24 - Electroconvulsive Therapy | Manual of Psychiatric Therapeutics: Practical Psychopharmacology and Psychiatry (Little, Browns Paperback Book Series)

Editors: Shader, Richard I.

Title: Manual of Psychiatric Therapeutics, 3rd Edition

> Table of Contents > 24 - Electroconvulsive Therapy

Electroconvulsive Therapy

Laura J. Fochtmann

Chester Pearlman

Richard I. Shader

Since its first use in 1938, electroconvulsive therapy (ECT) has been effective for several psychiatric disorders. The development of effective psychotropic medications in the 1950s reduced the need for ECT, and negative media portrayals influenced public perceptions of ECT. Patients were often fearful, and the availability of ECT decreased, particularly in the public sector. More recently, studies have led to technical improvements and an enhanced knowledge of risk-to-benefit considerations. Because few psychiatrists in the United States administer ECT, more than one psychiatrist is usually involved. In such cases, communication among them must be more frequent and more detailed than with the medical-surgical model. Psychiatrists who do not provide ECT also must understand the indications and procedures to make appropriate referrals.

This chapter reviews the clinical use of ECT. It covers indications and assessment of potential benefits and risks. An overview of ECT administration is provided. More detailed practice guidelines can be found in the reports of the American Psychiatric Association Committee on Electroconvulsive Therapy and the Royal College of Psychiatrists Special Committee on ECT (see Additional Reading).

I. Indications

A. General Factors

Patients who have failed adequate medication trials are the most appropriate candidates for ECT. A past response of patients or their biologic relatives to ECT also suggests that ECT will be efficacious. For this reason, some patients prefer ECT as a first-line treatment. ECT is also indicated as primary treatment for severe symptoms (extreme suicidality or inanition) where the risk of a failed drug trial would be unwise. In such patients, ECT usually produces more rapid improvement, particularly for vegetative symptoms. An advantage of ECT for the medically ill is that exposure is brief and that it is associated with careful monitoring and ready management of complications. For pregnant patients, ECT may be advantageous by avoiding the long-term use of potentially teratogenic medications.

B. Diagnostic Considerations

Major depression. ECT is the gold standard for treatment of depressive episodes both in major depressive or bipolar disorder. Greater efficacy of ECT has been shown in comparisons with tricyclics, monoamine oxidase inhibitors, and selective serotonin reuptake inhibitors. Response rates are 80% to 90%, although some evidence exists to indicate that failure of an adequate drug trial is associated with a lower response rate. Those with comorbid personality disorders tend to do less well, but older patients and those with psychomotor retardation respond better. ECT is also highly effective in patients with psychotic features. It is somewhat superior to, and is often better tolerated than, therapy with an antidepressant combined with an antipsychotic agent. In patients with dementia, depression often contributes to cognitive impairment and makes caregiving more difficult. When other treatments are unsuccessful, ECT can safely and effectively treat such depression. A somewhat increased risk of acute delirium or more persistent cognitive impairment may be present, but this risk can be minimized by the adjustment of treatment frequency or other parameters if indicated.
Mania. ECT is the only antidepressant treatment that is also effective for mixed or manic episodes. In acute mania, ECT is generally reserved
P.379

for those who do not respond to the usual combinations of mood stabilizers (e.g., lithium, valproic acid), benzodiazepines (e.g., clonazepam), or antipsychotic agents (e.g., olanzapine). ECT is indicated as the primary treatment in severe excited states and for those who are unresponsive to medication or who have required frequent physical restraint during previous episodes. Comparison studies have shown equivalent or superior efficacy to that of lithium or antipsychotic agents. Before the advent of lithium, maintenance ECT was the standard treatment; it remains useful for patients who do not tolerate medications or who relapse in spite of them.
Catatonia. This syndrome involves symptoms such as mutism, severe hypokinesia, and dystonia (catalepsy or rigidity). It is much more common in mood disorders than it is in schizophrenia. Although the initial treatment is usually with a benzodiazepine (e.g., lorazepam), ECT is highly effective if symptom resolution is incomplete. Although benzodiazepine use is generally minimized during an ECT course to avoid possible interference with efficacy, it may augment the ECT response in some catatonics. In neuroleptic malignant syndrome, which shares many signs and symptoms with catatonia, some patients will respond better to ECT than to other treatments.
Schizophrenia and related disorders. Although schizophrenia, schizophreniform, and schizoaffective disorders are generally less responsive to ECT than are the mood disorders, ECT may also improve symptoms of these conditions. Hallucinations, delusions, and changes in mood show more improvement when ECT is given in combination with an antipsychotic agent. Patients with longer duration of illness, insidious onset, or premorbid paranoid or schizoid personality traits do less well. However, even with an extensive history of minimal response to antipsychotic agents, some improve with ECT.
Other diagnoses. For some medical conditions, ECT may benefit the syndrome itself. Presumably due to the dopaminergic effects of ECT, patients with Parkinson disease may show an increased benefit of antiparkinson drugs and improvements in rigidity or on off periods. In intractable seizures, ECT has been beneficial by increasing the seizure threshold and stimulating the release of endogenous anticonvulsant substances involved in termination of seizure activity.

C. Age-Related Considerations

Age is not relevant in determining appropriateness for ECT. For those under 18 years of age, concurrence should be obtained from a consultant familiar with the treatment of children and adolescents (two should be obtained for those under 13 years of age). Because adult psychiatrists administer most ECT, this process ensures the consideration of the unique aspects of treatment of children and adolescents. As was discussed above, the elderly often tolerate ECT better than drugs.

D. Considerations Relating to Medical Conditions

Pre-ECT medical and anesthetic evaluation. This involves a complete medical history; a physical examination, including assessment of the teeth; laboratory screening tests, including a pregnancy test for women of childbearing age; electrocardiogram; and chest x-ray, as defined by local policies for brief procedures involving general anesthesia. Preanesthesia assessment by an anesthesiologist includes questions about adverse reactions to barbiturates and a personal or family history of abnormal responses to succinylcholine (e.g., prolonged apnea or malignant hyperthermia) and the assignment of American Society of Anesthesiologists (ASA) risk status according to the ASA system (Barash et al., 1991).
Medical conditions associated with increased risk. Although no absolute contraindications to ECT exist, some of the following conditions are associated with increased risk: severe or unstable cardiovascular
P.380

disorders, such as recent myocardial infarction, unstable angina, poorly compensated congestive heart failure, severe valvular heart disease, aneurysm, or vascular malformations, and recent gastrointestinal bleeding. In the central nervous system, a substantially increased risk is associated with increased intracranial pressure. Others with increased risk are those ranked as ASA status 4 (i.e., incapacitating illness that is a threat to life [heart failure, renal failure]) or ASA status 5 (i.e., a moribund patient not expected to survive 24 hours [ruptured aneurysm, head trauma with increasing intracranial pressure]). ECT is not done in ASA 5 patients.

E. Technical and Pharmacologic Modifications to Optimize Benefit and to Minimize Risk

Medical comorbidities. Table 24.1 lists suggested ways to minimize risks of specific conditions. Some standard medications, such as hypoglycemic agents and diuretics, should be held until after ECT, and the dosage of antiparkinson agents should also be decreased by 50%. Antihypertensive, antianginal, antiarrhythmic, and antireflux agents; bronchodilators; glaucoma medications; corticosteroids; and other medications that may reduce side effects should be given with a minimum amount of water before ECT.
Modifications in psychotropic medications. Antipsychotic agents and antidepressants can generally be continued during an ECT course. Some evidence of synergistic effects of antipsychotic agents in patients with schizophrenia does exist. The use of antidepressants is less clear. One report has been made of possible synergism, and no evidence of increased risk is seen. Concern among anesthesiologists regarding hypertensive reactions with monoamine oxidase inhibitors is based on the use of indirect sympathomimetic agents during emergency surgery and is not relevant to current practice. Lithium doses should be reduced or discontinued when feasible because of the increased potential for neurotoxicity. With unilateral electrode placement, benzodiazepines may interfere with the efficacy due to the necessity for greatly suprathreshold stimulation. For this electrode placement, agents without active metabolites should be used and no benzodiazepines should be given for 8 hours before ECT. In patients receiving anticonvulsants for seizure disorders or mood stabilization, bitemporal ECT is advised to avoid the necessity for greatly suprathreshold stimulation. Morning doses of these medications should be held before ECT, and sometimes the preceding evening doses should be held as well.

Modifications in electrode placement and treatment frequency. At comparable stimulus intensities, fewer cognitive effects occur with right unilateral ECT than with bitemporal ECT. However, unlike bitemporal placement, optimum benefit with unilateral ECT requires stimulation that is at least five to six times the threshold. Medications that increase the seizure threshold (e.g., benzodiazepines, barbiturates, anticonvulsants) are likely to diminish the efficacy of unilateral ECT. Although the seizure threshold is usually much lower with unilateral ECT, United States Food and Drug Administration restrictions on the energy output of ECT devices in the United States may make sufficient suprathreshold stimulation impossible. Several strategies have been used to minimize the cognitive side effects while maximizing response. One is to begin with right unilateral ECT with stimulus intensity significantly above the seizure threshold and to change to bitemporal placement if minimal response is seen after about 2 weeks. Another strategy is to begin with bitemporal ECT and to change to unilateral placement or twice weekly treatment if cognitive side effects become a problem. Other electrode placements are less studied, but these may provide comparable therapeutic efficacy to bitemporal placement with less impact on cognition. They include bifrontal positioning and a

P.381

P.382

right frontotemporal left frontal electrode placement (see Additional Reading).

TABLE 24.1. MEDICAL CONDITIONS AND ELECTROCONVULSIVE THERAPY MODIFICATIONS

Medical Condition	Suggested Modification
Addison disease or other steroid-dependent conditions	Additional steroid boluses may be needed at the time of electroconvulsive therapy (ECT)
Aneurysms	Use antihypertensive agents to blunt the surge in blood pressure that occurs with seizure induction
Asthma or chronic obstructive pulmonary disease	Pre-ECT bronchodilators and additional oxygenation may be required; discontinue theophylline or decrease dose to avoid prolonging seizures
Bone or joint disease	If severe, increase muscle relaxant dosage
Cardiac arrhythmias	Lidocaine interferes with seizure induction; bradycardia or asystole may require treatment with anticholinergic agents; tachycardias may require treatment with -adrenergic receptor antagonists; in patients with pacemakers, cardiology consultants should comment on whether a change to a fixed firing rate is indicated; in patients with implanted defibrillators, cardiology consultants should comment on whether this function should be disabled during ECT
Central nervous system tumor or other mass lesion	Antihypertensive agents, steroids, diuretics, or hyperventilation can minimize the likelihood of increases in intracranial pressure
Diabetes mellitus	More frequent glucose monitoring may be needed; dosages of oral hypoglycemic agents or insulin may need to be adjusted
Esophageal reflux	H₂-receptor antagonists or sodium citrate can be given before ECT to minimize and neutralize stomach acids; metoclopramide can promote gastric emptying; cricoid pressure at the time of ECT may decrease aspiration risk; consider intubation for high-risk patients
Glaucoma	Prescribed antiglaucoma medications should be given before ECT
Hyperthyroidism	If clinically significant, give -adrenergic receptor antagonists to decrease the risk of thyroid storm
Intracranial shunt	Ensure shunt patency before ECT
Ischemic cerebrovascular disease	Avoid induction of low blood pressure with ECT
Myasthenia gravis	Decrease the dose of succinylcholine
Parkinson disease	Decrease the dose of dopaminergic agonists, such as L-dopa, to minimize post-ECT delirium
Pheochromocytoma	Give medications to block -adrenergic receptors, -adrenergic receptors, and tyrosine hydroxylase synthesis
Porphyria	Use only nonbarbiturate anesthetics
Pregnancy	Depends on the stage of pregnancy and recommendations of obstetrical consultants; modifications may include altered patient position with a wedge placed under the right hip at the time of ECT after 20 wk gestation, altered doses of anesthetic medications due to pharmacokinetic changes of pregnancy; to prevent and reduce risks of aspiration, minimize the use of anticholinergic agents, use antireflux agents and consider intubation near-term; ensure good oxygenation without hyperventilation; fetal monitoring, including fetal heart rate before and after ECT past 14 16 wk gestation and a 30-min to 60-min fetal heart rate strip (nonstress test with a tocometer) before and after ECT past 24 wk gestation
Seizure disorders	Optimal anticonvulsant dosages
Traumatic brain injury	Avoid placing stimulus electrodes directly over a skull defect
Upper motor neuron disease	Decrease doses of succinylcholine
Abbreviation: H₂, histamine-2.

II. Electroconvulsive Therapy Procedures

A. Number and Frequency of Treatments

The number of treatments in an ECT course is determined by the patient's response. With major depressive episodes, the time course with ECT is similar to that with antidepressant drugs. Some show improvement after one or two ECT treatments, but full response usually requires 3 to 4 weeks. With schizoaffective disorder or schizophrenia, longer courses may improve the outcome. Regardless of diagnosis, however, benefit is unlikely if no response is seen after 4 weeks. In the United States, ECT is generally administered three times per week on nonconsecutive days. Twice weekly ECT is also effective, and it has fewer cognitive effects, especially with bitemporal placement.

B. Informed Consent

From a legal perspective, capacity to consent to a standard procedure like ECT is presumed. However, when the clinical evaluation of the patient suggests that this capacity may be impaired by cognitive deficits or psychotic beliefs, a formal assessment of capacity is warranted before obtaining informed consent. To have this capacity, a patient must show a clear understanding of the risks, benefits, and alternatives to the proposed treatment. For individuals lacking this capacity, individual state laws govern consent procedures. A typical method involves surrogate consent from a family member, legal guardian, or health care proxy, but judicial proceedings for court-ordered treatment may be required. In such instances, consideration should be given to opinions expressed by the patient when this capacity was not compromised by acute illness. The consent process for a patient or surrogate involves the rationale; a description of the procedure; a range of likely number of treatments; possible complications, including memory disturbance; activity restrictions during and after the ECT course; and the likely need for some form of continuation or maintenance treatment. For persons influenced by negative media imagery, consultation with other

P.383

psychiatrists or educational videotapes may be helpful. If videotapes are used, the clinician should review them before giving them to the patient to ensure their appropriateness for that particular patient, as well as their current accuracy.

C. Treatment Environment

As a procedure involving general anesthesia, ECT is administered in an area with appropriate monitoring equipment and emergency medical support. When the ECT treatment area is distant from the psychiatric unit, familiar personnel and occasional preanesthetic medication may minimize patient anxiety.

D. Anesthetic Considerations

General factors. Throughout the procedure, monitoring should include electrocardiogram, serial blood pressures, and pulse oximetry. Ventilation with 100% oxygen should be delivered using positive pressure at a rate of at least 5 L per min with 15 to 20 breaths per min. Additional preanesthetic oxygenation is indicated for patients at risk of myocardial ischemia.
Anticholinergic medications. Though common in clinical ritual, such agents may increase the cardiovascular risk by inducing tachycardia and increased hypertension. Their appropriate use is restricted to documented bradyarrhythmias or to situations in which subconvulsive stimuli are used to determine the seizure threshold, especially in patients treated with -adrenergic receptor antagonists. Rarely, anticholinergic medications are also indicated when the oral secretions are excessive.
Anesthetic medications. Methohexital (0.5 to 1 mg per kg intravenously [i.v.]) is generally used to induce ultrabrief general anesthesia for ECT. Some practitioners use thiopental (1.5 to 3.5 mg per kg i.v.). Etomidate (0.15 to 0.35 mg per kg i.v.) does not appear to impact the seizure threshold or duration, and it may be useful when adequate seizure induction is difficult. Propofol (0.75 to 1.5 mg per kg i.v.) is useful for patients with barbiturate allergies or conditions requiring minimal hypertensive reaction, but it requires bitemporal ECT because of its significant increase in seizure threshold. Ketamine (2 to 3 mg per kg i.v.) has also been used to enhance efficacy with unilateral ECT, but postictal delirium is a significant complication.
Muscle relaxants. Muscle relaxants are used to minimize motor convulsive activity and also to facilitate airway management. The usual muscle relaxant agent used for ECT is the depolarizing neuromuscular blocking agent succinylcholine (0.5 to 1 mg per kg i.v.). In patients particularly susceptible to fractures, more complete relaxation is required via the use of higher doses of succinylcholine. The rare patients who lack plasma cholinesterases are unable to metabolize succinylcholine rapidly; they are at risk of prolonged apnea. For these patients, nondepolarizing (competitive) neuromuscular blocking agents, such as mivacurium or atracurium, can be used because they have a prompt onset of action, they are not long acting¹, and they are nonenzymatically degraded at pH 7.4. Succinylcholine also causes transiently increased serum potassium. Patients at increased risk for succinylcholine-induced hyperkalemia (e.g. burns, muscle trauma, paraplegia, or other conditions involving neuromuscular degeneration) should be treated with a nondepolarizing agent. In all patients, muscle relaxation must be ensured before delivering the electrical stimulus. With succinylcholine, muscle relaxation is unlikely to be sufficient before fasciculations have stopped. Adequacy of relaxation may also be checked by lower extremity reflexes or the response to peripheral nerve stimulation. Muscle relaxation
P.384

is never complete, however, and residual muscle activity produces carbon dioxide that may contribute to cardiac arrhythmias. Thus, adequate postictal ventilation is essential.
Cardiovascular medications. The initial parasympathetic activation with stimulus delivery and subsequent sympathetic activation with seizure induction are associated with alterations in heart rate and blood pressure. These changes are generally self-limited. Rarely, anticholinergic agents are needed acutely to treat asystole or bradyarrhythmias. More often, medications are used to attenuate tachycardia and hypertension. For example, in patients with cardiac disease, -adrenergic receptor antagonists may be used to decrease tachycardia and cardiac workload. The most commonly used is esmolol (0.25 to 0.5 mg per kg), an ultra short-acting ₁-selective adrenergic receptor antagonist, whose brevity allows use in conditions where -adrenergic receptor antagonists are typically contraindicated. Short-acting antihypertensive agents may also be used to reduce the cardiac workload and to minimize the risk of increased intracranial pressure and intracranial bleeding. The most effective agents are calcium channel blockers, such as nifedipine (10 mg crushed and swallowed 0.5 hours before ECT) or nicardipine (1.25 to 5 mg i.v.). Nitroglycerin (as a paste, sublingual tablet, or sublingual spray) is less reliable, but it may facilitate coronary artery dilation. These agents are short acting, and they do not cause hypotension or orthostasis after recovery from ECT.

E. Seizure Elicitation

Stimulus parameters. The use of a constant current, brief, pulse stimulus is recommended because sine waveforms produce increased cognitive impairment with only rare reports of increased efficacy. Adjustments in stimulus intensity are most efficiently accomplished by altering the duration of the pulse train.
Stimulus dosing. Stimulus dosing with ECT resembles individualized dosing of psychotropic agents and aims to minimize adverse cognitive effects while still producing an effective response. Such procedures include formula-based dosing procedures and stimulus titration. With empirical stimulus titration, an initial dose is selected, generally based on age and gender, that would rarely be expected to elicit a generalized seizure. If no seizure occurs, the stimulus energy is increased, generally by about 50%, and another stimulus is given after a delay of at least 20 seconds. This upward titration continues until a generalized seizure occurs or until four or five stimulations have been unsuccessful. The empirically estimated threshold is taken to be the stimulus intensity at which a seizure occurred. At subsequent treatment sessions, stimulus energy is used that is 1.5 to 2.5 times the seizure threshold for bilateral ECT or 2.5 to 6 times the seizure threshold for unilateral ECT. Common formula-based strategies apply the observation that seizure thresholds vary with age younger patients require lower stimulus intensities, whereas the elderly require much higher stimulus intensities to produce an adequate seizure. Thus, for bitemporal ECT, half of the patient's age as a percentage of maximum stimulus output is a good estimate of seizure threshold. More complex formula-based dosing strategies include variations in seizure threshold by gender and electrode positioning. Regardless of the method by which an initial stimulus dose is selected, adjustments in dosing are generally necessary during a series of ECT. For many patients, the seizure threshold may increase significantly, requiring a comparable increase in stimulus energy by the end of the treatment course.
Stimulus administration. For efficacy and safety, ensuring optimal contact between the skin and the electrode is important; this is produced by cleansing the underlying skin and applying conducting paste to the electrodes. To prevent injury to the teeth and tongue from contraction of
P.385

the masseter muscles induced by the stimulus, a flexible protective device is placed between the teeth before stimulus delivery and the chin is supported manually.
Seizure monitoring. The duration of the induced seizure may be monitored by electroencephalogram (EEG) or by convulsive movements. Although motor signs are usually clear, visualization can be ensured by placing a blood pressure cuff on a wrist or ankle and inflating it above the systolic pressure before the muscle relaxant is administered. When unilateral placement is used, the cuff is placed ipsilateral to the electrodes to be sure of seizure generalization. EEG monitoring is done with a frontomastoid lead placement. With unilateral ECT, the leads are placed contralateral to the electrodes to ensure generalized seizure induction. EEG monitoring is essential for detection of prolonged nonconvulsive seizure activity. Prolonged seizures (longer than 180 seconds by motor or EEG) are rare, and these require termination by an i.v. benzodiazepine or additional barbiturate anesthetic and oxygenation throughout the seizure activity. More often, delivery of the electrical stimulus induces only a brief seizure or none at all. For seizures lasting less than 15 seconds, a neuronal refractory period requires a delay of 45 to 90 seconds before restimulation with a stimulus intensity of 1.5 to 2 times that of the original. When no seizure activity is noted, the delay can be as short as 20 seconds, but restimulation is done with a similarly increased intensity.

When inducing an adequate seizure at a given treatment session is not possible, several factors should be considered. Hyperventilating the patient before seizure induction and ensuring adequate hydration are always helpful. In addition, reducing the anesthetic dose or using an anesthetic with less anticonvulsant effect may be possible. When feasible, the doses of benzodiazepines and other anticonvulsant medications should be decreased and held before ECT. If the benzodiazepine dosage cannot be reduced, the benzodiazepine receptor antagonist flumazenil may be given immediately before ECT to achieve an adequate seizure. Except as noted, no relation is observed between seizure duration and clinical outcome. Although the specific features of the EEG waveforms are a subject of ongoing research, whether these may predict efficacy is not clear.

III. Complications

A. Mortality

Despite the increasing use of ECT in patients with serious medical conditions, morbidity and mortality remain low. The report of the American Psychiatric Association Committee on ECT estimates mortality at 1 in 10,000 patients or 1 in 80,000 treatments, approximately the same as that associated with minor surgical procedures or with childbirth.

B. Cardiovascular Effects

Electrocardiogram changes after ECT include ST segment depression, T-wave inversion, premature ventricular contractions, and, rarely, ventricular tachyarrhythmias. They result from the sympathetic activation, and they are rarely associated with myocardial enzyme changes. A history of arrhythmia increases the risk of arrhythmia with ECT. Similarly, a history of myocardial ischemia increases the risk of morbidity due to ischemia with ECT. The modifications in treatment procedures discussed in section II.D.5 have mostly eliminated such complications.

C. Adverse Cognitive Effects

These show a great deal of individual variability. The most frequent are anterograde or retrograde amnesia. With anterograde amnesia, new information is forgotten more rapidly, whereas with retrograde amnesia, recall of autobiographical events and public information shows spotty deficits. Typically, the recall of public information is affected more than that of personal details, and recent events are affected more than distant ones.

P.386

Amnesia is most prominent during and immediately after a course of ECT, and it fades within a few weeks. Some patients report longer lasting difficulty with memory, and they may show inability to recall items remembered before ECT when tested 6 months later. Because no problem is seen with relearning, distinction from normal forgetting may be difficult. Patients with prominent white matter hyperintensities in the head of the caudate nucleus and prefrontal subcortex on magnetic resonance imaging (most parkinsonian patients) are at increased risk of more severe cognitive effects, including delirium (see Chapter 5). Cognitive effects are also more pronounced with sine wave stimulation, bitemporal brief pulse stimulation given thrice weekly, and possibly high stimulus doses relative to the seizure threshold. Modifications that minimize cognitive side effects include changing electrode placement and decreasing treatment frequency or stimulus intensity. Several nootropic agents have been given with ECT in the attempt to counteract such effects, but none has shown clinical efficacy despite benefit in animal trials. In most patients, improvement in psychiatric symptoms with ECT leads to improved attention, concentration, and global cognitive performance. Thus, patients with persisting subjective memory complaints should be assessed for residual depression. Because ECT affects memory and cognition, many individuals have expressed concern about possible brain damage. ECT does produce increases in cerebral blood flow and metabolism and a transient increase in blood brain barrier permeability, but extensive study with computed tomography, magnetic resonance imaging, and neuronal enzyme markers has shown no evidence of injury to brain tissue.

D. Other Adverse Effects

Prolonged apnea. Prolonged apnea is rare, and it results from the slow metabolism of succinylcholine. If adequate oxygenation is maintained, it usually resolves spontaneously without sequelae after 30 to 60 minutes.
Postictal agitation. Delirium and agitation upon awakening are rare but troublesome complications of ECT. Young men are most susceptible, but no relation to previous history or other treatment factors has been determined. Although this is usually managed supportively, physical restraint (see Chapter 26) may be necessary, and i.v. doses of a benzodiazepine may be useful.
Headache. Headache is common, occurring in up to 45% of patients. It generally responds to rest under dim lighting or to analgesics, such as acetaminophen; nonsteroidal antiinflammatory agents; or 5-hydroxytryptamine (5-HT)_1B/D agonists, such as sumatriptan.
Nausea or vomiting. Nausea and vomiting may be associated with headaches, anesthesia, or recent changes in psychotropic medication dosages. These symptoms are usually well controlled with dopamine receptor antagonists, such as prochlorperazine or metoclopramide.
Muscle aches. This is a known complication of succinylcholine, and it is treated symptomatically with analgesics.
Treatment-emergent mania. All antidepressants may cause a switch to hypomania or mania, but only ECT can treat it. Thus, more ECT leads to mood stabilization. If continuation or maintenance ECT is not used, treatment with a mood-stabilizing agent is advisable.

IV. Ambulatory or Outpatient Electroconvulsive Therapy

With the increasingly restrictive criteria for inpatient psychiatric care, ECT is increasingly being provided in an outpatient setting. In such instances, the patient's symptoms must be safely manageable as an outpatient. Those with significant suicide risk, substantial psychotic or cognitive impairment, or a risk of serious medical complications should be treated as inpatients. Outpatients must be willing and able to comply with specific requirements during the ECT course. These include avoiding activities that could be influenced by the cognitive effects of ECT, complying with medication regimens, and abstaining from

P.387

oral intake for about 8 hours before each ECT. Outpatients must also be able to make reliable reports of changes in their medical condition or of the adverse effects of ECT or anesthesia. Except for stable outpatients receiving maintenance ECT, this requires at least one significant other or caregiver to ensure the patient's safety and compliance with the treatment plan and to transport the patient to and from treatment sessions. For a more detailed discussion, see the report by Fink et al. (1996).

V. Continuation or Maintenance Electroconvulsive Therapy and Continuation Pharmacotherapy

After a successful course of ECT, continuation therapy is advisable to reduce the risk of relapse. Although some may respond to a class of antidepressants that they have not already failed, many will fare best with maintenance ECT. The decision to use this approach is complicated by the unpredictable duration of spontaneous remission that is obscured by maintenance drug treatment guidelines. When ECT is used, treatments are given at intervals ranging from weekly to monthly, with treatment schedules adjusted to clinical response. As suggested by guidelines for drug treatment, the need to continue ECT is assessed at intervals that are based on the number of previous episodes, with patient preferences playing a major role in the decision to continue treatment. One recent study suggested that over 80% of responders to ECT who were treated with an oral placebo relapsed within 6 months after ECT if no additional treatment was given. In this study, nortriptyline combined with lithium was found to reduce the relapse rate to 39%. The use of nortriptyline alone led to a 60% relapse rate.

VI. Possible Mechanisms of Action

Although the mechanism of ECT is unknown, some data are suggestive. The anticonvulsant hypothesis is based on observations that the therapeutic outcome is correlated with the magnitude of increase in seizure threshold during the ECT course, especially with right unilateral stimulation. A relation to the increasing role of anticonvulsants in mood disorders may be present. The neuroendocrine hypothesis suggests that therapeutic benefits result from diencephalic stimulation and the consequent release of mood-regulating neuropeptides. With electroconvulsive shock (ECS), the animal analogue of ECT, many neuropeptides (e.g., thyrotropin-releasing hormone, somatostatin, -endorphin, enkephalin, neurokinin A, neuropeptide Y, nerve growth factor, brain-derived neurotrophic factor) show increased levels in discrete brain regions. The difficulty is determining which, if any, of these is responsible for benefits of ECT. A similar problem exists with neurochemical hypotheses of ECT. With a time course like clinical ECT, ECS increases brain levels of norepinephrine. Repeated ECS also increases cortical and amygdalar ₁-adrenergic receptors while decreasing the numbers of ₂- and -adrenergic receptors in the cortex and hippocampus. Dopamine levels are also increased with repeated ECS, particularly in the striatum. Although dopamine D₂ receptor numbers are generally unchanged, altered dopaminergic responses are seen with repeated ECS, and these relate to the increased numbers of nigrostriatal dopamine D₁ receptors. ECS also leads to acute increases in brain serotonin, but basal serotonin levels remain unchanged. Cortical 5-HT₂ receptor numbers are increased after repeated ECS, but only in male rats. This contrasts with the down-regulation of 5-HT₂ receptors with antidepressant drugs. Repeated, but not single, ECS also affects other receptors, enzymes, and neuronal ion channels. Second messengers (e.g., cyclic adenosine monophosphate [AMP] and phosphatidyl inositol) and multiple transcriptional regulatory factors are also differentially modulated. Despite the limitations of animal models, these findings establish that electrically induced seizures produce a complex array of neurochemically specific changes that are localized to specific neuroanatomic regions. Consequently, they refute the idea that ECT acts as a global reset button for brain electrical activity. In addition, these results suggest that specific neurochemical alterations may relate to the specific effects of ECT. Future research will need to detail the mechanisms by which such neurochemical changes are translated into the therapeutic benefits of ECT.

P.388

VII. Useful Websites

Readers may wish to contact the following for more useful information: the National Institute of Mental Health (800-421-4211 or http://www.nimh.nih.gov/), the American Psychiatric Association APA Answer Center for Electroconvulsive Therapy (http://www.psych.org/), or the JAMA Patient Pages (see Patient Page Index for March 14, 2001; http://www.jama.com/).

ADDITIONAL READING

General References

Abrams R. Electroconvulsive therapy, 3rd ed. New York: Oxford University Press, 1997.

American Psychiatric Association Committee on Electroconvulsive Therapy. The practice of electroconvulsive therapy: recommendations for treatment, training, and privileging, 2nd ed. Washington, D.C.: American Psychiatric Association, 2001.

Beyer JL, Weiner RD, Glenn MD. Electroconvulsive therapy: a programmed text, 2nd ed. Washington, D.C.: American Psychiatric Press, 1998.

Kellner CH. Handbook of ECT. Washington, D.C.: American Psychiatric Press, 1997.

Royal College of Psychiatrists Special Committee on ECT. The ECT handbook: the second report of the Royal College of Psychiatrists' Special Committee on ECT. London: Royal College of Psychiatrists, 1995.

Specific References

Avramov MN, Stool LA, White PF, et al. Effects of nicardipine and labetalol on the acute hemodynamic response to electroconvulsive therapy. J Clin Anesth 1998;10:394 400.

Bailine SH, Rifkin A, Kayne E, et al. Comparison of bifrontal and bitemporal ECT for major depression. Am J Psychiatry 2000;157:121 123.

Barash PL, Cullen BF, Stoelting RK. Handbook of clinical anesthesia. Philadelphia: J B Lippincott, 1991:4.

Beale MD, Kellner CH. Proposed titration schedule. Convuls Ther 1997;13:44.

Bloch Y, Levcovitch Y, Bloch AM, et al. Electroconvulsive therapy in adolescents: similarities to and differences from adults. J Am Acad Child Adolesc Psychiatry 2001;40:1332 1336.

Ende G, Braus DF, Walter S, et al. The hippocampus in patients treated with electroconvulsive therapy: a proton magnetic resonance spectroscopic imaging study. Arch Gen Psychiatry 2000;57:937 943.

Fall PA, Granerus AK. Maintenance ECT in Parkinson's disease. J Neural Transm 1999;106:737 741.

Fink M. Delirious mania. Bipolar Disord 1999;1:54 60.

Fink M, Abrams R, Bailine S, et al. Ambulatory electroconvulsive therapy: report of a task force of the Association for Convulsive Therapy. Association for Convulsive Therapy. Convuls Ther 1996;12:42 55.

Fochtmann LJ. Animal studies of electroconvulsive therapy: foundations for future research. Psychopharmacol Bull 1994;30:321 444.

Folk JW, Kellner CH, Beale MD, et al. Anesthesia for electroconvulsive therapy: a review. J ECT 2000;16:157 170.

Glass RM. Electroconvulsive therapy. JAMA 2001;285:1346 1362.

Krystal AD, Watts BV, Weiner RD, et al. The use of flumazenil in the anxious and benzodiazepine-dependent ECT patient. J ECT 1998;14:5 14.

Lisanby SH, Maddox JA, Prudic J, et al. The effects of electroconvulsive therapy on memory of autobiographical and public events. Arch Gen Psychiatry 2000;57:581 592.

Manly DT, Oakley SP Jr, Bloch RM. Electroconvulsive therapy in old-old patients. Am J Geriatr Psychiatry 2000;8:232 236.

P.389

McCall WV, Reboussin DM, Weiner RD, et al. Titrated moderately suprathreshold vs fixed high-dose right unilateral electroconvulsive therapy: acute antidepressant and cognitive effects. Arch Gen Psychiatry 2000;57:438 444.

Petrides G, Divadeenam KM, Bush G, et al. Synergism of lorazepam and electroconvulsive therapy in the treatment of catatonia. Biol Psychiatry 1997;42:375 381.

Rabheru K, Persad E. A review of continuation and maintenance electroconvulsive therapy. Can J Psychiatry 1997;42:476 484.

Rao V, Lyketsos CG. The benefits and risks of ECT for patients with primary dementia who also suffer from depression. Int J Geriatr Psychiatry 2000;15:729 735.

Rasmussen KG, Jarvis MR, Zorumski CF. Ketamine anesthesia in electroconvulsive therapy. Convuls Ther 1996;12:217 223.

Sackeim HA. The anticonvulsant hypothesis of the mechanisms of action of ECT: current status. J ECT 1999;15:5 26.

Sackeim HA, Haskett RF, Mulsant BH, et al. Continuation pharmacotherapy in the prevention of relapse following electroconvulsive therapy. JAMA 2001;285:1299 1307.

Sackeim HA, Prudic J, Devanand DP, et al. A prospective, randomized, double-blind comparison of bilateral and right unilateral electroconvulsive therapy at different stimulus intensities. Arch Gen Psychiatry 2000;57:425 434.

Swartz CM. Asymmetric bilateral right frontotemporal left frontal stimulus electrode placement for electroconvulsive therapy. Neuropsychobiology 1994;29:174 178.

Tharyan P, Adams CE. Electroconvulsive therapy for schizophrenia. Cochrane Database Syst Rev 2002;2:CD000076.

Walter G, Rey JM, Mitchell PB. Practitioner review: electroconvulsive therapy in adolescents. J Child Psychol Psychiatry 1999;40:325 335.

Weiner RD, Coffey CE, Krystal AD. Electroconvulsive therapy in the medical and neurologic patient. In: Stoudemire A, Fogel BS, Greenberg D, eds. Psychiatric care of the medical patient. New York: Oxford University Press, 2000:419 428.

¹Although short-acting, atracurium has a longer duration of action than mivacurium; cholinergic reversal with neostigmine or edrophonium can be used to terminate its effects.