused

by stressful medical events. However, anxiety in the cardiac patient can also be

caused by a number of general medical conditions and medications commonly

associated with cardiac care (Table 26-1).

Table 26-1

Selected General Medical Causes of Anxiety

Among Cardiac Patients in the General

Hospital

Cardiac Events

Myocardial ischemia

Atrial and ventricular arrhythmias

Congestive heart failure

Other Medical Conditions

Pulmonary embolism

Asthma/chronic obstructive pulmonary disease (COPD) exacerbation

Hyperthyroidism

Hypoglycemia

Medications

830

Sympathomimetics

Thyroid hormone

Bronchodilators

Stimulants

Corticosteroids

Illicit Substances

Cocaine or amphetamine intoxication

Alcohol or benzodiazepine withdrawal

Not uncommonly, cardiac events cause anxiety. Myocardial ischemia,

arrhythmias, and HF can each cause anxiety owing to the sympathetic

discharge associated with these conditions and because of what they may

represent to the patient (e.g., the fear of dying, the worsening of medical illness,

the loss of role identity). Other general medical conditions may cause or

exacerbate anxiety in the cardiac patient; important among these is pulmonary

embolism in the sedentary cardiac patient. Anxiety may also be a side effect of

medications administered to cardiac patients, such as sympathomimetics.

Anxiety can also result from substance intoxication or withdrawal that may be

causing or exacerbating acute cardiac issues (e.g., cocaine intoxication, alcohol

withdrawal). Finally, impaired sleep in the hospital (as the result of an

unfamiliar setting, frequent nursing interventions, and significant noise) can

lead to or exacerbate anxiety.

The general hospital psychiatrist should consider general medical causes of

anxiety when evaluating cardiac patients; this is especially true when the

anxiety has developed during an uneventful hospitalization, when the patient

has no history of anxiety, or when anxiety persists despite appropriate

treatment.

Psychopharmacologic Issues in the Anxious Cardiac Patient

Agents used to treat anxiety in the general hospital patient include

benzodiazepines, antidepressants, and antipsychotics. Benzodiazepines are the

medications most frequently used in the treatment of anxiety in cardiac

patients. These medications rapidly relieve anxiety and appear to have a

number of beneficial cardiovascular effects.

Benzodiazepines.

Among patients with myocardial ischemia or infarction, benzodiazepines

reduce catecholamine levels and decrease coronary vascular resistance.44

Although β-blockers have similar effects, anxious patients tend to have

831

elevations in vital signs, catecholamines, and coronary pressures as the result of

their anxiety, despite the use of β-blockers; benzodiazepines can effectively

treat these abnormalities. In addition, there is some evidence that

benzodiazepines may inhibit platelet aggregation and raise the ventricular

fibrillation (VF) threshold.45 Furthermore, benzodiazepines are generally well

tolerated by the general hospital population; low rates of hypotension, virtually

no anticholinergic effects, and very low rates of respiratory compromise

develop when standard doses of benzodiazepines are used. Benzodiazepines

also appear to be safe even in seriously ill patients, with low rates of adverse

events. Although clinicians may be concerned about the development of

benzodiazepine dependence, when these agents are used in the acute care

setting, at adequate doses and for appropriate indications, the risk of

dependence is minimal. Benzodiazepines may even have beneficial effects on

cardiovascular outcomes in specific populations, such as those with cocaineinduced

chest pain.

One important caveat for the use of benzodiazepines is that they can

exacerbate confusion and paradoxically worsen agitation in patients with

delirium or dementia; therefore, other agents (e.g., antipsychotics) may be more

appropriate for the treatment of anxiety, fear, and distress in the delirious or

demented cardiac patient.

Antidepressants.

Antidepressants can also be used in the treatment of anxiety in the general

hospital. However, these agents often take several weeks to work and are best

used to treat primary anxiety disorders, such as PD, GAD, or PTSD. For acutely

anxious cardiac patients in the general hospital, when antidepressants are

prescribed, it is often wise to co-administer a benzodiazepine to acutely reduce

anxiety during a vulnerable cardiovascular state. Antidepressants will be

discussed more extensively in the section on depression.

Antipsychotics.

Antipsychotics can also be used for the treatment of heightened anxiety in the

general hospital. Though no agents have specific approvals for anxiety

disorders, the use of antipsychotics as adjunct treatment for anxiety in nonmedical

populations is now standard clinical practice. These agents have the

additional beneficial effects of symptomatically treating co-morbid delirium,

and they do not cause the paradoxical disinhibition that is sometimes associated

with benzodiazepines. Antipsychotics, however, can cause orthostasis and

anticholinergic effects (associated with low-potency typical agents and, to a

lesser degree, some atypical agents) and may be associated with prolongation of

the corrected QT (QTc) interval. Many atypical agents also carry a risk of

weight gain, which may further predispose patients to adverse cardiac

outcomes. Antipsychotics will be discussed more extensively in the section on

delirium.

Other Agents.

832

The anticonvulsant gabapentin has been used in the acute treatment of anxiety.

Gabapentin is associated with essentially no risk of physiologic dependence,

and does not cause orthostasis or anticholinergic effects. Its efficacy in the

treatment of acute anxiety in hospitalized cardiac patients has not been formally

studied. Gabapentin is also used at times to treat post-operative pain and

alcohol withdrawal.

Approach to the Anxious Cardiac Patient

The psychiatric consultant is frequently called to cardiac floors to assess and

treat anxiety. A careful, stepwise approach to these consultations can ensure an

accurate diagnosis and appropriate treatment.

Consider a Broad Differential Diagnosis for the Patient's Distress.

A primary role of the general hospital psychiatrist is to accurately characterize a

patient's distress as anxiety, denial, depression, delirium, or another psychiatric

phenomenon. Patients who appear anxious and tremulous may in fact be

disoriented, paranoid, and frightened—that is, delirious. Therefore, the

consultant should be careful in the interview to assess affect, behavior, and

cognition.

If the patient's primary psychiatric symptom appears to be anxiety, the

consultant should then consider the potential contribution of medications or

medical symptoms to this anxiety. As noted earlier, there is a long list of

conditions that can cause or exacerbate anxiety, and the consultant should

carefully consider these and recommend appropriate diagnostic studies, if

appropriate. It may be especially useful to note correlations between anxiety

levels and the initiation or discontinuation of potentially offending medications

or substances.

Evaluate Sources of Anxiety and Assess How the Patient Has Dealt With

Difficult Situations in the Past.

A careful psychiatric interview of the anxious patient will help determine what

factors are causing his or her anxiety. Is the preoperative CABG patient anxious

because a relative died during cardiac surgery years ago? Is the patient with an

AICD fearful that his defibrillator will painfully discharge again? By

determining the sources of anxiety, the consultant will be able to address these

anxieties through education, reassurance, medication, or brief psychotherapy.

A related task is to determine the patient's coping style and coping strengths.

How does he or she manage anxiety outside of the hospital? How has he or she

managed difficult situations in the past? The consultant can use this

information to identify the patient's strengths and determine the best approach

to the patient's anxieties.

Another question for patients involves stimulation and control. Some cardiac

patients crave control and wish to know every detail of their care; they feel

anxious when they do not feel that they have comprehensive information about

their illness and when they are not part of all treatment decisions. In contrast,

833

other patients find such information and the pressure to make decisions overstimulating

and feel less anxious when told only the general details of their

condition.

Recommend Appropriate Behavioral and Therapeutic Interventions.

Having learned about the patient's sources of anxiety, coping strengths, and

preferences regarding control, the consultant is in an excellent position to

design a treatment plan that reduces a patient's anxiety. For example, if a

patient reports that the hospitalization is overwhelming, members of the

treatment team can be encouraged to limit detailed information and reassure

the patient that they see this condition frequently (if true) and that they plan to

provide excellent care to the patient. On the other hand, for the patient whose

anxiety increases with the perceived lack of control, the treatment team can be

encouraged to provide the patient with detailed information and written

materials. The patient should also be included in treatment decisions; inclusion

in even small decisions (e.g., the best time for dressing changes) can allay

anxiety and allow the patient to feel in control.

In other cases, worried cardiac patients simply need to express their anxieties

to someone. If the consultant, treating physician, and nursing staff can set aside

short periods to reflectively listen to the patient's fears, this investment of time

often results in significantly less anxiety, greater compliance with treatment,

and less chaos for patient and staff alike. If the patient seems to have an

insatiable desire to discuss his or her fears, staff can be taught to consistently set

aside time to listen to the patient while setting limits on his or her time; for

example, a nurse may tell the patient that she will sit with him or her for 5

minutes at the beginning, middle, and end of the shift to talk about his or her

worries. If the patient attempts to engage the nurse in further conversation

about this topic, the nurse can calmly tell the patient that they can discuss it at

their next appointment.

Intelligently Use Psychiatric Medications for Specific Target Symptoms.

Benzodiazepines are often the agents of choice for the anxious cardiac patient. If

the anxiety appears to be short-term or situation-specific (e.g., whenever a

procedure is performed), a short-acting benzodiazepine can be used on an asneeded

basis, (e.g., lorazepam 0.5 to 1.0 mg as needed for acute anxiety).

However, for most patients, longer-acting benzodiazepines given on a standing

basis provide the smoothest and most consistent reduction of anxiety. Most

anxious cardiac patients can be started on clonazepam 0.5 mg at night or twice

per day; doses can be adjusted upward if this dose is well tolerated and anxiety

persists. In general, these agents can be discontinued on discharge from the

hospital if they were only used on a short-term basis.

Benzodiazepines may not be the agents of choice for patients with acute or

chronic organic brain syndromes (e.g., delirium, dementia, traumatic brain

injury), tenuous respiratory function (including obstructive sleep apnea), or a

history of substance dependence. For these patients antipsychotics or

gabapentin are often useful, alleviating agitation and confusion in delirious

834

patients while also reducing anxiety. We often start with doses of quetiapine at

12.5 to 25 mg at night or gabapentin at 100 mg three times daily.

Antidepressants may be useful in the treatment of primary anxiety disorders

and when depression is co-morbid with anxiety; we sometimes co-administer a

benzodiazepine to reduce initial anxiety associated with initiation of

antidepressants.

Return Frequently to See the Patient.

Anxious patients generally are relieved to see a familiar face, especially one that

has attempted to understand and address their anxiety. Such frequent followup,

therapeutic in itself, allows for careful monitoring of behavioral and

pharmacologic interventions.

Case 1

Mr. A, a 53-year-old executive without a psychiatric history, was admitted for

CABG after his cardiac catheterization revealed three-vessel cardiac disease.

Initially, he had an uneventful perioperative course. However, on the day after

his operation, psychiatry was consulted to assess his capacity to leave the

hospital against medical advice.

On interview, Mr. A was alert, oriented, lucid, and initially quite angry. He

reported, “I have no assurance that I'm getting the right care; the doctors and

nurses come in and out of my room and bark orders to one another but they

don't include me at all. They haven't even listened to the fact that I always take

my sleeping pill at 9 p.m. every night instead of 11 p.m. like they give it to me.

I'm fed up.” By the end of his tirade, Mr. A's anger had changed to fear and

anxiety.

The consultant told Mr. A that he would bring his concerns to the team. The

consultant met with members of the treatment team and encouraged them to

provide as much information as possible about his care and to allow him to

mandate his treatment when possible (e.g., getting his sleeping pill at 9 p.m.).

The consultant, nurse, and Mr. A then met together so that Mr. A could express

his concerns and the nurse and consultant could outline the ways their

procedures would change so that he could have more information and more

control. Mr. A agreed to this plan and also agreed to clonazepam 0.5 mg twice

per day to reduce his anxiety.

The consultant checked back frequently with Mr. A to assess his response to

this treatment plan. Small changes in the plan were instituted at his request,

and his anxiety steadily decreased. He was discharged to cardiac rehabilitation

and thanked the nursing staff for their “compassionate care.”

835

Depression in the Cardiac Patient

Over the past two decades, substantive research has firmly established a

bidirectional link between depression and cardiac disease: Patients with

depression are more likely to develop cardiac disease, and patients with cardiac

disease are more likely to suffer from depression.46 Multiple studies have also

demonstrated that patients with depression and cardiac disease have worse

outcomes than those without depression. These findings have underscored the

importance of the general hospital psychiatrist's role in identifying depressed

cardiac patients and considering appropriate treatments.

Depression in Patients With Established Cardiac

Illness

Depression is common among patients with CAD, with prevalence rates of

major depressive disorder (MDD) hovering around 20%.47 This rate of MDD is

greater than the prevalence of depression in the general population (approx.

15%), and is even higher for patients with more serious cardiac disease.

Roughly 15% to 30% of post-ACS patients; 20% to 35% of patients with HF; and

24% to 33% of patients with an AICD meet criteria for MDD.48–51 Rates of

suicidal ideation are also increased among patients with cardiac disease.52

Furthermore, studies indicate that most patients who are found to have major

depression during a cardiac hospitalization have a history of MDD predating

their cardiac event, and depressive symptoms often persist following

discharge,53,54 with more than half of patients with post-ACS depression

remaining depressed after 1 year.

Despite the high prevalence of MDD and the risks it carries for cardiac

patients, less than 15% of post-MI depressed patients are recognized as such.55

Several factors likely account for these low rates of recognition. The pattern of

depression (with hostility, listlessness, and withdrawal being more common

than sad mood) is often somewhat atypical;56 furthermore, depression is often

seen as a normal consequence of a serious medical event, such as an ACS.

Finally, most patients with ACS have brief inpatient stays, and it may be

difficult to assess a patient's mood or to obtain psychiatric consultation during

this limited time frame.

A review of post-MI depression delineated a number of putative risk factors

for the development of post-MI depression.50 These risk factors included

smoking, hypertension, female gender, social isolation, medical complications

during acute hospitalization, a history of depression, and first-time prescription

of benzodiazepines (suggesting potential co-morbidity between anxiety and

depression).

In short, depression is common among cardiac patients, and it has been best

studied in those with recent ACS. Post-ACS depression is highly prevalent,

poorly recognized, and frequently persistent.

836

Depression as a Risk Factor for Cardiac Disease

Evidence from many community studies over the past 20 years indicates that

depression is an independent risk factor for cardiovascular disease. Several

studies have found that patients with depressive symptoms are 1.5 to 3.5 times

more likely to have an ACS than are those without such symptoms, while those

with major depression have an even greater risk.57–59 This increased

vulnerability holds true for female patients as well as those over the age of 65,

and has been demonstrated in studies with a follow-up period of more than 25

years.48,60,61 In a meta-analysis, van der Kooy and colleagues examined 28

relevant studies comprising more than 80,000 subjects, and demonstrated that

depression was associated with an increased risk of cardiovascular disease (RR

1.46).62 This analysis found that the strongest association was made when

patients were diagnosed by clinical interview and that studies using depression

scales demonstrated a dose–response relationship between depressed mood

and the development of cardiovascular disease.

In addition to predicting the development of CAD in healthy people,

depression has also been associated with significantly higher rates of cardiac

death and overall mortality among patients with established CAD. Numerous

studies have found that depressive symptoms after ACS are associated with

increased morbidity and mortality in the subsequent 5 years.63–67 These effects

on mortality appear to be largely independent of the severity of cardiac disease,

demographic variables, medications, or other confounding factors. One early

study found that depressive symptoms immediately after MI were associated

with a four-fold to six-fold increase in risk of cardiac mortality in the next 6 to

18 months.64,65 Bush and colleagues found that even minimal depressive

symptoms were associated with an elevated risk of cardiac mortality, though

more severe depressive symptoms more strongly predicted cardiac mortality.66

The Stockholm Female Coronary Risk Study found that women with two or

more depressive symptoms had a two-fold increased risk of future cardiac

events (ACS, cardiovascular mortality, and re-vascularization procedures) over

5 years compared with women who had one or no depressive symptoms.68

Similarly, a meta-analysis of 22 studies of post-MI subjects found that post-MI

depression was associated with a 2- to 2.5-fold increased risk of negative

cardiovascular outcomes.69 Patients experiencing a first episode of depression in

the aftermath of an ACS may be at the highest risk for negative cardiac

outcomes.70,71 In light of this overwhelming evidence, the American Heart

Association in 2014 declared that depression was a risk factor for adverse

events following an ACS.72

Depressive symptoms also appear to predict cardiac morbidity and mortality

in other cardiac populations. In depressed patients hospitalized for cardiac

illness, each additional point on the Patient Health Questionnaire-9 depression

rating scale was independently associated with a 9% greater risk of cardiac readmission

over the next 6 months.73 At least three studies have found that pre-

CABG depressive symptoms have been associated with an increase in cardiac

morbidity at 6- or 12-month follow-up.74–76 Among patients undergoing cardiac

837

transplantation, persistent depression was associated with increased rates of

incident CAD and mortality.77,78 Depression is a risk factor for incident HF, and

several studies have shown that depression predicts higher mortality

independent of demographic factors or clinical status.48,79,80 Patients with

depression at time of AICD placement have higher rates of all-cause mortality

over the subsequent 4 years, and depressed patients with atrial fibrillation have

higher rates of recurrence following cardioversion.81,82

Numerous mechanisms have been offered to explain the link between

depression and cardiovascular disease. Behavioral hypotheses include

continued poor health habits (e.g., smoking, lack of exercise) and noncompliance

with medical care. Depressed patients appear to have decreased

adherence to medication regimens, and depressed cardiac patients attend

cardiac rehabilitation programs less frequently than their non-depressed

peers.83,84 Furthermore, patients who are depressed after ACS are less likely to

follow recommendations about diet, exercise, and smoking cessation.85

Physiologic mechanisms that have been proposed in the link between

depression and worsened outcomes in cardiac disease include inflammation,

endothelial dysfunction, platelet activation and aggregation, and autonomic

nervous system dysfunction. Two studies have suggested at least a minor

contribution from inflammatory factors, including increased levels of C-reactive

protein and interleukin-6 in both depression and cardiac disease.86,87 Depression

has been associated with impaired endothelial dysfunction in those with and at

risk for heart disease.59,88 Depressed patients with CAD appear to have

increased platelet aggregation,89 and this could increase vulnerability to

myocardial ischemia.90 In terms of autonomic system dysfunction, depressed

patients, as well as those with CAD and HF, have been shown to have

decreased heart rate variability (HRV). Further, depressed patients with CAD

have greater decreases in HRV than patients with depression or CAD alone,

and more severe depression correlates with a greater reduction in HRV.91,92

Depressed patients also have increases in baseline levels of circulating

catecholamines and exaggerated elevations in catecholamine levels during

stress;93 elevated catecholamine levels can result in increased myocardial

oxygen demand and elevations in blood pressure and heart rate. Furthermore,

elevated catecholamine levels have been associated with infarct initiation,

infarct extension, and the development of VF in patients with MI. Perhaps the

most obvious link between depression, increased sympathetic drive, and

cardiac illness is seen in the phenomenon of Takotsubo cardiomyopathy, in

which negative psychological factors, such as depression, grief, or acute loss can

directly lead to a reversible cardiomyopathy that mimics an acute MI and can

be fatal.94

The link between depression in cardiac patients and increased cardiac

morbidity is likely mediated by multiple factors. One crucial question that

remains unanswered is whether treatment of depression among cardiac

patients improves cardiac prognosis. The vast majority of studies have not

demonstrated a link between antidepressant therapy and improved cardiac

outcomes, though the Enhancing Recovery in Coronary Heart Disease

838

(ENRICHD) trial found a 40% lower risk of either death or non-fatal MI in

patients treated with antidepressant medications.95

Differential Diagnosis of Depression in the Cardiac

Patient

As with anxiety in the cardiac patient, there are a number of medical conditions

and medications that can cause or exacerbate depressive symptoms. Table 26-2

lists a number of these medical influences on mood. Conditions associated with

depressed mood that are common in the cardiac patient include

hypothyroidism (both idiopathic and secondary to amiodarone, which is

frequently prescribed to cardiac patients), Cushing's syndrome (Cushing's

disease or symptoms secondary to steroid administration), neoplasm (especially

pancreatic), vitamin B12 and folate deficiencies, and depression associated with

vascular dementia.

Table 26-2

Selected General Medical Causes of

Depression Among Cardiac Patients in the

General Hospital

Medical Conditions

Hypothyroidism (idiopathic or amiodarone-induced)

Cushing's syndrome

Vitamin B12 or folate deficiency

Neoplasm (especially pancreatic, lung, or central nervous system tumors)

Vascular dementia

Movement disorders (e.g., Parkinson's disease or Huntington's disease)

Medications

Methyldopa

Reserpine

Corticosteroids

Interferon

839

Illicit Substances

Chronic alcohol or benzodiazepine abuse

Cocaine or amphetamine withdrawal

A number of medications sometimes used in cardiac populations have been

associated with depression. Steroids, methyldopa, and reserpine have each been

linked with increased rates of depression. Substances can also influence mood;

chronic alcohol use and withdrawal from cocaine or amphetamines commonly

lead to depression. β-Blockers have long been associated with depression;

however, a recent re-examination of the literature suggests a minimal

association between β-blockers and depression.96

Because depression can have significant effects on cardiac and psychosocial

outcome in cardiac patients and because the causes of depression are often

reversible or treatable, the general hospital psychiatrist should consider these in

all depressed cardiac patients.

Psychopharmacologic Issues in the Depressed

Cardiac Patient

Antidepressants are effective in the treatment of depression for patients with

cardiac disease. However, older antidepressants (i.e., tricyclic antidepressants

[TCAs] and monoamine oxidase inhibitors [MAOIs]) have effects that make

their use in cardiac patients difficult. TCAs have anticholinergic effects

(including tachycardia) and can cause orthostasis. Even more concerning is the

propensity of TCAs to prolong cardiac conduction and to have a proarrhythmic

effect in some patients as a result of their quinine-like properties.97,98

One study, controlling for medical and demographic factors, found that

depressed patients on TCAs had more than a two-fold risk of MI, whereas

selective serotonin re-uptake inhibitors (SSRIs) did not enhance risk of MI.99 For

these reasons, the use of TCAs is generally not recommended for patients with

pre-existing cardiac disease.

SSRIs are now considered the best first-line medications for treating

depression in patients with cardiac disease. None cause problematic orthostasis,

and, with the exception of paroxetine, they are generally not associated with

significant anticholinergic effects. The safety of SSRIs, particularly sertraline,

has been well-established in post-MI populations. The multi-center Sertraline

AntiDepressant Heart Attack Randomized Trial (SADHART) used a doubleblind,

placebo-controlled design to administer sertraline or placebo to 369 post-

MI patients.100 The investigators found that, compared with placebo, sertraline

significantly improved depressive symptoms and was not associated with

changes in ejection fraction, cardiac conduction, or adverse cardiac events. Two

important limitations of this study were that patients did not begin treatment

until an average of 34 days after their MI and that the patients were followed

for only 24 weeks. The Canadian Cardiac Randomized Evaluation of

840

Antidepressant and Psychotherapy Efficacy (CREATE) trial found citalopram

safe and effective in treating depression in 284 patients with CAD.101 More

recently, the SADHART-CHF study found sertraline to be safe for treating

patients with significant HF, though sertraline did not separate out from

placebo in terms of reducing depressive symptoms or improving cardiac

morbidity.102 Our clinical experience with SSRI administration in cardiac

patients has also found SSRIs to be safe, and we have safely prescribed SSRIs in

post-MI patients earlier than 1 month after the MI when indicated by the

severity of depression or the follow-up circumstances.

Though SSRIs had long been considered safe from the standpoint of cardiac

conduction, the Food and Drug Administration (FDA) in 2011 issued a warning

regarding the potential for citalopram to prolong the QTc interval and increase

the risk for lethal ventricular arrhythmias such as torsades de pointes (TDP).103

This warning was based on a thorough QT study ordered by the FDA, which

showed a dose-dependent increase in QTc with citalopram, with an absolute

increase of 18.5 msec at doses of 60 mg daily. Though a dose-dependent

increase was also shown for escitalopram, the magnitude was smaller (10.7

msec at 30 mg), and no warning was issued. In 2012, the warning was

downgraded to note that citalopram is not recommended at doses >40 mg in the

general population, is not recommended at doses >20 mg in patients over the

age of 65 or with pre-existing liver disease, is not recommended for patients

with congenital long-QT syndrome, and should be discontinued in patients

with QTc >500 ms. Several subsequent studies as well as a meta-analysis have

suggested that citalopram does indeed separate out from other SSRIs in its

propensity to prolong the QTc,104,105 though not all findings have supported a

dose-dependent relationship.106 From a clinical standpoint, though the

magnitude of increase is small and likely to be insignificant for most patients,

many no longer use citalopram as a first-line agent in those with a history of or

significant risk factors for heart disease, preferring sertraline instead, given its

established safety.

Newer antidepressants are less well-studied than SSRIs. Venlafaxine can

elevate blood pressure, which may preclude its use as a first-line agent in

patients with cardiac disease. Duloxetine has not been associated with QTcprolongation

or other cardiac side-effects, and may be a reasonable second-line

agent.103 Bupropion, at therapeutic doses, does not have adverse effects on

blood pressure, heart rate, or other cardiovascular parameters, and has been

shown to reduce rates of smoking.107 Furthermore, a study of bupropion in

depressed patients with CAD found that this agent had a favorable

cardiovascular side effect profile in this specific population.108 Mirtazapine has

few effects on cardiac conduction or vital signs, even in overdose.109 The

Myocardial Infarction Depression Intervention Trial (MIND-IT), a 24-week

randomized, placebo-controlled study, found mirtazapine to be safe in 209 post-

MI patients with depression, and mirtazapine often has more immediate effects

on sleep than do other antidepressants.110 However, mirtazapine is highly

associated with weight gain as a result of its interaction with histamine

receptors, limiting its use in patients with cardiac disease.

841

Psychostimulants have also been shown to be rapidly acting, efficacious

antidepressants in medically hospitalized patients.111,112 Though they may

elevate blood pressure or heart rate, stimulants may be indicated in cardiac

patients whose depression requires rapid treatment (e.g., depression that is

severe, is negatively affecting rehabilitation owing to anergia or minimal oral

intake, or is affecting the patient's capacity to make medical decisions).

Stimulants are relatively contraindicated in patients with a history of

ventricular tachycardia, recent MI, HF, uncontrolled hypertension or

tachycardia. However, in many cardiac patients, psychostimulants can be used

safely with slow dosage titration, beginning with 2.5 to 5.0 mg in the morning

and increasing up to 20 mg per day.

Other Treatment Modalities for the Depressed

Cardiac Patient

Several studies have also examined the efficacy of psychotherapy to treat

depression in cardiac patients. Though both cognitive-behavioral therapy (CBT)

and interpersonal therapy (IPT) have been shown to reduce depressive

symptoms in patients with cardiac disease, effects are typically short-lived, and

IPT in particular was shown to be inferior to clinical management in terms of

depression scores.95,101 One study suggested that CBT significantly improved

depressive symptoms at 3 and 9 months in patients who had undergone CABG

in the past year.113

Increasing evidence suggests that collaborative care, stepped care, and

blended care models are highly effective in treating depression in cardiac

populations. These programs use a non-physician care manager to identify and

monitor psychiatric conditions while transmitting care recommendations from

a study team psychiatrist to primary medical providers. Both pharmacologic

and psychotherapeutic approaches are often utilized. Collaborative care

programs have been successful for depressed patients undergoing CABG,114 for

patients with recent ACS,115,116 and even when started in the hospital during an

admission for acute cardiac illness.117,118 In addition to improving depression,

these programs have been shown to be cost-effective and possibly even costsaving,

119 and have resulted in reduced rates of cardiac readmissions and death

during the study period.120

Approach to the Management of the Depressed

Cardiac Patient

The approach to the depressed cardiac patient is in many ways similar to the

approach to the anxious cardiac patient. The consultant must first confirm that

depression is the primary psychiatric symptom and evaluate for co-morbid

psychiatric conditions. Furthermore, the psychiatrist must consider the

presence of medical conditions or medications that can cause or exacerbate

mood symptoms. Once these steps have been completed, an approach to

842

treatment involves an identification of the patient's coping strengths and

support network; it may also involve the weighing of the risks and benefits of

antidepressant treatment.

Routine Screening.

One of the biggest obstacles to diagnosing depression in patients with cardiac

disease is failure to adequately screen. Since 2008, the American Heart

Association has recommended routine screening for depression using the

Patient Health Questionnaire-2 (PHQ-2) and PHQ-9 in patients with cardiac

disease in all settings.121 Given the resource burden and lack of evidence that

screening alone improves outcomes, we recommend screening only in the

setting of adequate treatment options. It is important to remember that cardiac

patients are also at increased risk for suicidal ideation, and screening for

thoughts of suicide is a crucial component of depression screening.

Consider Appropriate Psychiatric and Medical Differential Diagnoses.

A positive screen for depression should be followed by consideration of

alternative explanations. As with consultations on cardiac floors for apparent

anxiety, we also find that consultations for apparent depression often reveal

that a patient's distress is caused by another psychiatric syndrome, such as

hypoactive delirium or somatic symptoms secondary to cardiac illness.

The consultant should also note the course of depressive symptoms to see if

the onset or worsening of such symptoms correlated with the administration of

a new medication or new physical symptoms or if there are other indications

that a physical disorder might be implicated in the evolution of the depressive

symptoms. The general hospital psychiatrist should also order laboratory tests

and other studies as indicated.

Attempt to Identify the Patient's Coping Style and the Triggers for

Depressive Symptoms.

Determining the external factors that exacerbate depressive symptoms may

help the consultant reduce the patient's stressors. The consultant can use this

information to implement solutions that are psychotherapeutic in nature (e.g.,

discussing mortality and life goals) or more concrete (e.g., having family

members call the patient to let him know he is missed and important to them).

Identification of the patient's coping strengths—especially, how the patient has

previously managed difficult situations—will inform the treatment team's

approach to the patient.

Of particular interest to the psychiatric consultant are recent data indicating

that patients with CAD who use a repressive coping style are at particular risk

of adverse cardiac events and death.122 Perhaps because these individuals report

low levels of anxiety and depression, they were once thought to be at low

psychological risk for clinical events. However, it is now thought that these

individuals often fail to detect or report significant emotional distress, which

could contribute to their increased risk of MI and death. The consultant should

be particularly mindful of patients whose emotional cool seems at variance with

843

the severity of their clinical circumstances.

Make Use of Existing Social Supports or Help Develop a Network.

Social support has been associated with superior medical outcomes in

depressed patients after MI;123 therefore, if such social support does not exist,

the consultant can work with the treatment team to consider options to improve

the patient's support system. Such options could include participation in

cardiac rehabilitation, having visiting nurses, or joining a support group.

Carefully Consider the Use of Antidepressant Medication.

SSRIs, particularly sertraline, appear to be both safe and effective for patients

with CAD, with few cardiovascular side effects. Citalopram may carry a

slightly higher risk of QTc-prolongation and should therefore be used more

thoughtfully in this population. Bupropion also has few drug–drug interactions

and may be the agent of choice in patients with co-morbid MDD and a desire to

stop smoking.

The risks and benefits of antidepressant medications should be more

carefully considered in patients with recent MI and probably by extension all

patients with severe cardiac disease. For most patients who have just had a MI

or a CABG, we typically do not prescribe antidepressants for the onset of

depressive symptoms within days after MI, both because such patients have not

yet met criteria for MDD and because extensive data establishing the safety of

these agents in the post-MI or post-cardiac surgery period do not exist. These

patients should be encouraged to follow-up with a psychiatrist for further

monitoring; alternatively, direct coordination with their primary care physician

or cardiologist may allow for repeat screening in two weeks, with a plan for

intervention if indicated. For patients who have evidence of pre-existing

untreated MDD, however, initiation of an SSRI in the immediate post-ACS

period may be indicated. Severe depression that impairs one's ability to

adequately participate in rehabilitation or self-care, or the return of depressive

symptoms in a patient with a prior history of severe depression, may also be

indications for more aggressive treatment with medications. If available,

involvement in collaborative care or blended care programs also appears to be

an excellent treatment strategy for patients.

Case 2

Mr. B, a 52-year-old gentleman with a history of MDD, was admitted to the

hospital with chest pain. His electrocardiogram showed ST-segment

depression in the anterolateral leads, his cardiac enzymes were elevated, and

he was ruled in for an MI. Though he had not been depressed in the year

before admission, he developed depressive symptoms in the days after his MI;

psychiatric consultation was obtained.

On interview, Mr. B was dysphoric but alert, oriented, and able to actively

engage in conversation with the interviewer. He reported depressed mood,

anhedonia, and low energy, along with disturbed concentration and appetite;

844

he denied significant anxiety. He denied feeling suicidal or being unable to

care for himself. Mr. B reported one episode of relatively mild MDD 3 years

earlier that responded well to sertraline (100 mg/day, for 1 year); he had also

had several episodes of “feeling low” for 3 to 5 days that spontaneously

resolved. He appeared to be invested in getting better, he had a strong social

support network, and he planned to follow up with his cardiologist shortly

after his hospitalization.

Given Mr. B's relatively mild current depressive symptoms, his history of

having only one mild episode of MDD, and his ability and willingness to

follow up with his cardiologist, the consultant decided to defer antidepressant

treatment while Mr. B was in the hospital. The consultant contacted Mr. B's

cardiologist and they agreed that sertraline should be started (given Mr. B's

history of good response to this medication) if he continued to be depressed at

his follow-up appointment in 2 weeks.

Mr. B had an uneventful medical course and was discharged 3 days after his

MI. He followed up in 2 weeks with his cardiologist; he remained depressed

and was started on sertraline. He tolerated the sertraline (100 mg per day) well,

and his depressive symptoms subsided over the next 8 weeks.

845

Delirium in Cardiac Patients

Despite advances in the treatment of cardiac illness and the use of non-invasive

procedures, general hospital patients with cardiac disease continue to suffer

delirium at high rates (ranging from 3% to 72% depending on the specific

illness and type of procedure).124 The general hospital psychiatrist should be

aware of the special issues in the diagnosis and management of delirium in

cardiac patients.

Epidemiology

Delirium and Cardiac Disease

The incidence of delirium is 17% among those admitted to a cardiac floor,125

20% among those admitted for ACS,126 up to 25% among those undergoing

cardiac surgery,127 and as high as 34% among those receiving IABP therapy.128

Reports of risk factors for the development of delirium in cardiac patients

have varied, but it is universally agreed that the etiology of delirium in cardiac

patients is multi-factorial, with different factors varying in importance from

patient to patient. Among biological and iatrogenic factors, there are multiple

preoperative risk factors for delirium (including a history of MI or stroke,

diabetes, aortic insufficiency, decreased cardiac output, dehydration, electrolyte

imbalance, and the use of anticholinergic drugs).127,129 Intraoperatively, the use

of on-pump CABG surgery is associated with an increased rate of intracerebral

microemboli and dysfunction of several neurotransmitter systems

(serotonergic, noradrenergic, dopaminergic, and anticholinergic), both of which

may contribute to delirium. Postoperatively, sleep deprivation, use of narcotic

or sedative/hypnotic medications, and possibly some cardiac medications (e.g.,

digoxin) may cause or contribute to delirium.124

Delirium and Medical Outcome

Delirium in cardiac patients has been shown to be associated with longer

intensive care unit (ICU) stays, ICU re-admission, longer hospital stays, greater

prevalence of falls, greater chance of discharge to a nursing facility, and

increased mortality at 30 days and 1 year.125,130–132 Older individuals who

develop delirium in the setting of cardiac surgery have been found to have

poorer short-term function in terms of independent activities of daily living.133

Differential Diagnosis of Delirium in the Cardiac

Patient

In the delirious cardiac patient, a number of specific causes should be carefully

considered (Table 26-3). Central nervous system (CNS) hypoperfusion is a

common mechanism of delirium in the cardiac population; this can result from

poor cardiac output caused by HF or myocardial ischemia, from co-morbid

846

carotid disease, from CNS bleeding (in the setting of anticoagulation), or from

relative hypotension. The phenomenon of relative hypotension deserves special

mention. Patients with baseline uncontrolled hypertension who are admitted

with myocardial ischemia or another cardiac event are often placed on one or

more antihypertensives, and blood pressure is run “low,” with systolic blood

pressure typically between 100 and 120 mmHg. Such patients are likely to have

significant baseline hypertension, which may lead to stiffening of cerebral

vessels with impaired ability to autoregulate. When these patients' blood

pressure is lowered to “normal” (significantly below their baseline blood

pressure), cerebral hypoperfusion and ischemia may result, leading to delirium.

Table 26-3

Selected Causes of Delirium Among Cardiac

Patients in the General Hospital

Central Nervous System Hypoperfusion

Myocardial infarction/ischemia

Cerebrovascular accident (ischemic or hemorrhagic)

Hypovolemia (due to dehydration or bleeding)

Relative hypotension

Other General Medical Conditions

Electrolyte abnormalities (especially sodium with diuretic administration)

Thyroid abnormalities

Hypertensive encephalopathy

Hypoxia (during pulmonary edema)

Infections (e.g., pneumonia, urinary tract infections)

Alcohol withdrawal

Cardiopulmonary bypass

Medication-Related Causes

Digoxin toxicity

Narcotic analgesics

847

Benzodiazepines

Anticholinergic medications

H2-blockers

Other common causes of delirium in the cardiac patient include hypoxia

during HF, hypertensive encephalopathy, electrolyte abnormalities (e.g.,

hyponatremia in the context of diuretic therapy), and medication effects (e.g.,

digoxin toxicity). The general hospital psychiatrist should rule out each of these

potential etiologies of delirium in the cardiac patient.

Psychopharmacologic Issues in the Delirious

Cardiac Patient

The use of medications is an important component of a multi-pronged

approach to the psychiatric management and treatment of delirium; such an

approach also includes monitoring and ensuring safety, educating the patient

and family regarding the illness, and implementing environmental and

supportive interventions (e.g., placing the patient near the nursing station and

frequently re-orienting the patient). This chapter will not discuss the treatment

of delirium in depth, but it will touch on the topics relevant to the delirious

cardiac patient.

In general, antipsychotic agents are used for the management of delirium.

They can reduce agitation and psychotic symptoms and may help normalize the

sleep–wake cycle. They usually are not the primary treatment for delirium, but

they can reduce the risk of patient harm and alleviate patient distress until the

etiology is identified and effectively treated. In general, they are quite safe.

Haloperidol has been the agent most widely used in the management of

delirium. This agent can be given orally or intramuscularly, but the intravenous

(IV) form is both more rapidly acting and much less associated with the

development of extrapyramidal symptoms (EPS); prospective study and clinical

experience have found the rate of EPS with IV haloperidol to be minimal.134

Haloperidol generally has no significant effects on heart rate, blood pressure, or

respiratory status, and it has essentially no anticholinergic effects.

Haloperidol has, however, been associated with the development of TDP.

More than 70 cases of TDP have been reported to the FDA with IV

haloperidol,103 though TDP remains a very rare phenomenon, given the millions

of delirious patients treated. Many of these cases were also confounded by the

use of other QTc-prolonging medications and other risk factors for QTc

prolongation. TDP appears more common at high doses (>35 mg/day) of

haloperidol, though it has also occurred at low doses.135

Other typical antipsychotics sometimes used to manage delirium, such as

chlorpromazine, are also associated with QTc prolongation and TDP. In the case

of droperidol, concerns about its propensity to cause TDP have significantly

reduced its availability and use.

848

More recently, atypical antipsychotics, especially risperidone, quetiapine, and

olanzapine, have been used in the management of delirium. Though these

agents are generally considered safe in cardiac populations, risperidone and

quetiapine can cause orthostatic hypotension, and quetiapine and olanzapine

have some anticholinergic effects. Many atypical antipsychotics also cause

weight gain and predispose patients to metabolic syndrome, though the risk is

likely lower with short-term use. There have also been concerns about the

potential for atypical antipsychotics to cause QTc prolongation in patients with

cardiac disease. In healthy volunteers, ziprasidone causes the greatest mean

QTc prolongation of the atypicals, and although it has only been associated

with TDP in a handful of cases, its use in cardiac patients is not

recommended.136 In a meta-analysis comparing side-effects of antipsychotic

agents, ziprasidone and iloperidone caused the most QTc prolongation, while

aripiprazole and lurasidone performed the best.137 To date, there are no case

reports of lurasidone causing QTc-prolongation or TDP. Finally, in elderly

patients with dementia, atypical antipsychotics have been associated with

mortality related to cardiac events (some of which may represent episodes of

ventricular arrhythmia such as TdP); this has led to an FDA black box warning

for these medications and highlights the caution needed when prescribing these

medications in certain populations.103

Because hypokalemia and hypomagnesemia have been associated with the

development of TDP, it is recommended that patients receiving antipsychotics

for delirium have these electrolytes monitored and repleted as needed. If

possible, an ECG should be checked prior to the initial dose of antipsychotic,

and checked again 30–60 minutes following the dose. We recommend daily

ECGs for at least two days for patients receiving standing doses of

antipsychotics. If the QTc is >500 ms or increases by more than 25%, a careful

risk–benefit analysis is recommended before proceeding with further dosing.

Nonetheless, because TDP is a very low base-rate phenomenon and difficult to

predict even in the setting of a prolonged QT interval, and because agitation in

delirium may predispose patients to even greater risks, such as removing

central lines and other devices, there may be an indication for ongoing use of

antipsychotic agents even with significantly lengthened QTc intervals.

When QTc prolongation or other concerns preclude the use of antipsychotic

medications, other second-line medications may be used for management of

delirium in cardiac patients. Valproic acid is often helpful in reducing agitation

and frontal disinhibition, and is relatively safe in cardiac populations.

Increasingly, α2 agonists, such as clonidine and dexmedetomidine, are being

used to manage delirium. These agents should be used with some caution in

cardiac patients, given their risk of hypotension and bradycardia, respectively.

Trazodone is another agent sometimes used to mitigate agitation, though it also

carries a risk of orthostasis. Finally, benzodiazepines are sometimes given in

combination with an antipsychotic to reduce the dose of antipsychotic needed;

though these agents may worsen confusion, they are effective in the short term,

when sedation is urgently needed.

Emerging evidence has suggested a possible role for delirium prophylaxis in

849

cardiac patients undergoing surgery or those with particularly high risk for

delirium, though further studies are needed before this becomes common

clinical practice. In a study of 126 patients undergoing cardiac surgery, those

receiving risperidone after awakening postoperatively had significantly lower

rates of delirium compared to those receiving placebo.138 Dexmedetomidine, a

potent and highly selective α2 agonist, has been increasingly studied for

delirium management and prophylaxis. A recent meta-analysis found reduced

rates of postoperative delirium in patients receiving intraoperative

dexmedetomidine, and other studies have shown lower delirium rates when

dexmedetomidine was used for ICU sedation after surgery instead of

propofol.139,140

The Practical Management of the Delirious Cardiac Patient

As with other conditions involving cardiac patients, the management of

delirium involves careful diagnosis and the consideration of co-morbid

conditions. Once diagnosis and etiology have been established, the general

hospital psychiatrist can then implement optimal behavioral and nonpharmacologic

strategies and intelligently use psychotropic agents that reduce

medical risk while effectively decreasing symptoms.

Make an Informed Diagnosis of Delirium, and Carefully Consider Potential

Etiologies.

Delirium is characterized by an acute onset, disorientation, poor attention,

fluctuation of levels of consciousness, and alterations in sleep–wake cycle.

Psychotic symptoms, anxiety, worry, and reports of depressed mood may or

may not be present. A careful review of the chart and cognitive evaluation (that

considers orientation, attention, and executive function) can allow the

consultant to use these factors to distinguish delirium from other psychiatric

illnesses. Once a diagnosis has been made, the psychiatrist should work to

consider all possible causes of delirium. The cause of delirium is frequently

multi-factorial; therefore, the identification of one potential contributing factor

of delirium should not preclude the search for further potential abnormalities

leading to an acute confusional state. The consultant should pay special

attention to the initiation and termination of medications and their relationship

to the onset of delirium; a careful review of nursing medication sheets often

reveals a wealth of information that can provide important answers regarding a

delirium of unknown etiology.

Aggressively Treat All Potential Etiologies of Delirium.

Treating the core etiology of delirium is the only way to definitively reverse

delirium; all other behavioral and pharmacologic remedies are symptomatic

treatments that reduce risk and increase comfort until the primary etiologies of

the delirium resolve. Therefore, treatment of urinary tract infections, vitamin B12

deficiency, mild metabolic abnormalities, and other seemingly minor

contributing factors to delirium is absolutely crucial.

850

Use Non-Pharmacologic Strategies to Minimize Confusion and Ensure

Safety.

Having the patient situated near the nursing station or in other areas where the

patient can be monitored frequently, can reduce the risk of falls, wandering, or

other dangerous actions. Placing the patient in a room with a window and a

clock—to help orient patients to day–night cycles—can also be useful. The use

of mittens, sitters, or locked restraints may be required when a delirious

patient's inability to safely navigate places him or her at risk; in almost all cases,

medication should be given in combination with physical restraint to reduce

discomfort and risk of harm while in restraints. The presence of reassuring

family members or friends at the bedside can mitigate paranoia and agitation,

whereas visitors that over-stimulate the patient may worsen symptoms. The

consultant may recommend that the team either encourage or dissuade

interaction with certain visitors depending on the response of the patient's

symptoms to the visitors.

Use Antipsychotic Medications to Reduce Agitation and Psychotic

Symptoms and Regulate the Sleep–Wake Cycle.

Pharmacologic management should be strongly considered in those with

hyperactive delirium or in those with significant psychotic symptoms, such as

paranoia or hallucinations. IV haloperidol remains the “gold standard” for

managing delirium. The protocol used at the MGH (Table 26-4) for the use of IV

haloperidol considers risk factors for TDP and uses a progressive dosing

schedule. An initial dose (from 0.5 to 10 mg based on the age and size of the

patient and the extent of agitation) is selected and administered to the patient. If

the patient is not calm within 20 to 30 minutes, the dose is doubled and

continues to be doubled every 20 to 30 minutes until the patient is calm. This

effective dose is then used when and if the patient again becomes agitated.

Though most patients require standard doses of haloperidol (2 to 10 mg), some

patients have required (and safely received) thousands of milligrams for

agitation.141

Table 26-4

Massachusetts General Hospital Protocol for

IV Haloperidol in Agitated Delirious Patients

Check Pre-Haloperidol QTc Interval

If QTc >450 ms, proceed with care.

If QTc >500 ms, consider other options.

Check Potassium and Magnesium, and Correct Abnormalities

Aim for potassium >4 mEq/L, magnesium >2 mEq/L.

851

Give Dose of Haloperidol (0.5–10 mg) Based on Level of Agitation

and Patient's Age and Size

Goal is to have patient calm and awake.

Haloperidol precipitates with phenytoin and heparin; flush line before giving

haloperidol if these agents have been used in the same intravenous tubing.

Wait 20–30 minutes. If patient remains agitated, double dose.

Continue to double dose every 30 minutes until patient is calm.

Follow QTc Interval to Ensure That QTc Is Not Prolonging

If QTc increases by 25% or becomes >500 ms, consider alternative treatments.

Once Effective Dose Has Been Determined, Use That Dose for Future

Episodes of Agitation

Depending on likely course of delirium, may schedule haloperidol or give on

as-needed basis.

For example, may divide previous effective dose over next 24 hours, giving

every 6 hours.

Or may simply give effective dose as needed for agitation.

Consider small dose at night to regulate sleep–wake cycle in all delirious

patients.

If an agitated, delirious patient is or becomes unable to receive IV haloperidol

(e.g., because of QTc prolongation), there are a number of other options.

Atypical antipsychotics may be used, though it should be noted that they may

also carry a risk for QTc prolongation and may be associated with other side

effects, such as orthostasis. Consideration may be given to alternative agents

discussed above, such as valproic acid, clonidine, dexmedetomidine, or

trazodone, or to adding a small amount of benzodiazepine to the antipsychotic

dose.

Once the agitated delirious patient has been safely and adequately sedated,

there is often a question of whether to schedule antipsychotic medication or to

use it on an as-needed basis for agitation. Such a decision may depend on the

likely duration of the delirium, if this can be determined. For example, if the

delirium is secondary to CNS hypoperfusion in a patient with low cardiac

output on an IABP, such delirium may well be prolonged, and scheduling of an

antipsychotic would be reasonable. In contrast, delirium in an elderly cardiac

patient resulting from narcotic administration may be short-lived once the

narcotic has been eliminated, and standing antipsychotics may not be needed.

852

In most cases of delirium, we have found that it is often reasonable to schedule

a low dose of IV haloperidol or an oral atypical antipsychotic at bedtime to help

regulate the sleep–wake cycle, which is often seriously perturbed in delirious

patients. We have found that by ensuring adequate sleep at appropriate hours,

delirious cardiac patients have the best possible chance to recover.