# 01 - 30.1 Electroconvulsive Therapy

# 30.1 Electroconvulsive Therapy

Brain Stimulation Methods
 30.1 Electroconvulsive Therapy
Convulsive therapies for major psychiatric illnesses predate the modern therapeutic era,
with the use of camphor reported as early as the 16th century and the existence of
several accounts of camphor convulsive therapies from the late 1700s to the mid-1800s.
Unaware 
of 
the 
history 
of 
camphor 
convulsive 
therapy, 
the 
Hungarian
neuropsychiatrist Ladislas von Meduna made the observation that the brains of
epileptics had greater than normal numbers of glial cells, whereas those of
schizophrenics had fewer, and he hypothesized that there might be a biological
antagonism between convulsions and schizophrenia. Following animal experimentation,
camphor was (again) selected as the appropriate agent to use for the therapeutic
induction of seizures. In 1934, the first catatonic psychotic patient was successfully
treated using intramuscular injections of camphor in oil to produce therapeutic seizures.
Lucio Bini and Ugo Cerletti were interested in the use of electricity to induce seizures,
and, after a series of animal experiments and observation of the use of electricity
commercially, they were able to safely apply current across the heads of animals for this
purpose. In 1938, the first electroconvulsive treatment (ECT) course was administered to
a delusional and incoherent patient, who improved with one treatment and remitted
after 11 treatments. Electrical induction of convulsive therapy could be made more
reliable and shorter acting than chemically induced convulsive therapies, and, by the
early 1940s, it had replaced them. In 1940, the first use of ECT occurred in the United
States.
In an effort to reduce the retrograde memory problems that persisted for some
patients after the initial recovery period post-ECT, explorations of nondominant
electrode placement and alternative, more efficient waveforms were undertaken in
subsequent decades. The practice of ECT also benefited from the introduction of
controlled trials methodology, which demonstrated its safety and efficacy, and from
refinements made in diagnostic systems and the process of informed consent. In the
1980s and 1990s efforts to ensure uniformly high standards of practice were under way
with the publication of recommendations for treatment delivery, education, and training
by professional organizations in the United States, England, Scandinavia, and Canada,
among others.
With the widespread use of pharmacological agents as first-line treatments for major
psychiatric disorders, ECT is now more commonly used for patients with resistance to
those treatments, except in the case of life-threatening illness due to inanition, severe
suicidal symptoms, or catatonia. Although the failure of subconvulsive stimulation to

induce the remission of psychiatric illness and the effectiveness of chemical convulsive
therapy suggested that the seizure was necessary and sufficient for therapeutic benefit
with ECT, it is now known that there is a dose–response relationship with right
unilateral ECT and that bilateral ECT is likely to be ineffective with ultrabrief pulse
widths. Work continues to explore the underlying mechanisms and biological
characteristics of effective ECT treatments, with interest in having the treatment focus
on appropriate neural networks with a more efficient stimulus as a method of reducing
cognitive side effects. With the growing understanding that depression is a chronic
disease for many patients, more emphasis has been placed on continuation and
maintenance treatments following an acute course of ECT. Utilization of ECT has
diminished since the middle of the 20th century; but because ECT remains the most
effective treatment for major depression and a rapidly effective treatment for lifethreatening psychiatric conditions, ECT, unlike its contemporaneous somatic therapies,
such as insulin coma, remains in the active treatment portfolio of modern therapeutics.
Its use has shifted from public to private institutions, and it is estimated that
approximately 100,000 patients have received ECT annually over the past few decades
in the United States (Table 30.1-1).
Table 30.1-1
Milestones in the History of Convulsive Therapy

The Nobel Laureate Paul Greengard has suggested that the term electrocortical therapy
might be used to replace the current term electroconvulsive therapy. Greengard has
acknowledged that if the mechanism of action of ECT, as yet unknown, turns out to be
subcortical, then the term might have limited use. Until that time, however, the authors
of this text think Greengard’s suggestion deserves consideration. It would help diminish
the fear associated with the word convulsion and help destigmatize a very effective
treatment method.
ELECTROPHYSIOLOGY IN ELECTROCONVULSIVE THERAPY
Neurons maintain a resting potential across the plasma membrane and may propagate
an action potential, which is a transient reversal of the membrane potential. Normal
brain activity is desynchronized; that is, neurons fire action potentials asynchronously.
A convulsion, or seizure, occurs when a large percentage of neurons fire in unison. Such
rhythmical changes in the extracellular potential entrain neighboring neurons,
propagate the seizure activity across the cortex and into deeper structures, and
eventually engulf the entire brain in high-voltage synchronous neuronal firing. Cellular
mechanisms work to contain the seizure activity and to maintain cellular homeostasis,

and the seizure eventually ends. In epilepsy, any of possibly several hundred genetic
defects can alter the balance in favor of unrestrained activity. In ECT, seizures are
triggered in normal neurons by application through the scalp of pulses of current, under
conditions that are carefully controlled to create a seizure of a particular duration over
the entire brain.
The qualities of the electricity used in ECT can be described by Ohm’s law: E = IR, or I
= E/R, in which E is voltage, I is current, and R is resistance. The intensity or dose of
electricity in ECT is measured in terms of charge (milliampere-seconds or millicoulombs)
or energy (watt-seconds or joules). Resistance is synonymous with impedance and, in
the case of ECT, both the electrode’s contact with the body and the nature of the bodily
tissues are the major determinants of resistance. The skull has a high impedance; the
brain has a low impedance. Because scalp tissues are much better conductors of
electricity than bone, only about 20 percent of the applied charge actually enters the
skull to excite neurons. The ECT machines that are now widely used can be adjusted to
administer the electricity under conditions of constant current, voltage, or energy.
MECHANISM OF ACTION
The induction of a bilateral generalized seizure is necessary for both the beneficial and
the adverse effects of ECT. Although a seizure superficially seems as though it is an allor-none event, some data indicate that not all generalized seizures involve all the
neurons in deep brain structures (e.g., the basal ganglia and the thalamus); recruitment
of these deep neurons may be necessary for full therapeutic benefit. After the
generalized seizure, the electroencephalogram (EEG) shows about 60 to 90 seconds of
postictal suppression. This period is followed by the appearance of high-voltage delta
and theta waves and a return of the EEG to preseizure appearance in about 30 minutes.
During the course of a series of ECT treatments, the interictal EEG is generally slower
and of greater amplitude than usual, but the EEG returns to pretreatment appearance 1
month to 1 year after the end of the course of treatment.
One research approach to the mechanism of action for ECT has been to study the
neurophysiological effects of treatment. Positron emission tomography (PET) studies of
both cerebral blood flow and glucose use have shown that, during seizures, cerebral
blood flow, use of glucose and oxygen, and permeability of the blood–brain barrier
increase. After the seizure, blood flow and glucose metabolism are decreased, perhaps
most markedly in the frontal lobes. Some research indicates that the degree of decrease
in cerebral metabolism is correlated with therapeutic response.
Seizure foci in idiopathic epilepsy are hypometabolic during interictal periods; ECT
itself acts as an anticonvulsant because its administration is associated with an increase
in the seizure threshold as treatment progresses. Recent data suggest that for 1 to
2 months following a session of ECT, EEGs record a large increase in slow-wave activity
located over the prefrontal cortex in patients who responded well to the ECT. Highintensity, bilateral stimulation produced the best response; low-intensity, unilateral
stimulation, the weakest. These data are of unclear significance, however, because the

specific EEG correlate disappeared 2 months after ECT, whereas the clinical benefit
persisted.
ECT affects the cellular mechanisms of memory and mood regulation and raises the
seizure threshold. The latter effect may be blocked by the opiate antagonist naloxone
(Narcan).
Neurochemical research into the mechanisms of action of ECT has focused on changes
in neurotransmitter receptors and, recently, changes in second-messenger systems.
Virtually every neurotransmitter system is affected by ECT, but a series of ECT sessions
results in downregulation of postsynaptic β-adrenergic receptors, the same receptor
change observed with virtually all antidepressant treatments. The effects of ECT on
serotonergic neurons remain controversial. Various research studies have reported an
increase in postsynaptic serotonin receptors, no change in serotonin receptors, and a
change in the presynaptic regulation of serotonin release. ECT has also been reported to
effect changes in the muscarinic, cholinergic, and dopaminergic neuronal systems. In
second-messenger systems, ECT has been reported to affect the coupling of G-proteins to
receptors, the activity of adenylyl cyclase and phospholipase C, and the regulation of
calcium entry into neurons.
Recently, there has been increased interest in structural changes in the brain
associated with psychiatric syndromes and response to treatment. This has been
particularly so for microscopic changes associated with electroconvulsive stimulation, as
well as antidepressant and other medications. In animals, mostly rodents, synaptic
plasticity in hippocampus, including mossy fiber sprouting, alterations in cytoskeletal
structure, increased connectivity in perforant pathways, promotion of neurogenesis, and
suppression of apoptosis have been observed. Many of these structural events are also
observed, although to a lesser extent, with antidepressant medications such as fluoxetine
(Prozac). These reports have also galvanized controversy over various aspects of the
technical validity of the observations. It is unknown whether such changes occur
clinically and, if they do, what significance to efficacy and cognitive side effects might
be discovered.
INDICATIONS
Major Depressive Disorder
The most common indication for ECT is major depressive disorder, for which ECT is the
fastest and most effective available therapy. ECT should be considered for use in
patients who have failed medication trials, have not tolerated medications, have severe
or psychotic symptoms, are acutely suicidal or homicidal, or have marked symptoms of
agitation or stupor. Controlled studies have shown that up to 70 percent of patients who
fail to respond to antidepressant medications may respond positively to ECT. Table
30.1-2 presents the indications for the use of ECT.
Table 30.1-2
Indications for the Use of Electroconvulsive Therapy

ECT is effective for depression in both major depressive disorder and bipolar I
disorder. Delusional or psychotic depression has long been considered particularly
responsive to ECT; but recent studies have indicated that major depressive episodes with
psychotic features are no more responsive to ECT than nonpsychotic depressive
disorders. Nevertheless, because major depressive episodes with psychotic features
respond poorly to antidepressant pharmacotherapy alone, ECT should be considered
much more often as the first-line treatment for patients with the disorder. Major
depressive disorder with melancholic features (e.g., markedly severe symptoms,
psychomotor retardation, early morning awakening, diurnal variation, decreased
appetite and weight, and agitation) is considered likely to respond to ECT. ECT is
particularly indicated for persons who are severely depressed, who have psychotic
symptoms, who show suicidal intent, or who refuse to eat. Depressed patients less likely
to respond to ECT include those with somatization disorder. Elderly patients tend to
respond to ECT more slowly than do young patients. ECT is a treatment for major
depressive episode and does not provide prophylaxis unless it is administered on a longterm maintenance basis.
Manic Episodes
ECT is at least equal to lithium (Eskalith) in the treatment of acute manic episodes. The
pharmacological treatment of manic episodes, however, is so effective in the short term
and for prophylaxis that the use of ECT to treat manic episodes is generally limited to
situations with specific contraindications to all available pharmacological approaches.
The relative rapidity of the ECT response indicates its usefulness for patients whose
manic behavior has produced dangerous levels of exhaustion. ECT should not be used for

a patient who is receiving lithium, because lithium can lower the seizure threshold and
cause a prolonged seizure.
Schizophrenia
Although an effective treatment for the symptoms of acute schizophrenia, ECT is not for
those of chronic schizophrenia. Patients with schizophrenia who have marked positive
symptoms, catatonia, or affective symptoms are considered most likely to respond to
ECT. In such patients, the efficacy of ECT is about equal to that of antipsychotics, but
improvement may occur faster.
Other Indications
Small studies have found ECT effective in the treatment of catatonia, a symptom
associated with mood disorders, schizophrenia, and medical and neurological disorders.
ECT is also reportedly useful to treat episodic psychoses, atypical psychoses, obsessivecompulsive disorder, and delirium and such medical conditions as neuroleptic malignant
syndrome, hypopituitarism, intractable seizure disorders, and the on–off phenomenon of
Parkinson’s disease. ECT may also be the treatment of choice for depressed suicidal
pregnant women who require treatment and cannot take medication; for geriatric and
medically ill patients who cannot take antidepressant drugs safely; and perhaps even
for severely depressed and suicidal children and adolescents who may be less likely to
respond to antidepressant drugs than are adults. ECT is not effective in somatization
disorder (unless accompanied by depression), personality disorders, and anxiety
disorders.
CLINICAL GUIDELINES
Patients and their families are often apprehensive about ECT; therefore, clinicians must
explain both beneficial and adverse effects and alternative treatment approaches. The
informed-consent process should be documented in the patients’ medical records and
should include a discussion of the disorder, its natural course, and the option of
receiving no treatment. Printed literature and videotapes about ECT may be useful in
attempting to obtain a truly informed consent. The use of involuntary ECT is rare today
and should be reserved for patients who urgently need treatment and who have a
legally appointed guardian who has agreed to its use. Clinicians must know local, state,
and federal laws about the use of ECT.
Pretreatment Evaluation
Pretreatment 
evaluation 
should 
include 
standard 
physical, 
neurological, 
and
preanesthesia examinations and a complete medical history. Laboratory evaluations
should include blood and urine chemistries, a chest X-ray, and an electrocardiogram
(ECG). A dental examination to assess the state of patients’ dentition is advisable for
elderly patients and patients who have had inadequate dental care. An X-ray of the

spine is needed if other evidence of a spinal disorder is seen. Computed tomography
(CT) or magnetic resonance imaging (MRI) should be performed if a clinician suspects
the presence of a seizure disorder or a space-occupying lesion. Practitioners of ECT no
longer consider even a space-occupying lesion to be an absolute contraindication to
ECT, but with such patients the procedure should be performed only by experts.
Concomitant Medications.
 Patients’ ongoing medications should be assessed for
possible interactions with the induction of a seizure, for effects (both positive and
negative) on the seizure threshold, and for drug interactions with the medications used
during ECT. The use of tricyclic and tetracyclic drugs, monoamine oxidase inhibitors,
and antipsychotics is generally considered acceptable. Benzodiazepines used for anxiety
should be withdrawn because of their anticonvulsant activity; lithium should be
withdrawn because it can result in increased postictal delirium and can prolong seizure
activity; clozapine (Clozaril) and bupropion (Wellbutrin) should be withdrawn because
they are associated with the development of late-appearing seizures. Lidocaine
(Xylocaine) should not be administered during ECT because it markedly increases the
seizure threshold; theophylline (Theo-Dur) is contraindicated because it increases the
duration of seizures. Reserpine (Serpasil) is also contraindicated because it is associated
with further compromise of the respiratory and cardiovascular systems during ECT.
Premedications, Anesthetics, and Muscle Relaxants
Patients should not be given anything orally for 6 hours before treatment. Just before
the procedure, the patient’s mouth should be checked for dentures and other foreign
objects, and an intravenous (IV) line should be established. A bite block is inserted in the
mouth just before the treatment is administered to protect the patient’s teeth and tongue
during the seizure. Except for the brief interval of electrical stimulation, 100 percent
oxygen is administered at a rate of 5 L a minute during the procedure until spontaneous
respiration returns. Emergency equipment for establishing an airway should be
immediately available in case it is needed.
Muscarinic Anticholinergic Drugs.
 Muscarinic anticholinergic drugs are
administered before ECT to minimize oral and respiratory secretions and to block
bradycardias and asystoles, unless the resting heart rate is above 90 beats a minute.
Some ECT centers have stopped the routine use of anticholinergics as premedications,
although their use is still indicated for patients taking β-adrenergic receptor antagonists
and those with ventricular ectopic beats. The most commonly used drug is atropine,
which can be administered 0.3 to 0.6 mg intramuscularly (IM) or subcutaneously (SC) 30
to 60 minutes before the anesthetic or 0.4 to 1.0 mg IV 2 or 3 minutes before the
anesthetic. An option is to use glycopyrrolate (Robinul) (0.2 to 0.4 mg IM, IV, or SC),
which is less likely to cross the blood–brain barrier and less likely to cause cognitive
dysfunction and nausea, although it is thought to have less cardiovascular protective
activity than does atropine.

Anesthesia.
 Administration of ECT requires general anesthesia and oxygenation.
The depth of anesthesia should be as light as possible, not only to minimize adverse
effects but also to avoid elevating the seizure threshold associated with many
anesthetics. Methohexital (Brevital) (0.75 to 1.0 mg/kg IV bolus) is the most commonly
used anesthetic because of its shorter duration of action and lower association with
postictal arrhythmias than thiopental (Pentothal) (usual dose 2 to 3 mg/kg IV), although
this difference in cardiac effects is not universally accepted. Four other anesthetic
alternatives are etomidate (Amidate), ketamine (Ketalar), alfentanil (Alfenta), and
propofol (Diprivan). Etomidate (0.15 to 0.3 mg/kg IV) is sometimes used because it does
not increase the seizure threshold; this effect is particularly useful for elderly patients
because the seizure threshold increases with age. Ketamine (6 to 10 mg/kg IM) is
sometimes used because it does not increase the seizure threshold, although its use is
limited by the frequent association of psychotic symptoms with emergence from
anesthesia with this drug. Alfentanil (2 to 9 mg/kg IV) is sometimes coadministered with
barbiturates to allow the use of low doses of the barbiturate anesthetics and, thus,
reduce the seizure threshold less than usual, although its use can be associated with an
increased incidence of nausea. Propofol (0.5 to 3.5 mg/kg IV) is less useful because of its
strong anticonvulsant properties.
Muscle Relaxants.
 After the onset of the anesthetic effect, usually within a
minute, a muscle relaxant is administered to minimize the risk of bone fractures and
other injuries resulting from motor activity during the seizure. The goal is to produce
profound relaxation of the muscles, not necessarily to paralyze them, unless the patient
has a history of osteoporosis or spinal injury or has a pacemaker and, therefore, is at
risk for injury related to motor activity during the seizure. Succinylcholine (Anectine),
an ultrafast-acting depolarizing blocking agent, has gained virtually universal
acceptance for the purpose. Succinylcholine is usually administered in a dose of 0.5 to 1
mg/kg as an IV bolus or drip. Because succinylcholine is a depolarizing agent, its action
is marked by the presence of muscle fasciculations, which move in a rostrocaudal
progression. The disappearance of these movements in the feet or the absence of muscle
contractions after peripheral nerve stimulation indicates maximal muscle relaxation. In
some patients, tubocurarine (3 mg IV) is administered to prevent myoclonus and
increases in potassium and muscle enzymes; these reactions can be a problem in
patients with musculoskeletal or cardiac disease. To monitor the duration of the
convulsion, a blood pressure cuff may be inflated at the ankle to a pressure in excess of
the systolic pressure before infusion of the muscle relaxant to allow observation of
relatively innocuous seizure activity in the foot muscles.
If a patient has a known history of pseudocholinesterase deficiency, atracurium
(Tracrium) (0.5 to 1 mg/kg IV) or curare can be used instead of succinylcholine. In such
a patient, the metabolism of succinylcholine is disrupted, and prolonged apnea may
necessitate emergency airway management. In general, however, because of the short
half-life of succinylcholine, the duration of apnea after its administration is generally
shorter than the delay in regaining consciousness caused by the anesthetic and the

postictal state.
Electrode Placement.
 Historically, most practitioners have used bifrontotemporal
electrode placement because of its reliability in producing efficacy and its ease of use.
This electrode placement is also associated with more short-term and long-term adverse
cognitive effects and is more likely to produce delirium, which may require interrupting
a course of ECT and perhaps even terminating it before optimal therapeutic effects have
been obtained. Hence, when bifrontotemporal ECT is used, attention should be paid to
restricting the dose to a moderately suprathreshold level to attenuate adverse cognitive
effects as much as possible. It should be emphasized that the combination of ultrabrief
pulse and bifrontotemporal electrode placement has not been demonstrated to be
effective. Treatment with bilateral electrode placements, particularly a bifrontal
configuration, is more likely to manifest EEG seizure without motor seizure, and EEG
monitoring can be particularly useful in detecting its occurrence.
Newer electrode placements include bifrontal configuration and asymmetrical
placements. There are limitations to these strategies, imposed by the fact that the high
impedance of the skull and scalp causes spreading of the electrical stimulus and restricts
possibilities for localization of the stimulus. Bifrontal electrode placement, with
positioning far enough laterally to minimize interference with impedance relations, has
been investigated, and there have been several demonstrations that bifrontal electrode
placements are equally effective to bifrontotemporal and adequately dosed right
unilateral electrode configurations. Evidence of advantages in sparing of cognitive
effects is quite preliminary, and adequately powered investigations with more extensive
and sensitive cognitive batteries are needed. Seizure threshold is likely to be relatively
higher with bifrontal ECT.
The relatively better cognitive side effect profile of right unilateral ECT should
encourage wider use now that the efficacy of this electrode placement can be ensured
with adequate dosing strategies. In contrast to bilateral ECT, a dose closer to 500
percent above the seizure threshold is more likely to ensure efficacy. ECT devices in the
United States are restricted to an output in the range of 504 to 576 mCi. Approximately
90 percent of patients have seizure thresholds that can accommodate optimal dosing
with brief-pulse right unilateral ECT, and the combination of right unilateral electrode
placement with ultrabrief pulse width extends the range of US devices so that most
patients can be treated within these constraints. Individuals with an exceptionally high
seizure threshold may require bilateral electrode placements to remain within the device
restrictions. Maximizing interelectrode distance by using the d’Elia placement may also
be optimal. Many other right unilateral placements have been described, but there is
little work to support their use (Fig. 30.1-1).

FIGURE 30.1-1
Electrode placements. Position 1 represents the frontotemporal position, used for both
electrodes, one on each side of the head, in conducting bilateral electroconvulsive
therapy (ECT). For right unilateral ECT, one electrode is in the right frontotemporal
position, and the other is just to the right of the vertex at position 2. (Courtesy of
American Psychiatric Association, with permission.)
There has been some concern that left-handed patients may require different electrode
placement than right-handed patients, especially if unilateral placement is desired. Even
when handedness is lateralized to the left, the anatomic localization of language
function in 70 percent of left-handed individuals is the same as in those who are righthanded. Furthermore, there is evidence for independent lateralization of affect, with the
right hemisphere involved in sustaining depressed mood regardless of handedness.
Because of limited indications that affective function and efficacy of ECT are associated
with handedness, handedness is not generally used to guide the choice of electrode
placement.
Electrical Stimulus
The electrical stimulus must be sufficiently strong to reach the seizure threshold (the
level of intensity needed to produce a seizure). The electrical stimulus is given in cycles,
and each cycle contains a positive and a negative wave. Old machines use a sine wave;
however, this type of machine is now considered obsolete because of the inefficiency of
that wave shape. When a sine wave is delivered, the electrical stimulus in the sine wave
before the seizure threshold is reached and after the seizure is activated is unnecessary
and excessive. Modern ECT machines use a brief pulse waveform that administers the

electrical stimulus usually in 1 to 2 milliseconds at a rate of 30 to 100 pulses a second.
Machines that use an ultrabrief pulse (0.5 milliseconds) are not as effective as brief
pulse machines.
Establishing a patient’s seizure threshold is not straightforward. A 40 times variability
in seizure thresholds occurs among patients. In addition, during the course of ECT
treatment, a patient’s seizure threshold may increase 25 to 200 percent. The seizure
threshold is also higher in men than in women and higher in older than in younger
adults. A common technique is to initiate treatment at an electrical stimulus that is
thought to be below the seizure threshold for a particular patient and then to increase
this intensity by 100 percent for unilateral placement and by 50 percent for bilateral
placement until the seizure threshold is reached. A debate in the literature concerns
whether a minimally suprathreshold dose, a moderately suprathreshold dose (one and a
half times the threshold), or a high suprathreshold dose (three times the threshold) is
preferable. The debate about stimulus intensity resembles the debate about electrode
placement. Essentially, the data support the conclusion that doses of three times the
threshold are the most rapidly effective and that minimal suprathreshold doses are
associated with the fewest and least severe cognitive adverse effects.
Induced Seizures
A brief muscular contraction, usually strongest in a patient’s jaw and facial muscles, is
seen concurrently with the flow of stimulus current, regardless of whether a seizure
occurs. The first behavioral sign of the seizure is often a plantar extension, which lasts
10 to 20 seconds and marks the tonic phase. This phase is followed by rhythmic (i.e.,
clonic) contractions that decrease in frequency and finally disappear. The tonic phase is
marked by high-frequency, sharp EEG activity on which a higher frequency muscle
artifact may be superimposed. During the clonic phase, bursts of polyspike activity occur
simultaneously with the muscular contractions but usually persist for at least a few
seconds after the clonic movements stop.
Monitoring Seizures.
 A physician must have an objective measure that a bilateral
generalized seizure has occurred after the stimulation. The physician should be able to
observe either some evidence of tonic-clonic movements or electrophysiological evidence
of seizure activity from the EEG or electromyogram (EMG). Seizures with unilateral ECT
are asymmetrical, with higher ictal EEG amplitudes over the stimulated hemisphere than
over the nonstimulated hemisphere. Occasionally, unilateral seizures are induced; for
this reason, at least a single pair of EEG electrodes should be placed over the
contralateral hemisphere when using unilateral ECT. For a seizure to be effective in the
course of ECT, it should last at least 25 seconds.
Failure to Induce Seizures.
 If a particular stimulus fails to cause a seizure of
sufficient duration, up to four attempts at seizure induction can be tried during a course
of treatment. The onset of seizure activity is sometimes delayed as long as 20 to 40

seconds after the stimulus administration. If a stimulus fails to result in a seizure, the
contact between the electrodes and the skin should be checked, and the intensity of the
stimulus should be increased by 25 to 100 percent. The clinician can also change the
anesthetic agent to minimize increases in the seizure threshold caused by the anesthetic.
Additional procedures to lower the seizure threshold include hyperventilation and
administration of 500 to 2,000 mg IV of caffeine sodium benzoate 5 to 10 minutes
before the stimulus.
Prolonged and Tardive Seizures.
 Prolonged seizures (seizures lasting more than
180 seconds) and status epilepticus can be terminated either with additional doses of the
barbiturate anesthetic agent or with IV diazepam (Valium) (5 to 10 mg). Management
of such complications should be accompanied by intubation, because the oral airway is
insufficient to maintain adequate ventilation over an extended apneic period. Tardive
seizures—that is, additional seizures appearing some time after the ECT treatment—may
develop in patients with preexisting seizure disorders. Rarely, ECT precipitates the
development of an epileptic disorder in patients. Such situations should be managed
clinically as if they were pure epileptic disorders.
Number and Spacing of Treatments
ECT treatments are usually administered two to three times a week; twice-weekly
treatments are associated with less memory impairment than thrice-weekly treatments.
In general, the course of treatment of major depressive disorder can take 6 to 12
treatments (although up to 20 sessions are possible); the treatment of manic episodes
can take 8 to 20 treatments; the treatment of schizophrenia can take more than 15
treatments; and the treatment of catatonia and delirium can take as few as 1 to 4
treatments. Treatment should continue until the patient achieves what is considered the
maximal therapeutic response. Further treatment does not yield any therapeutic benefit,
but increases the severity and duration of the adverse effects. The point of maximal
improvement is usually thought to occur when a patient fails to continue to improve
after two consecutive treatments. If a patient is not improving after 6 to 10 sessions,
bilateral placement and high-density treatment (three times the seizure threshold)
should be attempted before ECT is abandoned.
Multiple-Monitored Electroconvulsive Therapy.
 Multiple-monitored ECT
(MMECT) involves giving multiple ECT stimuli during a single session, most commonly
two bilateral stimuli within 2 minutes. This approach may be warranted in severely ill
patients and in those at especially high risk from the anesthetic procedures. MMECT is
associated with the most frequent occurrences of serious cognitive adverse effects.
Maintenance Treatment
A short-term course of ECT induces a remission in symptoms but does not, of itself,
prevent a relapse. Post-ECT maintenance treatment should always be considered.

Maintenance therapy is generally pharmacological, but maintenance ECT treatments
(weekly, biweekly, or monthly) have been reported to be effective relapse prevention
treatments, although data from large studies are lacking. Indications for maintenance
ECT treatments can include rapid relapse after initial ECT, severe symptoms, psychotic
symptoms, and the inability to tolerate medications. If ECT was used because a patient
was unresponsive to a specific medication, then, following ECT, the patient should be
given a trial of a different medication.
Failure of Electroconvulsive Therapy Trial
Patients who fail to improve after a trial of ECT should again be treated with the
pharmacological agents that failed in the past. Although the data are primarily
anecdotal, many reports indicate that patients who had previously failed to improve
while taking an antidepressant drug do improve while taking the same drug after
receiving a course of ECT treatments, even if the ECT seemed to be a therapeutic failure.
Nonetheless, with the increased availability of drugs that act at diverse receptor sites, it
is less often necessary to return to a drug that has failed than it was formerly.
ADVERSE EFFECTS
Contraindications
ECT has no absolute contraindications, only situations in which a patient is at increased
risk and has an increased need for close monitoring. Pregnancy is not a contraindication
for ECT, and fetal monitoring is generally considered unnecessary unless the pregnancy
is high risk or complicated. Patients with space-occupying central nervous system lesions
are at increased risk for edema and brain herniation after ECT. If the lesion is small,
however, pretreatment with dexamethasone (Decadron) is given, and hypertension is
controlled during the seizure and the risk of serious complications minimized for these
patients. Patients who have increased intracerebral pressure or are at risk for cerebral
bleeding (e.g., those with cerebrovascular diseases and aneurysms) are at risk during
ECT because of the increased cerebral blood flow during the seizure. This risk can be
lessened, although not eliminated, by control of the patient’s blood pressure during the
treatment. Patients with recent myocardial infarctions are another high-risk group,
although the risk is greatly diminished 2 weeks after the myocardial infarction and is
even further reduced 3 months after the infarction. Patients with hypertension should be
stabilized on their antihypertensive medications before ECT is administered. Propranolol
(Inderal) and sublingual nitroglycerin can also be used to protect such patients during
treatment.
Mortality
The mortality rate with ECT is about 0.002 percent per treatment and 0.01 percent for
each patient. These numbers compare favorably with the risks associated with general

anesthesia and childbirth. ECT death is usually from cardiovascular complications and is
most likely to occur in patients whose cardiac status is already compromised.
Central Nervous System Effects
Common adverse effects associated with ECT are headache, confusion, and delirium
shortly after the seizure while the patient is coming out of anesthesia. Marked confusion
may occur in up to 10 percent of patients within 30 minutes of the seizure and can be
treated with barbiturates and benzodiazepines. Delirium is usually most pronounced
after the first few treatments and in patients who receive bilateral ECT or who have
coexisting neurological disorders. The delirium characteristically clears within days or a
few weeks at the longest.
Memory.
 The greatest concern about ECT is the association between ECT and
memory loss. About 75 percent of all patients given ECT say that the memory
impairment is the worst adverse effect. Although memory impairment during a course of
treatment is almost the rule, follow-up data indicate that almost all patients are back to
their cognitive baselines after 6 months. Some patients, however, complain of persistent
memory difficulties. For example, a patient may not remember the events leading up to
the hospitalization and ECT, and such autobiographical memories may never be
recalled. The degree of cognitive impairment during treatment and the time it takes to
return to baseline are related, in part, to the amount of electrical stimulation used
during treatment. Memory impairment is most often reported by patients who have
experienced little improvement with ECT. Despite the memory impairment, which
usually resolves, no evidence indicates brain damage caused by ECT. This subject has
been the focus of several brain-imaging studies, using a variety of modalities; virtually
all concluded that permanent brain damage is not an adverse effect of ECT. Neurologists
and epileptologists generally agree that seizures that last less than 30 minutes do not
cause permanent neuronal damage.
Other Adverse Effects of Electroconvulsive Therapy
Fractures often accompanied treatments in the early days of ECT. With routine use of
muscle relaxants, fractures of long bones or vertebrae should not occur. Some patients,
however, may break teeth or experience back pain because of contractions during the
procedure. Muscle soreness can occur in some individuals, but it often results from the
effects of muscle depolarization by succinylcholine and is most likely to be particularly
troublesome after the first session in a series. This soreness can be treated with mild
analgesics, including nonsteroidal anti-inflammatory drugs (NSAIDs). A significant
minority of patients experience nausea, vomiting, and headaches following an ECT
treatment. Nausea and vomiting can be prevented by treatment with antiemetics at the
time of ECT (e.g., metoclopramide [Reglan], 10 mg IV, or prochlorperazine
[Compazine], 10 mg IV; ondansetron [Zofran] is an acceptable alternative if adverse
effects preclude use of dopamine receptor antagonists).

ECT can be associated with headaches, although this effect is usually readily
manageable. Headaches often respond to NSAIDs given in the ECT recovery period. In
patients with severe headaches, pretreatment with ketorolac (Toradol) (30 to 60 mg IV),
an NSAID approved for brief parenteral use, can be helpful. Acetaminophen (Tylenol),
tramadol (Ultram), propoxyphene (Darvon), and more potent analgesia provided by
opioids can be used individually or in various combinations (e.g., pretreatment with
ketorolac and postseizure management with acetaminophen-propoxyphene) to manage
more intractable headache. ECT can induce migrainous headache and related symptoms;
sumatriptan (Imitrex) (6 mg SC or 25 mg orally) may be a useful addition to the agents
described above. Ergot compounds can exacerbate cardiovascular changes observed
during ECT and probably should not be a component of ECT pretreatment.
INVESTIGATIONS IN ELECTRICAL BRAIN STIMULATION TREATMENT
There is interest in continued refinements of ECT techniques. Common themes in these
approaches are focusing the treatment spatially to optimize dosing in brain areas
associated with putative neural networks involved in depression and other
psychopathologies that are indications for ECT, diminishing dosing in areas associated
with adverse cognitive effects, and improving the efficiency of a noninvasive electrical
stimulus in direction and amplitude, even to a subconvulsive level. This research is
parallel to investigations in magnetic stimulation (e.g., repetitive transcranial magnetic
stimulation) and to the renaissance of invasive electrical techniques (e.g., vagal nerve
stimulation and deep brain stimulation).
REFERENCES
Byrne P, Cassidy B, Higgins P. Knowledge and attitudes towards electroconvulsive therapy among health care professionals
and students. J ECT. 2006;22(2):133.
Cristancho MA, Alici Y, Augoustides JG, O’Reardon JP. Uncommon but serious complications associated with
electroconvulsive therapy: Recognition and management for the clinician. Curr Psychiatry Rep. 2008;10:474.
Hooten WM, Rasmussen KG Jr. Effects of general anesthetic agents in adults receiving electroconvulsive therapy: A
systematic review. J ECT. 2008;24:208.
Ingram A, Saling MM, Schweitzer I. Cognitive side effects of brief pulse electroconvulsive therapy: A review. J ECT.
2008;24:3.
Kellner CH, Knapp RG, Petrides G, Rummans TA, Husain MM. Continuation electroconvulsive therapy vs
pharmacotherapy for relapse prevention in major depression: A multisite study from the Consortium for Research in
Electroconvulsive Therapy (CORE). Arch Gen Psychiatry. 2006;63:1337.
Lapidus KA, Shin JS, Pasculli RM, Briggs MC, Popeo DM, Kellner CH. Low-dose right unilateral electroconvulsive therapy
(ECT): Effectiveness of the first treatment. J ECT. 2013;29(2):83–85.
Munk-Olsen T, Laursen TM, Videbech P, Rosenberg R, Mortensen PB. Electroconvulsive therapy: Predictors and trends in
utilization from 1976 to 2000. J ECT. 2006;22(2):127.
Painuly N, Chakrabarti S. Combined use of electroconvulsive therapy and antipsychotics in schizophrenia: The Indian
evidence. A review and a meta-analysis. J ECT. 2006;22:59.
Prudic J. Electroconvulsive therapy. In: Sadock BJ, Sadock VA, Ruiz P, eds. Kaplan & Sadock’s Comprehensive Textbook of